Interesting articles, March 2022

Russia’s invasion of Ukraine has gone badly. In spite of Russia’s fearsome reputation and its recent military modernization, its battlefield performance has been disappointing and marred by many mistakes. The Ukrainians have fought harder than anyone anticipated, and have inflicted serious losses on their enemy. What was supposed to have been a quick, surgical operation to replace Ukraine’s regime with one friendlier to Moscow is turning into a bloody stalemate. Western sanctions against Russia for the invasion have been very severe, setting the country on course for major economic problems within a few months. It’s clear that Putin badly miscalculated when he decided to launch the war. Here’s a roundup of war articles:

A record high percent of Irish people want their country to join NATO. I think the alliance should focus on “infilling” rather than antagonizing Russia by expanding into ex-Soviet republics.
https://www.politico.eu/article/poll-more-irish-want-to-join-nato/

An interesting bit about NATO standards:
‘Finally, some national standards are recognized as not inferior to NATO standards and do not require revision (in Ukraine, for instance, these are those related to potable water quality). The fact that compliance with all NATO standards is not the norm even for leading states members is evidenced by the fact that no country of the Alliance has achieved the mark of 100%, although in some, including Germany (91%), Great Britain (83%), France, Norway (81% each), Canada (76%), the degree of the implementation of standards is very high…Let us assume that if the current pace of implementation is maintained, Ukraine will implement at least 90% of the existing standards of the Alliance in approximately 13–14 years.’
https://rpr.org.ua/en/news/ukraine-and-nato-standards-progress-under-zelenskyy-s-presidency/

Britain is planning on upgrading its tank fleet to indigenously made Challenger 3’s, but it’s not economical for them to keep such a small fleet of one-off tanks that are specific to their country. They should switch to either the U.S. M1 Abrams or German Leopard 2, and maybe sell their Challenger 2’s to smaller NATO countries, like the Balkan states.
https://www.thedrive.com/the-war-zone/44927/british-armys-next-generation-challenger-3-tank-is-now-under-construction

American and Chinese stealth fighters have encountered each other during patrols over the South China Sea.
https://www.flightglobal.com/defence/f-35s-have-encountered-j-20s-over-east-china-sea-usaf-general/147936.article

The U.S. Navy recovered an F-35C stealth fighter that crashed off an aircraft carrier and sank to the bottom of the ocean. If not for this, a Chinese ship would have hauled it up.
https://www.thedrive.com/the-war-zone/44560/navys-crashed-f-35c-recovered-from-the-bottom-of-the-south-china-sea

The U.S. Navy wants to start decommissioning its expensive, defective Littoral Combat Ships starting next year. The oldest one is only seven years old! The project has been a disaster.
https://www.thedrive.com/the-war-zone/44811/littoral-disaster-navy-wants-to-retire-10-littoral-combat-ships-according-to-report

This German antiaircraft system is basically a giant shotgun meant to destroy drone swarms.
https://www.youtube.com/watch?v=pb5_F4_Eod8

The Soviet VVA-14 was one of the weirdest but potentially most versatile aircraft ever built. Its designer had plans for even more dramatic variants that weren’t built.
https://youtu.be/UD7xiWWs-bs

NASA briefly experimented with “passive” communications satellites that were large balloons made out of radar-reflecting materials. One ground station would aim a powerful radio at it and broadcast a signal, which would bounce off the balloon and deflect at such an angle that another ground station thousands of miles distant would receive it.
https://en.wikipedia.org/wiki/Project_Echo

The Lowy Institute predicts that China’s economic growth with sharply decelerate during the 2020s, and that the country will stagnate starting in the 2030s.
https://www.lowyinstitute.org/publications/revising-down-rise-china

‘Oil demand will peak in 2025, years earlier than previously expected, the International Energy Agency said in its World Energy Outlook on Tuesday.’
https://fortune.com/2021/10/13/oil-demand-peak-2025-world-invest-trillions-renewables-iea-world-energy-outlook/

The “Hall–Héroult process” allowed pure aluminum to be extracted from bauxite for much lower cost than previous methods. Its discovery in 1886 marked the start of aluminum becoming a common material. The Hall–Héroult process was crucial to the development of heavier-than-air aircraft, as it allowed the Wright Brothers to make an engine that was light enough to fit on a plane while still being powerful enough to propel it.
https://www.acs.org/content/acs/en/education/whatischemistry/landmarks/aluminumprocess.html
https://www.wright-brothers.org/Information_Desk/Just_the_Facts/Engines_&_Props/1903_Engine.htm

Using CRISPR, scientists were able to self-fertilize a female mouse with her own eggs. One of the resulting 12 zygotes lived to adulthood. It was not a clone of her. A clone shares 100% of your DNA, and a natural child shares 50% of your DNA. This new technique could be used to make offspring that shared an unnatural amount of your DNA, like 75 or 85%.
https://phys.org/news/2022-03-mammalian-offspring-derived-unfertilized-egg.html

How will humans change in the next 10,000 years?
https://www.realclearscience.com/articles/2022/03/02/how_will_humans_change_in_the_next_10000_years_819486.html

My predictions about humans in 12,022 A.D.:
-There might still be some Homo sapiens, though they will be genetically engineered. Imagine today’s all-star athlete who graduated from high school at 17, went to MIT on an academic scholarship, and is also a model on the side being the average human by then. We could breed with them.
-There will be “human-looking” people that will have radically different genetics and anatomical/physiological features from us. Imagine a person who looks externally normal, but has bird lungs, octopus eyes, and a different number of chromosomes than you. They will be so different that they will count as different species, and we won’t be able to breed with them.
-There will be intelligent but nonhuman-looking life forms that are the products of many iterations of genetic engineering. Imagine something like a horse-sized spider with a big brain occupying most of its torso. It could trace its lineage back to a normal human that is alive today.
-Some members of those three groups of intelligent life forms will be meshed with technology that augments their abilities. There might be a Homo sapien with synthetic, self-healing organs that are superior to his old, natural organs, there might be a “human-looking” Homo neosapien that also has brain implants to make him smarter, and there might be intelligent spiders with nanomachines circulating in their bloodstreams to assist with various bodily processes.

‘[By 1,000 years from now] The bulk of technology will remain simple or semi-simple, while a smaller portion will continue to complexify greatly. I expect our cities and homes a thousand years hence to be recognizable, rather than unrecognizable. As long as we inhabit bodies approximately our size – a few meters and 50 kilos — the bulk of the technology that will surround us need not be crazily more complex. And there is good reason to expect we’ll remain the same size, despite intense genetic engineering and downloading to robots. Our body size is weirdly almost exactly in the middle of the size of the universe. The smallest things we know about are approximately 30 orders of magnitude smaller than ourselves, and the largest structures in the universe are about 30 orders of magnitude bigger. We inhabit a middle scale that is sympathetic to sustainable flexibility in the universe’s current physics. Bigger bodies encourage rigidity, smaller ones encourage empheralization. As long as we own bodies – and what sane being does not want to be embodied? – the infrastructure technology we already have will continue (in general) to work. Roads of stone, buildings of modified plant material and earth, not that different from our cities and homes 2,000 years ago. Some visionaries might imagine complex living buildings in the future, for instance, but most average structures are unlikely to be more complex than the formerly living plants we already use. They don’t need to. I think there is a “complex enough” restraint. Technologies need not complexify to be useful in the future. Danny Hillis, computer inventor, once confided to me that he believed that there’s a good chance that 1,000 years from now computers might still be running programming code from today, say a unix kernel and TCP/IP. They almost certainly will be binary digital. Like bacteria, or cockroaches, these simpler technologies remain simple, and remain viable, because they work. They don’t have to get more complex.’
https://kk.org/thetechnium/the-arc-of-comp/

What happens if you load an enormous amount of data on chemical reactions and human biology into an AI, and then task it with finding lethal compounds against us?

‘In less than 6 hours after starting on our in-house server, our model generated 40,000 molecules that scored within our desired threshold. In the process, the AI designed not only VX, but also many other known chemical warfare agents that we identified through visual confirmation with structures in public chemistry databases. Many new molecules were also designed that looked equally plausible. These new molecules were predicted to be more toxic, based on the predicted LD50 values, than publicly known chemical warfare agents.’
https://www.nature.com/articles/s42256-022-00465-9

The newest language models are measurably better than GPT-3, which was only released 18 months ago. However, they’re still a long way from being able to pass the Turing Test.
https://www.lesswrong.com/posts/yYkrbS5iAwdEQyynW/how-do-new-models-from-openai-deepmind-and-anthropic-perform

This economic model suggests we’re probably 140 years away (or, seven more doublings of global GDP away) from inventing an AGI. Once it is invented, the GDP growth rate will sharply accelerate within 10 years.
https://www.lesswrong.com/posts/ftdCgGmkQ3bPyDadA/phase-transitions-and-agi

A new poll on Americans’ attitudes towards futuristic technologies like autonomous cars, human genetic engineering, and brain implants has been conducted. I’m surprised at how positively they view them.
https://www.pewresearch.org/internet/2022/03/17/ai-and-human-enhancement-americans-openness-is-tempered-by-a-range-of-concerns/

For the first time, a person in a “locked-in” medical state has been able to communicate, thanks to a brain implant.
http://www.technologynetworks.com/neuroscience/news/for-the-first-time-a-completely-locked-in-patient-can-communicate-thanks-to-brain-implant-359819

This analysis predicts that the U.S. trucking industry will probably switch to a “transfer-hub” model where autonomous trucks transport goods over long, simple highway routes, while human drivers in smaller trucks move the cargoes over shorter distances at both ends.
https://www.nature.com/articles/s41599-022-01103-w

People prefer mates that are similar to themselves.
https://www.nature.com/articles/s41467-022-28774-y

Not only is IQ heritable, but specific types of cognitive talents are, too.
http://biorxiv.org/lookup/doi/10.1101/2022.02.05.479237

Small doses of radiation might actually benefit human health thanks to a process called “hormesis.” Note that nuclear power is so expensive partly because the power plants aren’t allowed to release any radiation at all to the surrounding environment, and that requirement is predicated on the assumption that any amount of radiation exposure hurts people.
https://www.biorxiv.org/content/10.1101/832949v1

The first person to receive an implanted pig heart has died. He survived for two months with the animal organ.
https://www.usatoday.com/story/news/health/2022/03/09/human-pig-heart-transplant-patient-dies/9437650002/

I agree that the fake meat industry has been overhyped. Though meat substitutes are cheaper and more convincing than ever, they will not render meat consumption extinct. Not even close. Lab-grown meats will eventually eliminate the need to kill animals for food, but the technology won’t be good enough until near the end of this century (it’s not as good or advancing as fast as its contemporary cheerleaders claim).
https://reason.com/2022/03/05/the-fake-meat-revolution-has-stalled/

Computers have translated pig noises into a basic “vocabulary” of emotional and mental states. A variety of technologies will let us communicate with animals in the future, and possibly to even share thoughts with them.
https://www.nature.com/articles/s41598-022-07174-8

Bulldogs are so inbred that genetic testing might be needed to prevent further breeding of unhealthy members of the species.
https://www.npr.org/2022/03/15/1085173405/bulldogs-health-breed-ban

There’s a plan to use genetic technology to resurrect the extinct Tasmanian tiger species. The last one died in 1936, and full genomes have been recovered from preserved tissue samples.
https://pursuit.unimelb.edu.au/articles/the-9-steps-to-de-extincting-australia-s-thylacine

“DAM-ATOLL” was a proposed structure that would generate electricity from ocean waves.
https://ui.adsabs.harvard.edu/abs/1979mroe.proc…86H/abstract

North Korea still operates some Japanese-made trains from the 1930s. Thanks to the country’s socialist economy, labor is practically free, making it financially possible to keep fixing the trains in spite of their age. Once robots have made labor free across the world, will it become common for manufactured objects of all kinds to stay in service much longer than they do now?
https://www.oryxspioenkop.com/2021/07/blast-from-past-north-koreas-whacky.html

The German gunboat Graf von Goetzen was launched in 1915 and sent to Tanzania (then a German colony) to dominate Lake Tanganyika. Though the Germans left, the ship didn’t, and it remains in service to this day as a ferry, renamed the Liemba.
https://en.wikipedia.org/wiki/MV_Liemba

We now know of 5,000 exoplanets. I remember when we discovered the first one, and what a big deal it was.
https://www.npr.org/2022/03/22/1088009414/there-are-more-than-5-000-confirmed-exoplanets-beyond-our-solar-system-nasa-says

There’s a new scientific paper claiming that ivermectin doesn’t treat COVID-19. Instead, it kills parasitic worms in people, boosting their immune systems just enough to let them survive COVID-19. Parasites are only common in tropical areas.
https://doi.org/10.1001/jamanetworkopen.2022.3079

The pandemic isn’t over: China just locked down one of its biggest and most important cities due to a surge in COVID-19 cases. It will have global economic consequences.
https://www.bbc.com/news/world-asia-china-60893070

Interesting articles, February 2022

Russia has invaded Ukraine for a variety of reasons.

The fortunes of war have turned in Ethiopia, and the government is driving the Tigray rebels back into the latter’s territory.
https://apnews.com/article/africa-kenya-ethiopia-addis-ababa-abiy-ahmed-114c113bf52114157abcb0b165ef3a50

Taiwan is finally upgrading its obsolete tank forces by getting rid of its 1950s-vintage M41 light tanks.
https://www.thedrive.com/the-war-zone/44406/taiwan-prepares-to-give-up-its-m41a3-walker-bulldog-tanks-after-more-than-six-decades

Nicaragua has Central America’s most powerful tank force by a large margin. It could even beat up Mexico’s!
https://laststandonzombieisland.com/tag/nicaragua-t-55/

DARPA has created an add-on kit that turns old UH-60 helicopters into unmanned drones.
https://www.lockheedmartin.com/en-us/news/features/2022/safe-reliable-and-uninhabited-first-autonomous-black-hawk-flight.html

This demonstration video shows how poor the world’s first Gauss rifle is. You’re better off with a 120-year-old rifle.
https://www.youtube.com/watch?v=eAHKS0nVlL4

The “Henry Long Ranger” is probably the most advanced lever-action rifle ever made. However, the big unsolved design flaws were making it easy to clean the barrel and to clean and remove the internal mechanism.
https://youtu.be/wsCO0XV5rwA

Before there was cable TV, there was cable radio.
https://en.wikipedia.org/wiki/Cable_radio

Facebook just had its first ever drop in daily users. They’re out of new people to add, and teens are gravitating towards newer social media apps like TikTok.
https://www.theguardian.com/technology/2022/feb/02/facebook-shares-slump-growth-fourth-quarter

‘Bioenergy with carbon capture and storage (BECCS) is the process of extracting bioenergy from biomass and capturing and storing the carbon, thereby removing it from the atmosphere. The carbon in the biomass comes from the greenhouse gas carbon dioxide (CO2) which is extracted from the atmosphere by the biomass when it grows. Energy is extracted in useful forms (electricity, heat, biofuels, etc.) as the biomass is utilized through combustion, fermentation, pyrolysis or other conversion methods. Some of the carbon in the biomass is converted to CO2 or biochar which can then be stored by geologic sequestration or land application, respectively, enabling carbon dioxide removal (CDR) and making BECCS a negative emissions technology (NET).’
https://en.wikipedia.org/wiki/Bioenergy_with_carbon_capture_and_storage

Thank God combined cycle power plants were invented, because otherwise, we’d always be hearing about magnetohydrodynamic generators.
https://en.wikipedia.org/wiki/Magnetohydrodynamic_generator

“Nucleosynthesis” is a process in which a light element is put inside a fusion reactor and bombarded with protons and neutrons, some of which it captures, converting it into a heavier element. Once we have built commercial fusion reactors, we could use the technique to make heavier elements like gold and uranium, and to ensure that we never ran out of any type of element. Terrestrial nucleosynthesis would also decrease the need for asteroid mining to satisfy Earth’s mineral needs.
https://twitter.com/friedberg/status/1492382218307575809

This very technical article estimates that the first holographic TVs will become commercially available in the 2040s. They will present slightly different images to a viewer’s right and left eyes, fooling their brains into perceiving the displayed images as being 3D. In short, you’ll be able to watch movies in 3D without having to wear the glasses. The first generation of holographic TVs will operate on the principle of “multiview horizontal parallax only” (HPO), meaning 1) several people watching the same TV set from different angles (e.g. – watching from the left, watching straight-on from the center, watching from the right) will perceive its footage as being 3D and 2) the picture will only look 3D if the viewers’ heads are level. If one of them tilted his head 90 degrees, so his ear was flat against his shoulder, the footage would look 2D instead of 3D. Improved holographic displays operating on more advanced principles will emerge in the decades after.
https://www.light-am.com/en/article/doi/10.37188/lam.2021.028

Gawaine Baillie stated of the fusee, “Perhaps no problem in mechanics has ever been solved so simply and so perfectly.”
https://en.wikipedia.org/wiki/Fusee_(horology)

‘In North America, gradient is expressed in terms of the number of feet of rise per 100 feet of horizontal distance. Two examples: if a track rises 1 foot over a distance of 100 feet, the gradient is said to be “1 percent;” a rise of 2 and-a-half feet would be a grade of “2.5 percent.”…The effect of grades on train operations is significant. For each percent of ascending grade, there is an additional resistance to constant-speed movement of 20 lbs. per ton of train. This compares with a resistance on level, straight track of about 5 lbs. per ton of train. A given locomotive, then, can haul only half the tonnage up a .25-percent grade that it can on the level. Descending grades carry their own penalties in the form of equipment wear and tear and increased fuel consumption.’
https://www.trains.com/trn/train-basics/abcs-of-railroading/grades-and-curves/

There are super-strong concrete mixes called “Macro-Defect-Free” (MDF) cement, but they are rarely used since they’re expensive. If you wanted to make a house to survive the apocalypse, consider walls made of steel-reinforced MDF cement.
https://abc-utc.fiu.edu/wp-content/uploads/sites/52/2018/03/ISU-Brent-macro-defect-free_Final.pdf

There are such things as engines made of ceramics rather than metal.
https://en.wikipedia.org/wiki/Ceramic_engine

There is such a thing as “powdered alcohol.” You add it to a beverage to make it alcoholic, like making Kool-Aid.
https://melmagazine.com/en-us/story/palcohol-controversy

The adoption of widespread tea drinking in England starting around 1700 led to a 12% decline in mortality since more adults were boiling their dirty water before drinking it.
https://eh.net/eha/wp-content/uploads/2016/08/Antman.pdf

An important health benefit of smoking is that it reduces appetite, in turn lowering a smoker’s body weight. From 2002-17, the decline in smoking in the U.S. caused a small increase in the country’s obesity rate.
https://www.nber.org/papers/w29701

For the last 20 years, efforts to create a new Lyme disease vaccine have failed because scientists made a wrong assumption about how an important part of the Lyme bacterium was shaped. Several candidate vaccines failed because they were designed to attach themselves to the wrong shape on the outsides of the bacteria.
https://www.science.org/content/blog-post/lyme-disease-molecule-revealed

At a meeting between French president Macron and Vladimir Putin, the latter refused to take a COVID-19 test out of fear the Russia would use it to secretly sequence his DNA, and use the information against him. Note my relevant prediction, which I said will come true in the 2030s:

“Trivially cheap gene sequencing and vastly improved knowledge of the human genome will give rise to a “human genome black market,” in which people secretly obtain DNA samples from others, sequence them, and use the data for their own ends. For example, a politician could be blackmailed by an enemy who threatened to publish a list of his genetic defects or the identities of his illegitimate children. Stalkers (of celebrities and ordinary people) would also be interested in obtaining the genetic information of the people they were obsessed with. It is practically impossible to prevent the release of one’s DNA since every discarded cup, bottle, or utensil has a sample.”
https://www.reuters.com/world/europe/putin-kept-macron-distance-snubbing-covid-demands-sources-2022-02-10/

The world record for sequencing a human genome is now 5 hours and 2 minutes.
http://med.stanford.edu/news/all-news/2022/01/dna-sequencing-technique.html

Jewish law can be interpreted as supporting human genetic engineering.
https://www.bbc.co.uk/bitesize/guides/zdd7sbk/revision/7

Unsurprisingly, humans attach greater value to the lives of “cute-looking” animals than to ugly ones. Aesthetics even override our concern for animals’ likely cognitive levels and ability to experience pleasure and pain. We’re going to have a very hard time explaining our preferences to AIs someday.
https://www.psychologytoday.com/gb/blog/animals-and-us/202201/why-people-care-more-about-beautiful-animals-ugly-ones

This article explores why some animals can detect earthquakes, volcanic eruptions, and even tornadoes in advance. Once we understand how they do it, we could build sensors that took advantage of the same principles.
https://www.bbc.com/future/article/20220211-the-animals-that-predict-disasters

“Tetrachromats” are people who have four types of photoreceptors in their eyes instead of the usual three, allowing them to see more colors. This article profiles one tetrachromat, and she finds the colors of nature more stimulating than the average person, while color schemes found in many buildings and manmade environments are bracing. I think our descendants will be able to see more colors than we can.
https://www.theguardian.com/society/2022/jan/30/im-really-just-high-on-life-and-beauty-the-woman-who-can-see-100-million-colours

Humans can tell how strong a man is based on the sound of his voice alone. This is true even if the man is speaking a language the listener doesn’t understand.
https://royalsocietypublishing.org/doi/10.1098/rspb.2010.0769

There are important upsides to global warming.
http://www.rationaloptimist.com/blog/how-global-warming-can-be-good/

As a reminder of how unexpected the current real estate price boom is, consider this July 2020 article from The Economist: “Many economists still expect house prices to fall over the whole of 2020—but such forecasts are looking increasingly shaky.”

In other words, as late as mid-2020, there was still no consensus over what direction the U.S. housing market was headed, and even the experts were unaware of how much prices were about to jump. Keep this in mind when reading today’s news articles where those same experts predict what the housing market will do over the next few years. I doubt a big crash in prices is coming, but the possibility can’t be written off.
https://www.economist.com/finance-and-economics/2020/07/02/americas-housing-market-is-so-far-unfazed-by-recession

And here’s an article from the August 22, 2020 edition of The Economist wondering why inflation rates are so low.
https://www.economist.com/schools-brief/2020/08/22/why-does-low-unemployment-no-longer-lift-inflation

From this month: ‘High prices continue to hit American shoppers as inflation rose faster than expected to 7.5 percent for the month of January over the previous year, exceeding the 40-year high set in December.’
https://www.nbcnews.com/business/inflation-numbers-trend-rising-prices-goods-soar-rcna15692

Consider the years of (highly expensive) training men go through just to become fighter pilots, the need for nearly continuous re-training to prevent them from forgetting skills, and the many hours of preparation and after-action work that they must do for each single hour of flight of actual combat missions. It’s amazing but also colossally inefficient. Once someone figures out how to build autonomous fighter planes that can be mass produced, programmed in a few hours at the factory, and uploaded with mission instructions in a few seconds, it will be the end of human pilots. They will be overwhelmed by machine fighter planes that can do missions around the clock, even if the human fighter pilots are better in 1:1 combat against them.
https://www.thedrive.com/the-war-zone/44099/why-fighter-pilots-work-for-eight-hours-intensely-to-only-fly-for-30-minutes

‘When fed information about a target individual’s mobile phone interactions, as well as their contacts’ interactions, AI can correctly pick the target out of more than 40,000 anonymous mobile phone service subscribers more than half the time…The findings suggest humans socialize in ways that could be used to pick them out of datasets that are supposedly anonymized.’
https://www.sciencenews.org/article/ai-identify-anonymous-data-phone-neural-network

Machines can now beat humans at car racing video games.
https://www.npr.org/2022/02/23/1080976330/ai-video-games-sony-playstation-gran-turismo-autonomous-car

A new algorithm, “AlphaCode,” can write computer programs as well as most human programmers.
https://deepmind.com/blog/article/Competitive-programming-with-AlphaCode

Computer generation of images based on text descriptions has improved.
http://arxiv.org/abs/2112.10741
https://youtu.be/eaSTGOgO-ss

One of Google’s most advanced narrow AIs was tasked with finding a better way to losslessly compress internet videos. It did, but its method was only 4% better than what human computer scientists had created. Across a surprising number of technologies, humans have already achieved optimums or come very close to them.
https://deepmind.com/blog/article/MuZeros-first-step-from-research-into-the-real-world

James Cameron now thinks that Skynet would destroy humanity by circulating deepfake videos on the internet to turn people against each other. “The news cycles happen so fast, and people respond so quickly, you could have a major incident take place between the interval between when the deepfake drops and when it’s exposed as a fake.”
https://www.pcmag.com/news/james-cameron-skynet-would-destroy-humanity-with-deepfakes-not-nukes

For the record, I think a hostile AI would use a variety of weapons and tactics against us in a war, including advanced digital disinformation campaigns. Furthermore, the COVID-19 pandemic shows how effective biological warfare would be against us, so we should expect that, and I also think Skynet would build some robots to kill us “kinetically,” though the vast majority of them wouldn’t look like T-800s.

To that end, here is what I expect AI progress will look like if it fits the patterns of past progress.

  • Major new methods or capabilities for AI will be demonstrated in systems that are generally pretty poor.
  • Under the right conditions, such as a multi-billion-dollar effort by a state actor, the first version of an important new AI capability or method may be sufficiently advanced to be a major global risk or of very large strategic value.
  • An early system with poor practical performance is likely to be followed by very rapid progress toward a system that is more valuable or dangerous
  • Progress leading up to an important new method or capability in AI is more likely to be accelerating than it is to be stagnant. Notable advances preceding a new capability may not be direct ancestors to it.
  • Although high-risk and transformative AI capabilities are likely to emerge in an environment of less uncertainty than today, the feasibility of such capabilities and which methods can produce them are likely to be contentious issues within the AI community right up until those capabilities are demonstrated.
    https://www.lesswrong.com/posts/iRhSi5KcnnD9qsAGo/observed-patterns-around-major-technological-advancements

Better narrow AI and better robots could widen the existing wealth gaps between countries and reduce international trade. Rich countries will build robots to do work for them that poor people currently do for them overseas.
https://www.imf.org/en/Publications/WP/Issues/2020/09/11/Will-the-AI-Revolution-Cause-a-Great-Divergence-49734

Most of the world has turned the corner of the COVID-19 pandemic. For example, masking rules are being rescinded across the U.S.
https://www.usatoday.com/story/news/health/2022/02/25/covid-cdc-indoor-mask-guidelines/6937810001/

Review: “The Final Cut”

Plot: At an unspecified point in the future, it has become common for people to implant their children with devices that record everything they see and hear. The implants, called “Zoes” (two syllables), are organic, are implanted at the fetal stage of life somewhere in the central nervous system, and “grow” as the child grows. The implants are unnoticeable, and people are only told they have them once they hit adulthood. For technical reasons, the audiovisual contents of the Zoes can’t be downloaded until after the person dies.

Robin Williams is the main character and protagonist. The film starts with a memory from his own childhood where he is hanging out alone during a day trip to the countryside and encounters another boy his age, who is also alone. The two get on friendly terms and explore an abandoned building together. While walking over a narrow beam, the other boy falls over the edge, lands on his head and immediately dies. Kid Robin Williams could have saved him by grabbing him as he was dangling from the edge, but he hesitated and the boy died. He runs away and never tells anyone else about this traumatic and shameful memory.

Years later, Robin Williams has found work as a “cutter”–a sort of futuristic video editor who downloads Zoe recordings from the recently deceased and then edits them down into two- or three-hour movies that show all the milestones and positive highlights of their lives. These recordings are usually shown at funerals, given to loved ones, and serve as semi-official records of what happened in a person’s life.

Robin Williams editing footage from a zoe

The editing process entails deleting recordings of bad things the person did (like spousal abuse, child molestation, and everyday acts of cruelty), leaving a happy but false representation of the person’s life. Robin Williams’ choice of this profession clearly stems from his own desire to assuage his own guilty memories of the childhood incident. His character’s last name–“Hakman”–brings the symbolism to an even more obvious level.

The movie’s main conflict arises when Robin Williams is asked to cut the Zoe footage for a wealthy businessman who recently died. After reviewing it, not only does Robin Williams realize the man was a secret pedophile, but he also finds clues that the dead boy from his own childhood might have actually survived and crossed paths with the businessman. Added to the mix is the fact that Robin Williams is under a short deadline to do the cut and return the original footage to the family, and a dangerous terrorist group wants to steal the Zoe footage for blackmail purposes.

The year is never revealed in The Final Cut. Also, aside from the Zoes, the film depicts a world identical to our own–there are no flying cars, laser guns, robots, etc. Most people don’t even have stainless steel dishwashers. It’s a cop-out and makes the film more of a fantasy than anything else. By the time Zoes exist, it will be so far in the future that nearly everything about the world will be different from today.

Analysis:

There will be brain implants that record what people see and hear. In principle, this technology is possible and we already have crude versions of it. Implants that can monitor brain activity and turn a person’s thoughts into written text were recently invented to help people with speech disabilities. More advanced implants that monitor the parts of the brain that processed vision and hearing could someday decode the things a person was seeing and hearing. Alternatively, implants could be attached to the optic and cochlear nerves to directly monitor the stimuli being received by the eyes and ears (respectively), before any of it had been processed by the brain.

Safe, affordable central nervous system implants with capabilities like “Zoes” won’t exist until sometime in the 22nd century. However, average people will be able to effectively do the same type of lifelogging by the end of this decade by wearing the new generation of augmented reality (AR) glasses that are coming.

Brain implants will have “organic” characteristics. The Zoes “grew” along with their hosts, and since they were permanent, lasted a lifetime, and didn’t need to be removed for maintenance, they must have had self-healing capabilities and the ability to extract energy from blood or body heat. The devices thus had “organic” characteristics.

Some technologies will eventually gain organic attributes, and it’s clear this would be especially advantageous for devices implanted in “wet” brains and bodies. As one example, storage of digital information can presently only be done using artificial substrates like hard disk drives and flash drives, but scientists are developing ways to do it using DNA, which is an organic molecule. DNA is an incredibly efficient way to store information (a microscopic amount of it in just one of your cells can hold close to 1 GB of data), and existing cellular self-repair mechanisms are excellent at protecting the data contained in DNA from decay. This might be the ideal data storage medium for brain implants considering the enormous amounts of audiovisual data that would need to be saved.

Beyond that, advanced nanomachines and/or micromachines could fully bridge the gap between organic and synthetic since they would be artificial microorganisms and would allow macro-scale machines to grow, heal, and to move their parts in totally organic ways. Some robots will have supple bodies and will be made of what could be thought of as “artificial cells,” and some humans will have synthetic implants and body parts that look biological and have some properties of organic tissue. The line between “natural” and “artificial” might disappear, leading to life forms combining the attributes of both in refined ways.

Of course, that milestone won’t be reached anytime soon. Again, we’ll probably have to wait until the 22nd century to see this level of technology.

People won’t be able to control their own implants. Another two of the film’s conceits are that people can’t turn their own Zoes off or view the footage they have captured. Only after a person dies can the footage be downloaded (presumably, this involves brain surgery) and viewed (by other people).

Things will never turn out this way. Users will always demand control over their devices and their data privacy, and they will find Zoes useless if they can’t view their own recordings. Actual brain implants we create in the future will be able to transmit and receive data to and from external devices, and will also have simple features allowing users to do things like delete and play back recordings, or temporarily deactivate. (Also consider the legal, employment, and social consequences for a person if it were known that he was always recording everything he was experiencing.) If, for some reason, brain implants lacked these features, then people would instead use AR glasses for their lifelogging.

Machines will be able to recognize what is happening in video footage. A scene I really liked in The Final Cut was where Robin Williams used his computer to scan through the wealthy businessman’s Zoe footage. The data file is thousands of hours long, and the computer rapidly shuffled through every second of it, recognized what the dead man was doing each moment, and categorized each clip appropriately. It automatically sorted clips into groups like “Eating,” “Watching TV,” “At work,” “Walking around,” and “Having sex.” With the basic level of sorting completed, Robin Williams could then go through the clips and use his human judgment to select the ones best representing the man’s major achievements, milestones, and positive traits.

Well before Zoes are invented, computers will become smart enough to do this. In just the last five years, major progress has been made teaching machines to understand what’s going on in video footage, to accurately transcribe speech and recognize sounds, and to identify people through biometrics. Within ten years, a person will be able to upload his lifelogging footage from his AR glasses to a computer and have it sorted with the same speed, accuracy and thoroughness as Robin Williams’ computer. They will even be able to identify locations based on visible landmarks and other clues, and to make other intelligent inferences about the contents of clips.

Far from being a parlor trick or something that is only useful to obsessive-compulsives, this technology could help ordinary people. For example, in 10 years you could ask: “Who was that guy in the white jacket that I talked to at that party last week?” and your AR glasses will understand your spoken question, scan through its stored footage, and answer you, perhaps also offering an instant replay of the episode. It will be like having superhuman memory.

Parents will put implants in their newborn children. In the film, Zoes are implanted in their hosts in early childhood, meaning the decision is made by a host’s parent. It may sound unrealistic for parents to have unnecessary brain surgeries done on their children, but once Zoe-like devices are cheap and surgical techniques are more advanced, it could become common. It might be considered a great blessing for parents to enable their kids to re-live episodes from their childhoods later on. Just don’t expect any of this until the 22nd century.

The FDA just approved the “Eversense E3” glucose monitoring implant. It is surgically placed in a person’s upper arm and can remain there for up to 180 days. More advanced and longer-lasting implants are sure to come.

What might become common much sooner is the installation of health monitoring implants in children. The devices would be smaller, simpler and cheaper than Zoes and would be placed in less vital parts of the body than the brain, making surgery far less risky. Such implants could monitor vital functions (e.g. – heart rate, blood pressure, respiration, temperature, cholesterol, hormone levels, diet, gene expression) and alert parents and doctors to health problems in their earliest phases, and to sudden medical emergencies. The implants might even double as location trackers for use if the children became lost or were kidnapped. If the price and risk are low enough, and the benefits are high enough, the natural parental instinct to do everything to protect one’s children could lead to monitoring implants becoming common in a few decades.

Will tech implants ever be worth it?

But in the interim, body-worn devices will satisfy those functions. As discussed in my Cloud Atlas review, external devices can do most of the same things implanted devices could, but at lower cost and without need for surgery. In my analysis of Ray Kurzweil’s 2019 predictions, I explained how smart watches had become affordable and could continuously monitor many of their wearers’ vital signs, warn them of irregular heartbeats, and alert the local paramedics if they detected “hard falls” followed by user nonresponse. More features, like blood pressure monitoring, will be added with time. Smart watches can also be used as tracking devices.

In my analysis of how accurate my predictions for the 2010s were, I also calculated that it was feasible in 2020 for an average person to record every waking hour of his life with a GoPro, and at a respectable 720p video resolution. The cost of storing the footage would be only $1 a day, putting the whole system well within the financial means of most people in rich countries. Of course, that would require the person to strap a small box to his forehead, which would look so silly few would do it. However, the new generation of AR glasses that will be commercially available by the end of this decade will be sleek and stylish, and have unobtrusive cameras. Hard disk prices will also keep declining, meaning it won’t be long until it costs mere pennies a day to store videos of one’s waking life.

With that in mind, AR glasses that give people the same audiovisual recording abilities as the Zoe brain implants will be affordable and available by the end of this decade. Smart watches that can closely monitor their wearers’ health and provide them with significant medical help will be available around the same time. Improved computer algorithms will be able to pool and analyze all of the data gathered by a person’s various devices to detect patterns and make sophisticated inferences. For instance, it could correlate your early-afternoon headaches with your cup of yogurt at breakfast, and inform you that you are probably going lactose intolerant. Your devices could give you real-time summaries of your health status and make hourly activity recommendations based on the day’s data (“Go for a walk”…”Breathe deeply to calm down”…”Take your medication”).

And very importantly, putting on these or other body-worn devices won’t require surgery, and if they ever broke or became obsolete, you could simply take them off and and throw them away. That won’t be true for body implants. So are cyborg implants merely another poorly conceived sci-fi trope, like laser pistols, which will never materialize?

No. Body implants like Zoes will ultimately make sense for humans to get, and will have important advantages over body-worn devices, but it will take a long time for the implants to become common.

AR glasses can only record what you are seeing and hearing, not what you are tasting, smelling, or feeling on your skin. Only a brain implant like a Zoe could capture those senses, as well as your moment-to-moment emotional states. If you wanted to truly re-live happy memories, an implant would be needed.

And while smart watch technology will reach impressive heights, it will be handicapped by its inability to access the wearer’s bloodstream. Devices inside a person’s body could monitor hormone levels, glucose levels, immune system activity, gene expression, toxin levels, and other important metrics, in addition to doing everything smart watches do. Implants could even stimulate your body with things like electric shocks to your heart, hormone dumps into your bloodstream, or neurotransmitter releases into your brain to counteract health problems. Even without any future cures for diseases or breakthroughs in reversing the aging process, such devices by themselves would significantly improve public health and lifespan.

These and other cybernetic devices will migrate into our bodies once we have found ways to make them totally unobtrusive and reliable, and once the cost and invasiveness of surgery dramatically improves (robot surgeons that work for free might help). Some limited ability to self-repair and to internally reconfigure to account for technology updates will also be needed, and the radically advanced nature of such technology is is why I don’t see the cyborg era dawning this century.

Four final points that weren’t covered in the film:

  • Ubiquitous surveillance will reduce bad behavior. If people know they’re probably being recorded and the recordings will be stored forever and possibly shared with millions of people, they’re less likely to commit crimes or behave uncivilly. The effect is greater if they know that biometric analysis like facial recognition or voice recognition can easily uncover their real identities from video footage. Thanks to everything being recorded and to the world being populated by intelligent machines and posthumans that will lack berserker emotions and extreme stupidity, the 22nd century will probably be a very polite era.
  • Having implants in your brain and body that monitor your surroundings, your behavior, and your physiological state could lead to a spooky condition where your personal assistant AI that is watching them could anticipate your thoughts, actions, and needs. If gifted with high enough intelligence and tasked with furthering your long-term enlightened self-interests, your AI could find clever ways to nudge or even control you. As a simple example, it might act like an angel on your shoulder and tell you through your ocular nerve “Don’t eat that pie. You’ve already consumed 2,300 calories today. You get a break on your health insurance premiums if I report you’ve been eating well.” More paternalistically, it might be able to release synthetic dopamine into your brain to calm you down from fits, or just plain take over your body if you were doing something highly illegal or self-destructive. Mind-influencing and mind-control could, along with ubiquitous surveillance, give rise to a very peaceful and harmonious world (or a dystopian one).
  • Ubiquitous surveillance will create interesting tensions between peoples’ memories and what actually happened. The film touches on this when the brother of a recently deceased man remarks to Robin Williams that the video clip of a childhood boat trip was at odds with his own recollection. It’s beyond the scope of this essay to discuss this issue in depth, but the replacement of fuzzy human memories with clear, unchanging recordings will be a two-edged sword. Past traumas and failures would never be forgotten, but people would also be able to see their own actions through unbiased lenses and to see themselves in a more honest light.
  • There will be “snitch apps” in the future. Once people have AR glasses, they will be able to download apps that automatically compare the faces of every person they encounter with mugshots of all known criminals and terrorists, and then report sightings to law enforcement. Even if just 0.1% of the population used these when in public, it would be highly effective. There might even be crowdsourced “Wikis” of non-criminal rude people (ex – “Karens” who had public outbursts made notorious by YouTube) whom you could also set your devices to look out for and to highlight for your avoidance or mockery. Likewise, your own reputation would be viewable to other people wearing their own AR glasses.

Links:

  1. “AI video analytics” is a rapidly developing field of technology devoted to improving machines’ ability to understand what they are seeing in video footage.
    https://www.machinedesign.com/automation-iiot/article/21171867/the-rise-of-ai-video-analytics
  2. Implants that can do simple functions like monitoring blood cholesterol levels already exist. As they get cheaper, smaller and better, they will get more common.
    https://www.bbc.com/news/health-21841829
  3. DNA can be used to store computer data.
    http://www.nytimes.com/2013/01/29/science/using-dna-to-store-digital-information.html?_r=0

Interesting articles, January 2022

The “Jalisco New Generation” cartel used a drone to drop bombs on people in Mexico, and then released the drone’s camera footage. Weaponized drones will get better, cheaper, and available to an ever-growing number of people as time passes.
https://www.dailymail.co.uk/news/article-10394093/Terrifying-footage-shows-people-running-lives-cartel-controlled-DRONE-drops-bombs.html

On the second anniversary of Iranian general Qasem Soleimani’s assassination by a U.S. drone strike, two Iranian drones attacked an American military base in Baghdad. Both were shot down by a U.S. antiaircraft system called “C-RAM.” A C-RAM unit has a built-in radar that identifies the locations and flight paths of enemy aircraft and missiles, and it uses the data to aim its heavy machine gun so the bullets intercept them. All the human operator has to do is push a button to allow the C-RAM to fire. At the 0:26 mark in the video, the C-RAM opened fire on one of the Iranian drones. Note the laser-like stream of bullets. The machine’s aim was perfect.
https://youtu.be/Ajkg8yfgug0

Contrast that with footage of human-aimed antiaircraft guns trying to shoot down Japanese kamikaze planes in WWII. You’ll see dozens of machine guns spewing out thousands of bullets at one plane, and missing–often being wildly off-target.
https://youtu.be/4mTECUWP0Hk?t=171

120,000 Russian troops have massed near Ukraine for a possible invasion of that country. A frenzy of diplomacy is now underway to head off a confrontation. Putin has said he wants several assurances from NATO countries to bolster Russia’s national security.
https://www.cnn.com/2022/01/14/politics/us-intelligence-russia-false-flag/index.html
https://www.thedrive.com/the-war-zone/43942/flotilla-of-russian-amphibious-warships-has-entered-the-english-channel

This analysis concludes that the human brain is actually a very efficient type of computer, and AIs brains will need to be similar to ours.
https://www.lesswrong.com/posts/xwBuoE9p8GE7RAuhd/brain-efficiency-much-more-than-you-wanted-to-know

As impressive as GPT-3 is, it’s still easy to confuse it with well-chosen questions. It’s nowhere near passing the Turing Test.
https://statmodeling.stat.columbia.edu/2022/01/13/chatbots-still-dumb-after-all-these-years/

At the same time, GPT-3 is a much better chatbot than the best of ten years ago. I think each iteration of GPT will make fewer conversational mistakes than the last, until some future version (GPT-6?) is finally “good enough” to pass the Turing Test.
http://lacker.io/ai/2020/07/06/giving-gpt-3-a-turing-test.html

Once “Metaverse” technology is mature and widespread, it will be common for sports fans to “hang out” in the bleachers of sports arenas during games. They could even pay for retired athletes and other popular commentators to sit with them in VR and narrate the games. Other fans will still be willing to buy tickets to sit in the real bleachers and attend the games in-person.
https://futuristspeaker.com/business-trends/when-nfl-football-moves-into-the-metaverse/

Elon Musk says Tesla’s humanoid robot will probably be 5ft 8in robot, have a screen on its face, be able to lift 150lb, and walk at 5 mph. This mostly matches my own predictions.
https://www.bbc.com/news/technology-60154782

Machines can now convert a series of still photos of a building or place into a hi-res, 3D mock-up. Eventually, there will be a hyper-realistic, 1:1 virtual version of the real world that people will be able to explore in VR.
https://youtu.be/yptwRRpPEBM

The once-ubiquitous BlackBerry cell phone has been thrown into the dustbin of history.
https://www.cnn.com/2022/01/01/tech/blackberry-end-of-life/index.html

Apple is now the world’s first $3 trillion company. It hit the $1 trillion mark only three years ago. My prediction that the first trillionaire would be minted in the 2050s might be too conservative.
https://www.cnn.com/2022/01/03/investing/apple-three-trillion-dollar-market-cap/index.html

“Frequency hopping” is a method of sending encrypted messages with radio signals. A message like a simple sentence is chopped up into bits, each of which is transmitted on a different frequency from the other. To anyone listening to just one radio wavelength, they will only hear a single, brief sound of the message. However, the intended recipient will hear the whole thing thanks to a special radio descrambler that knows how the bits are distributed across the frequency spectrum.
https://en.wikipedia.org/wiki/Spread_spectrum

Dial-up modems from 20+ years ago used sounds to send and receive data. This is why, if you established the connection and then picked up the phone, you’d hear loud static for a moment–the static was the digital data being conveyed as sounds. Telephone lines weren’t built with the future needs of the internet in mind–they were designed around the much less demanding needs of human speech and listening. As a result, they can’t handle more than 56 kilobytes of data transfer.
https://www.10stripe.com/articles/why-is-56k-the-fastest-dialup-modem-speed.php

Thanks to sound transmission of data, people use to also download video games from their radios.
https://interestingengineering.com/you-could-download-video-games-from-the-radio-in-the-1980s

You might remember BetaMax, but you’ve probably forgotten (or never heard about) “V2000.”
https://en.wikipedia.org/wiki/Video_2000

What old sci-fi films imagined 2022 would be like.
https://www.npr.org/2022/01/04/1070110448/2022-movies-past-future-soylent-green

In 1991, George Friedman published a book predicting the U.S. and Japan would fight a war by 2020. This is something to remember when thinking about his other geopolitical predictions, which also look likely to fail.
https://www.youtube.com/watch?v=PrbUX84LcXg

Six months ago, Peter Zeihan predicted “We are gonna have inflation in the last half of this year that is absolutely going to be higher than what we had in the early 80s [and] probably faster than what we had in the 70s.” He was basically right. The U.S. inflation rate for November was the highest since July 1982.
https://youtu.be/x_fpY63fcd8?t=3560

In August 2019, 74% of economists said the U.S. would have a recession by the end of 2021. They were right! The recession happened in early 2020 due to the COVID-19 pandemic.
https://apnews.com/article/economy-donald-trump-us-news-ap-top-news-global-trade-3d77bbfda266497699554a754c124735

Nuclear-powered civilian ships were introduced before the technology was fully ready, and were doomed by irrational public fears about radiation and by unfair press coverage.
https://www.youtube.com/watch?v=cYj4F_cyiJI

Here’s a load of encouraging data from Ramez Naam about the rise of solar and wind power and grid battery storage. He estimates that solar and wind power will generate electricity more cheaply than coal and natural gas in the U.S., China, and most other countries by 2030. The transition has already happened in some places.
https://www.dropbox.com/s/l6qr9x1zhvc4yq7/Naam%20Clean%20Energy%20Revolution%20-%20Chinese%20Edition%20-%20Shareable%20-%20Jan%202022.pdf?dl=0

Theoretically, we could also build antennas that used the planet’s radiation of excess heat into space to generate electricity.
https://www.pnas.org/content/111/11/3927

Theoretically, 10 quadrillion people could live on Earth. We’d just have to build 300,000-storey high skyscrapers to fit everyone, plus a bunch of other megaprojects to radiate the planet’s excess heat into space and regulate global sunlight levels.
https://hereticalupdate.substack.com/p/is-earth-running-out-of-resources

‘The result is that you see a distinct parabolic shape in the returns on investment for a tall building. The point of maximum return varies depending on the city, the type of construction and the location of building, and real estate professionals go to great effort to determine the economic building height for a given case. For an office building on a piece of valuable urban real estate, this has traditionally been considered to be in the neighborhood of 60 to 70 storeys tall. During planning for the Empire State Building, it was calculated that 75 storeys was the optimal height, and developers suggested that 70 storeys should be the maximum during the planning of One World Trade Center. But the existence of an increasing number of Manhattan supertall residential buildings suggests that this limit might be increasing, at least for luxury residential real estate.

Building height in excess of this “theoretical optimum” is often height for height’s sake, with the idea that an exceptionally tall building will have “prestige value” that more than compensates for the less efficient design. The (real or perceived) benefits of prestige, combined with the rising costs of servicing the upper floors, often results in buildings that achieve their height by adding large volumes of unoccupied space at the top. The Burj Khalifa, currently the tallest building, is perhaps the ultimate example of this, with the top 29% of the building being unoccupied space.’
https://www.worksinprogress.co/issue/why-skyscrapers-are-so-short/

The world’s oldest tortoise is 190.
https://www.guinnessworldrecords.com/news/2022/1/190-year-old-jonathan-becomes-worlds-oldest-tortoise-ever-688683

Some people who have become social media stars by uploading videos of their pets have gotten those pets cloned after they died so they can keep making (profitable) content. Will humans someday do the same thing?
https://www.the-sun.com/tech/4497683/instagram-influencer-clone-dog-cat/

The eruption of an underwater volcano near the Pacific island of Tonga produced a dust cloud big enough to cover all of England.
https://graphics.reuters.com/TONGA-VOLCANO/lgpdwjyqbvo/

The Martian meteorite that, in 1996, was claimed to have evidence of alien microbial life, has been proven not to. I remember when this was news.
https://apnews.com/article/space-exploration-science-business-planets-mars-33dc55ce0ebced0b65322b03ae06feda

The Big Five Personality traits predict how well-suited people are for various jobs. “Conscientiousness” level is the strongest predictor of general success.
https://www.sciencedirect.com/science/article/pii/S0001879121001275

Imagine a 1 – 10 rating scale for your full possible range of emotional experiences.

1 = Most miserable and painful you can be
5 = Feel neither good nor bad
10 = Most happy and blissful you can be

People intuitively think that the increments between each number are subjectively constant, so the felt difference between levels 9 and 10 is the same as the difference between levels 5 and 6. However, there’s evidence that the good and bad extremes are way, way, way more extreme, so that going from level 9 to level 10 is much more of a jump in subjective pleasure than going from level 5 to 6 is.
https://www.qualiaresearchinstitute.org/blog/log-scales

With training, most people can learn to control single neurons in their brains. In the far future, I think humans will be able to control their own thoughts, emotions, and gene expression, just by thinking about it.
https://www.biorxiv.org/content/10.1101/2020.05.05.079038v1.full

For the first time, a pig heart has been transplanted into a human without killing the person. The pig had been genetically engineered so its heart cells would be similar enough to human tissue to not be rejected by the human immune system.
https://www.bbc.com/news/world-us-canada-59944889

Genes inherited from Neanderthals cause inflammatory diseases in some people.
https://www.science.org/doi/full/10.1126/sciadv.abm1148

Getting infected with Epstein-Barr virus causes some people to later develop Multiple Sclerosis, a devastating, chronic disease. Developing a vaccine for the virus could ease enormous human suffering.
https://www.science.org/content/blog-post/multiple-sclerosis-causality

A new study adds to the growing body of medical evidence that alcohol consumption causes some kinds of cancer.
https://newatlas.com/health-wellbeing/alcohol-consumption-directly-cause-cancer-oxford-genetic-study/

Antibiotic-resistant bacteria now kill over 1 million people per year. HIV/AIDS and malaria killed 860,000 and 640,000 people, respectively, in 2019.
https://www.theguardian.com/society/2022/jan/20/antimicrobial-resistance-antibiotic-resistant-bacterial-infections-deaths-lancet-study

This article was published in June 2020, a few months into the COVID-19 pandemic. At the time, it was unclear what the effect on birthrates would be, and it could have been argued that more children that normal were going to be conceived since couples would have more time at their homes together. The article predicted the 2021 U.S. birthrate would be 300,000 – 500,000 lower than it was in 2020. The data are still coming in, but there was a decline, closer towards the lower end of their estimate.
https://www.brookings.edu/research/half-a-million-fewer-children-the-coming-covid-baby-bust/

COVID-19 has killed at least 3 million Indians. That’s 6-7 times higher than their government’s official death toll.
https://www.science.org/doi/abs/10.1126/science.abm5154

Review: “Soylent Green”

Plot: Welcome to 2022. Welcome to a grotesquely overpopulated, resource-depleted, polluted, and impoverished world. It’s a place where practically every tree has been cut down and every person herded into cities to make room for farms that nevertheless barely make enough food for everyone, where the air is sticky and thick with toxic smog and the stench of unwashed bodies and corpses, and where the hungry masses are perpetually on the brink of rioting. There’s no joy, hope, jobs, or even real food anymore–just little processed crackers rationed to the population. It’s a place where corrupt politicians and the executives of corporations collude to protect their own power and privileges at any cost, even if it means forcing the ultimate sacrilege on humanity.

Manhattan, 2022

Welcome to New York City. It’s a decaying and crime-ridden cauldron that is so crowded it’s literally standing room only in many of its apartments and streets. Charlton Heston knows this city well, and keeps busy in it. He’s a homicide detective, and of such esteem that he enjoys the privilege of having his own, small apartment, which he shares with only one other person: his elderly assistant named “Sol.” Their dreary routine is interrupted one day when they are assigned to investigate the murder of one of New York’s richest people–a man named “Simonson” who was a Board member at the “Soylent” corporation.

Soylent is an enormous food processing company that controls half the world’s food supply. Their “Red” and “Yellow” products are derived from plants, and are formed into crackers or loaves. Their latest product, “Green,” is said to be derived from plankton harvested from the ocean. Soylent Red, Yellow and Green are staple foods for New Yorkers, and probably billions of people beyond.

Soylent foods for sale at a New York market

As the investigation proceeds, Heston quickly realizes Simonson’s murder was no robbery gone bad, as it appeared at first glance. As he and Sol follow the clues, it leads them to mortal danger, a conspiracy involving some of the world’s most powerful men, and to a profoundly disturbing secret about the food supply.

Soylent Green was a laugh-out-loud inaccurate portrayal of the world in 2022. Yeah, I know we have our problems, but they don’t compare to the film’s dystopia. The fact that it was so far off the mark should be FOOD FOR THOUGHT for anyone who takes the current crop of doomsday global warming movies set in the future (e.g. – Geostorm, Snowpiercer, Interstellar) seriously.

That said, I still liked Soylent Green and think it’s worth watching so long as it isn’t taken seriously. The movie is well-paced and manages to depict a grim future without overdoing it to the point of being depressing. It’s both entertaining and serious, and at times genuinely tense. The acting is great all around, especially on the part of Charlton Heston, who is less cocky and has a slightly broader emotional range in this than in most of his other roles.

Analysis:

The world will be grossly overpopulated. At the beginning of Soylent Green, we’re told that New York City’s population has grown from roughly 8 million the year the film was released (1973) to 40 million in 2022. Population figures for other parts of the U.S. or for other countries are never given, but at one point Heston says other cities are “all like this,” implying the rest of the world is similarly overpopulated.

The U.S. population in 1973 was about 205 million, and the world population that year was 3.7 billion. If they quintupled like New York City, then in the film, the U.S. population in 2022 was 1 billion, and the world population was 18.5 billion.

Mercifully, the real figures are much lower: New York City has 8.8 million residents, the U.S. has 330 million, and the world has 7.9 billion. Soylent Green‘s prediction that Earth would be grossly overpopulated by 2022 was wrong, and the city in which it is set, New York, has only 11% more inhabitants now than it did in 1973. Instead of it being “standing room only,” the city is but marginally denser.

Outside of the tropical countries and Muslim world, population is growing very little and is even shrinking

Ironically, a growing number of thinkers and journalists today are worried about the opposite problem: population decline. The populations of rich countries are mostly shrinking, or are only slowly expanding thanks to immigration and immigrants having kids. Even middle income countries like China, Thailand and Brazil have seen sharp drops in birthrates and have almost stopped growing. While shrinking a shrinking population has benefits (more space per person, cheaper real estate, less traffic, less pollution created), they are probably outweighed by the downsides of economic decline.

That said, it would be a mistake to simply extrapolate current demographic trends into the future indefinitely and to conclude that the human race is doomed to extinction because people will refuse to have kids. A slew of technologies that will come into existence this century will raise birthrates in various ways: Existing assisted reproductive technologies like in vitro fertilization (IVF) will get cheaper, putting them within reach of lower income people. New reproduction technologies will be invented, allowing more people with fertility problems to have healthy kids. For example, post-menopausal women with no eggs will be able to have fertility labs synthesize ova for them that contain their DNA, and to insert it into themselves, younger surrogate mothers or, in the far future, artificial wombs. Robot servants will also ease household workloads, giving parents more time for child-rearing and making parenthood more appealing.

Along with raising birthrates, future technologies will let us grow the human population through the opposite mechanism, which is lowering mortality rates. Disease cures, therapeutic cloning of human organs, cybernetic replacements for organs and limbs, stem cell therapies that regenerate ageing tissues and organs inside the patient’s body, and many other medical advances, will slowly raise lifespans, and to such an extent that “medical immortality” will probably be available to well-resourced people by the end of this century. If people don’t die, then even a very low birthrate among them will lead to Soylent Green levels of overpopulation, though it might take centuries.

The environment will be devastated by pollution. The other aspect of Soylent Green‘s dystopian reality is severe pollution and concomitant environmental devastation. The outdoor scenes–which are already bleak-looking since they are full of derelict buildings, trash-strewn streets and crowds of poor people–are shrouded in a sickly greenish haze, which is certainly smog. New York City is devoid of trees, except a few saplings in a small, sealed arboretum (presumably necessary to protect them from air toxins) that only privileged people can enter.

A sealed arboretum containing NYC’s only trees

The oceans are also so poisoned and overfished that plankton are the only remaining edible sea life. The Soylent company processes harvested plankton into green crackers for human consumption, and the film’s big reveal is that it has been secretly transitioning their content to human flesh because even plankton is dying out. In other words, “SOYLENT GREEN IS PEOPLE!”

Manhattan in 2021

This depiction of 2022 is almost totally wrong. New York City still has trees growing outdoors–notably in the massive Central Park. Additionally, the U.S. actually had more trees in 2021 than it did in 1921! The amount of global tree cover also increased by 8% from 1982 to 2016.

Instead of disappearing, global seafood harvests have risen since Soylent Green was in theaters, and there are no signs of an impending collapse of wild fisheries, though fish catches have been flat since the 1990s, suggesting we’ve reached the limit of how many wild calories the seas can sustainably provide us. Fortunately, the human race has proven itself more competent at surmounting this barrier than it was in the movie, and a large and growing share of fish are now “farmed” instead of caught wild.

Though the oceans still supply us with plenty of calories, a large and growing share of seafood comes from “fish farms,” labeled “aquaculture” in this graph.

New York City’s air is not full of smog, and its air quality is in fact substantially better than it was when the film was released. As just one example, sulfur dioxide (SO2) concentrations in the City’s air have sharply dropped, from an average of 155 μg/m3 from 1970-72, to a mere 6.8 μg/m3 today (January 24, 2022). (SO2 is the main component of “smog,” and has an opaque appearance. It causes respiratory problems and acid rain.) Every other type of air pollution (i.e. – PM 2.5, ozone, lead, nitrogen dioxide (NO2)) has sharply dropped in New York City, the rest of America, and the rest of the developed world over the same timeframe, meaning they breathe cleaner air today than people did when Soylent Green was in theaters. This is due to a slew of environmental laws being enacted, including the U.S. Clean Air Act of 1963 and the Clean Water Act of 1972. (U.S. air and water pollution levels had actually been trending down for a short time before Soylent Green‘s 1973 release.)

Delhi, India during its November 2021 smog emergency

Unfortunately, those things aren’t true for the poorer half of the global population, and hundreds of millions of people in India and China endure toxic air, mostly due to weak air pollution laws or to lax enforcement of relevant laws. In fact, in November 2021, Delhi had a smog emergency lasting several days, during which the air became so poisonous that the government shut down the city’s schools. The news images of opaque air, crowded streets, poverty, and decay bear striking similarities to the dystopian New York of Soylent Green. The suffering of people in polluted places like northern India is why I judged “This depiction of 2022 is almost totally wrong.”

Winters in temperate areas will be warm thanks to global warming. Though the movie indicates it is set in the year 2022, no clues are given about the exact dates of its events. Based on the facts that most of the characters wear light clothing, and there are several scenes where they are visibly sweating, it would seem it is set in the summer. However, that assumption is upended by a remark Heston makes when contemplating whether to turn on an air conditioner (a rare luxury): “All the way up. We’ll make it cold. Like winter used to be.”

Evidently, global warming has gotten so severe that even in places with slightly cold climates like New York City are hot in the winter!

Fortunately, this prediction about 2022 also fell flat. Global warming has only had a tiny effect on the city’s temperature. According to NOAA data taken from a weather station that has been operating in Central Park since 1869, NYC’s average temperature for all of 1973 (the year Soylent Green was released) was 56.1°F, and the average for that December was 39.0°F. The average temperature for 2020 (the last year for which full data have been published) was 57.3°F, and that December’s average temperature was 39.2°F.

And on the day I analyzed this prediction (January 26, 2022), New York City’s high temperature was 29°F, and it was bracing for a major snowstorm.

There will be tablet computers. Though we never get a good look at them or see how they work, there appear to be simple tablet computers and PDAs in the film. Heston keeps one of them in his apartment, and in the film’s first scene, Sol reads notes about criminal cases off of it. The device is a piece of transparent plastic, about the size and shape of a magazine, with an opaque layer embedded within it bearing written characters.

Heston’s tablet computer, displaying Sol’s suicide note

It is strongly reminiscent of an actual tablet computer that lets users handwrite digital notes on its screen by using metal styluses. This prediction about 2022 was right.

The “ReMarkable 2” tablet, displaying something more cheerful than a suicide note. It is new for 2022.

People will have computer game machines in their homes. Early in the film, there’s a scene set in Simonson’s luxury condo suite. There we see an arcade-style video game. To be exact, it is “Computer Space,” which was the first commercially successful video game in history, and only made its debut two years before Soylent Green was released.

A privately-owned computer arcade game

In 2022, it is very common for people to have video game consoles in their homes and to play games on their computing devices. If anything, the film’s prediction is too conservative since it depicts video games as being only available to rich people, whereas in reality, even a teenager working a part-time job today could afford a quality console and several games.

The government will ration essential goods. Due to dwindling natural resources, an excessive population, and widespread poverty that leaves most people unable to afford anything, the government rations essential goods, notably food and water. Citizens visit government offices where clerks give them their allotments of money or ration cards, which they exchange with other people in New York to get essential goods. In other scenes, we see private merchants selling Soylent food products in an open-air market, and men in official uniforms using an outdoor water tap to fill the jugs belonging to people who need their daily water rations. The film also implies that other basics, like soap, writing paper, and pencils, are also very hard to get.

A rationing office run by the government

For the U.S. and the developed world more broadly, this is inaccurate. Staple foods, potable water, and everyday items like soap are very cheap. For example, by cooking their own meals at home, an adult could easily get their food budget under $10 per day, and by drinking only tap water or some type of beverage mix like “Tang,” get their daily drink budget below $1. A bar of personal soap cost $1.50, and will last a person for weeks.

A visit to a typical American grocery store in 2022, even in poorer parts of the country, will reveal a cornucopia of food and merchandise at low prices. Additionally, thrift stores are practically everywhere, and are bursting with wide varieties of decent-quality secondhand goods at very low prices. Electronic resources like Craigslist.org, Facebook Marketplace, and Freecycle are also major sources of cheap or even free items available locally. If anything, most of the world is now contending with a surfeit of essential goods, which too often are wasted, thrown out, or allowed to accumulate as unused clutter. Growth of the self-storage industry bears further testimony to this reality. People, Americans in particular, have too much stuff, not too little.

Prostitution will be legal. One of Soylent Green’s main characters is “Shirl” (pronounced almost the same as “Cheryl”), a young woman prostitute who is compensated with free housing and amenities in Mr. Simonson’s luxury condo. The arrangement is legal and accepted as normal, and it is later revealed that the condo building has several other prostitutes, euphemistically termed “furniture,” living in other units. Having a live-in prostitute is an expensive marker of high status, and Heston’s suspicions are raised when, while investigating Simonson’s death, he discovers the latter’s bodyguard has “furniture” in his own apartment in spite of a salary that should be insufficient.

Prostitutes having a party

In real life, prostitution is illegal in New York City, and in the rest of the U.S. except Nevada. There, it is confined to a small number of heavily regulated brothel houses. With varying restrictions, prostitution is legal in about 15 countries, mostly in Europe. Nevertheless, as the revelations about Jeffrey Epstein’s high-end prostitution ring–which included sex parties at his luxury Manhattan townhouse–show, it’s still easy for rich men to buy sex in New York.

A small number of industrial food companies will control the global food supply. “Soylent” is clearly the dominant food producer in the U.S., and perhaps the world. As Sol says after researching it: “Soylent controls the food supply for half the world.” It’s unclear who produces the other half, but other big companies and government agricultural agencies probably dominate it.

A small number of food processing companies own many common food and beverage brands. But does that mean they “control the global food supply”?

The world is certainly full of large, highly profitable food processing companies, but none is so big that it controls anywhere near half of the global food supply. Consider the top ten food and drink companies of 2020, along with their food sales for that year:

  1. PepsiCo, Inc. – $70.3 billion
  2. Nestle – $67.7 billion
  3. JBS – $50.7 billion
  4. Anheuser-Busch – $46.9 billion
  5. Tyson Foods – $43.2 billion
  6. Mars – $37.0
  7. Archer Daniels Midland – 35.4 billion
  8. The Coca-Cola Company – $34.3 billion
  9. Cargill – $32.4 billion
  10. Danone – $26.9 billion

If we assume that these ten companies produced all the calories consumed by all humans in 2020, and use revenues as a proxy for calories each produced, then the largest, PepsiCo, only controls 15.8% of the food supply.

Of course, the top 10 food processing companies aren’t really the only ones in existence. The source from which I got the above data actually lists revenue figures for the top 100 companies in the sector. If we include them in the calculation (BTW, rank #100 goes to the “Kewpie Corporation,” which made $3.6 billion in 2020 selling mostly mayonnaise, salad dressing, and baby food in Japan), then big companies sold $1,316 billion of food and beverages in 2020, and the biggest one, PepsiCo, only controls 5.3% of the global market. The top ten combined only control 33.8%.

The darkness of the country indicates what share of its population is engaged in sustenance farming.

Additionally, sustenance farming and the consumption of food made by small, local farms still provides most of the calories for large fractions of the population in Africa and southern Asia. These people eat little or nothing made by the big food processing companies, meaning PepsiCo’s control over global calories should be even lower than the paltry 5.3%.

In rich countries with declining culinary traditions, like the U.S., it is probably common for people to get most of their daily calories from processed foods. However, the foods are still made by several different, competing food processing companies, so there is no monopoly and hence no real-world equivalent to “Soylent.” Even if the biggest one of those companies decided to start secretly blending calories derived from corpses into its food products, only a minority of the U.S. population would end up eating it.

New York City’s population will be 90% white. All of Soylent Green‘s main characters and seemingly 90% of its extras are white. This includes rich, working-class, and poor people.

The reality is very different. The U.S. Census estimated that, in 2021, only 32.1% of New Yorkers were both white and non-Hispanic. Blacks were 24.3%, Asians were 14.1%, and multiracial people were 3.6%. It is surely one of the most racially diverse cities on Earth.

There will be mass unemployment. In the first scene, Heston remarks “There are 20 million guys out of work in Manhattan alone.” Even if this is exaggerated and the real number is only half that figure, and even if “guys” refers to both sexes, it would indicate a staggeringly high unemployment rate.

To be generous, let’s assume that Soylent Green‘s New York had an excellent dependency ratio of 80, meaning 80% of its population was in good health and able to work (children, old people, and disabled people comprise the other 20%). For comparison, NYC’s actual dependency ratio in 2021 was 54.7, and dependency ratios in the 80s have only happened after periods of extraordinary population growth, such as when the post-WWII baby boom generations in India and South Korea hit adulthood.

Eighty percent of 40 million is 32 million, meaning there were 32 million potential adult workers in the city. If 10 million of them (half of Heston’s figure) couldn’t find jobs, that equates to a 31.25% unemployment rate. To put that into perspective, during the Great Depression, the U.S. national unemployment rate peaked at 24.9%. Remarkably, even with optimistic assumptions, the job picture was worse than it had ever been in real life!

What happens if we adjust the calculations to be more bleak? For example what if we lower the dependency ratio to 65 (many of the New Yorkers looked unhealthy and seemed to have motivation problems, both of which would leave them unable to work) and accept Heston’s “20 million guys out of work” figure?

We get a 76.9% unemployment rate, which is unheard of. I can’t imagine a situation where that many willing people wouldn’t be able to find jobs, except maybe the first few weeks following a massive nuclear war. That said, I foresee a day when 76.9% of healthy adults won’t have gainful jobs due to machines doing the work for them, but most of those people won’t be “unemployed” since they’ll embrace (or at least, deal with) the new reality by devoting their time to things other than work, like socializing, video gaming, doing drugs, traveling, or indulging in personal hobbies and niche interests. You don’t count as “unemployed” if you’re not interested in working.

Oh, and what’s New York City’s actual unemployment rate? In December 2021, it was 8.8%, which is high by real-world U.S. standards, but absolutely stellar by Soylent Green‘s.

There will be mass homelessness. Along with lacking jobs, most of the people in the film seem to lack homes. Every morning, Heston has to literally jump over poor people who sleep on the staircase of his apartment. Many of New York’s streets are clogged with broken-down cars that people live in, and sleeping people literally cover the whole floor of his local church at night. Most of the city’s population might be chronically homeless.

Heston jumping over poor people

In reality, no more than 1% of New York City’s population is truly homeless, meaning they either sleep in public spaces or in homeless shelters. And unlike in Soylent Green, most of them only go without proper housing for brief lengths of time, and aren’t “chronically” homeless.

New York City will have epidemic levels of violent crime. Soylent Green begins with a murder, later in the film there’s a street riot where several police officers are attacked and people are shot, and in one scene, the police chief says there were 137 murders in the city over the previous 24 hours. In short, New York City is extremely violent. How accurate was this depiction?

If we assume 137 murders a day is typical, that’s equivalent to 50,005 per year, and a homicide rate of 125 per 100,000 residents. In reality, New York City had 485 murders for all of 2021, meaning its homicide rate is a mere 5.5 per 100,000 residents.

Among big American cities, the most murderous is Louisville, Kentucky, which had 188 murders in 2021, equating to a homicide rate of 30 per 100,000. That means no major urban area in the U.S. comes close to being as violent as Soylent Green‘s New York was.

That said, there are cities outside the U.S. that approach its heights of murder. In 2020, three Mexican cities–Celaya, Tijuana and Ciudad Juarez–had the highest murder rates in the world, at 109, 105, and 103 murders per 100,000 residents, respectively. So if the movie had been Soylent Verde and set just one country away, it would have been grimly accurate in this regard.

People will have battery banks in their homes. The small apartment that Heston and Sol share has a bank of what look like car batteries for storing electricity. A stationary bicycle connected to the batteries can be pedaled to recharge them. It’s unclear whether the battery bank is their sole source of electricity, or if it’s merely a backup power source in case of grid failures, and it’s also unclear how common the batteries are in other homes.

The battery bank

Batteries are much cheaper and more energy-dense today than they were when Soylent Green was in theaters. However, home battery banks remain uncommon due to the reliability of the electric grid and because the batteries are still too expensive to be worth it.

For example, a typical American home consumes 30 kilowatt hours (kWh) of electricity per day. A person who valued efficiency could reasonably reduce that to 24 kWh / day by buying high-efficiency appliances and by doing things like wearing sweaters instead of turning the heat up so high in the winter. A typical home storage battery such as the “Growatt 6 KW,” costs $4,490 and can only store 6 kWh of electricity, so four of the batteries would be needed to store just one day’s worth of power, for a total cost of $17,960, plus installation costs. The batteries’ storage capacities also degrade with time, meaning they usually need to be replaced after 10-15 years.

The “Growatt 6KW” residential battery

A better option for backup power is a gas-powered generator. While portable generators with wheels are the most familiar versions of the machines, the types generally used for residential backup power are stationary and look like large boxes right outside the houses they provide power to. One high-quality standby generator capable of meeting the 24 kWh / day requirement is the “Generac 72101,” and it costs $5,997 plus more for installation. It is connected to the house’s natural gas plumbing and automatically turns on whenever it detects an electrical grid outage. Best of all, if properly maintained and not overused, such a generator can last 20 years or more before needing replacement.

A Generac 24 kW backup generator installed outside a home

This means a home battery backup system costs three times more than an equivalent backup gas generator. Battery prices will need to drop by 66% to achieve parity. Such an improvement might be possible: Between 2010 and 2019, lithium-ion battery pack prices dropped 87%. However, the rate of yearly cost-improvement declined over that period and continues to do so, suggesting we’ve picked the low-hanging fruits for improving battery cost-performance, so don’t expect another 87% decline over the next 10 years. To get our 66% improvement, which might cause battery banks to become common in houses and apartments, I think 20 years or more of research and industrial efficiencies will be needed.

Assisted suicide will be legal. Discovering the awful truth about Soylent Green pushes Sol–already an old and world-weary man–over the edge, so he signs up for assisted suicide, which is euphemistically called “Going home.” Not only is it legal, it is barely regulated, and Sol merely has to walk into the nearest euthanasia clinic and sign a form to have it done. There’s no wait time, no “cool down period,” and no requirement for suicide requests to be vetted by a court, doctors, mental health specialists, or the applicant’s family.

Sol committing assisted suicide

This depiction of 2022 was partly accurate. Physician-assisted suicide is legal in 10 American states and Washington, DC. While the laws only allow their residents the right of suicide, it is easy for people from other parts of America to satisfy the requirement by moving in and living there for a short period of time.

Additionally, in those 10 states and DC, the applicant must provide medical evidence that he probably has six months or less to live thanks to poor health, and there are processes for adjudicating that evidence. (In effect, legal doctor-assisted suicide is available to anyone in the U.S. who can prove he has six months or less to live.) Professing that one is sick of living–even if the person can prove they are sincere–is insufficient. This means Sol, were he alive in the real world of 2022, would not be able to commit assisted suicide.

The procedure is also not legal in New York, though it is in neighboring New Jersey, and it’s possible the euthanasia clinic in the film was in the latter state. Less than a mile of water separates Manhattan from Jersey City, and Sol could have easily made the journey.

Cannibalism will be widespread. Like “Luke, I am your father,” the line “Soylent Green is people” has long been in our cultural consciousness, and is known even to those who haven’t seen the latter film. With that in mind, I feel no guilt exposing the movie’s climactic reveal: the Soylent company has been secretly turning corpses into crackers that millions (possibly billions) of unsuspecting people have been eating.

Soylent Green crackers been scooped into a bag at a food market

Again, and very fortunately, this prediction was wrong. Cannibalism is not widespread in 2022, or even practiced by anything but a miniscule number of disturbed people. It is probably as culturally taboo as it was in 1973, and even in rare cases where a person voluntarily allows themselves to be killed and eaten by a cannibal, the latter is arrested and charged with a crime.

However, as I’ve predicted, in vitro meat technology should be advanced enough by 2100 to let us grow human flesh and organs in labs, which would provide people a legal way to indulge in “cannibalism” without breaking laws related to murder or desecration of a corpse. As a result, a small number of people will eat human flesh, mostly for novelty, like how people try weird meats like alligator today, but some will eat it routinely because they like the taste or have a cannibal fetish.

Links:

  1. U.S. tree cover was higher in 2021 than it was in 1921.
    https://www.treehugger.com/more-trees-than-there-were-years-ago-its-true-4864115
  2. Globally, tree cover rose by 8% from 1982 to 2016.
    https://www.nature.com/articles/s41586-018-0411-9
  3. New York City had several smog crises during the mid 20th century. https://en.wikipedia.org/wiki/1966_New_York_City_smog
  4. “The Relation of Air Pollution to Mortality” (1976) determined that New York City’s average SO2 concentration from 1970-72 was 155 μg/m3.
    https://www.jstor.org/stable/45002384
  5. Delhi’s 2021 smog emergency
    https://indianexpress.com/article/cities/delhi/delhi-smog-high-levels-of-so2-no2-ozone-7618922/
  6. NOAA webpage featuring data from the weather station in Central Park, which has been operating since 1869. It shows how little average temperatures have risen in NYC since 1972.
    https://www.weather.gov/okx/CentralParkHistorical
  7. In the U.S., the self-storage industry has been growing at a healthy rate.
    https://www.nytimes.com/2021/09/21/business/self-storage-roars-back.html
  8. The top 100 food processing companies of 2020 by revenue
    https://www.foodengineeringmag.com/2021-top-100-food-beverage-companies
  9. The FAO’s “World Food and Agriculture Statistical Pocketbook” for 2018 contains data on sustenance farming.
    https://www.globalagriculture.org/fileadmin/files/weltagrarbericht/Weltagrarbericht/10B%C3%A4uerlicheIndustrielleLW/Pocketbook2018.pdf
  10. South Korea’s dependency ratio was 83.3 in 1970, and India’s was 81.2 in 1965.
    https://www.livemint.com/Opinion/3aGTvnsOvqfu22cfQbS4KN/Making-Indias-demography-its-destiny.html
  11. On any given day, about 1% of New Yorkers are homeless, meaning they spent the night sleeping in public or in a homeless shelter.
    https://www.bowery.org/homelessness/
  12. New York City had 485 murders in 2021.
    https://nypost.com/2022/01/01/nyc-recorded-485-murders-in-2021/
  13. Rankings of most murderous cities, 2020
    https://www.eluniversal.com.mx/nacion/mexico-con-el-top-6-en-el-ranking-de-ciudades-mas-violentas-del-mundo-informe
  14. In 2019, the typical American home used 30 kWh of electricity per day.
    https://blog.constellation.com/2021/02/25/average-home-power-usage/
  15. Residential backup batteries typically wear out after 10 – 15 years.
    https://www.pv-magazine.com/2021/09/23/how-long-do-residential-storage-batteries-last/
  16. Between 2010 and 2019, lithium-ion battery pack prices dropped 87%.
    https://www.dropbox.com/s/l6qr9x1zhvc4yq7/Naam%20Clean%20Energy%20Revolution%20-%20Chinese%20Edition%20-%20Shareable%20-%20Jan%202022.pdf?dl=0
  17. Doctor-assisted suicide is legal in 10 states and Washington, DC.
    https://euthanasia.procon.org/states-with-legal-physician-assisted-suicide/

My future predictions (2022 iteration)

If it’s January, it means it’s time for me to update my big list of future predictions! I used the 2021 version of this document as a template, and made edits to it as needed. For the sake of transparency, I’ve indicated recently added content by bolding it, and have indicated deleted or moved content with strikethrough.

Like any futurist worth his salt, I’m going to put my credibility on the line by publishing a list of my future predictions. I won’t modify or delete this particular blog entry once it is published, and if my thinking about anything on the list changes, I’ll instead create a new, revised blog entry. Furthermore, as the deadlines for my predictions pass, I’ll reexamine them.

I’ve broken down my predictions by the decade. Any prediction listed under a specific decade will happen by the end of that decade, unless I specify some other date (e.g. – “X will happen early in this decade.”).

2020s

  • Better, cheaper solar panels and batteries (for grid power storage and cars) will make clean energy as cheap and as reliable as fossil fuel power for entire regions of the world, including some temperate zones. As cost “tipping points” are reached, it will make financial sense for tens of millions of private homeowners and electricity utility companies to install solar panels on their rooftops and on ground arrays, respectively. This will be the case even after government clean energy subsidies are inevitably retracted. However, a 100% transition to clean energy won’t finish in rich countries until the middle of the century, and poor countries will use dirty energy well into the second half of the century.
  • Fracking and the exploitation of tar sands in the U.S. and Canada will together ensure growth in global oil production until around 2030, at which time the installed base of clean energy and batteries will be big enough to take up the slack. There will be no global energy crisis.
  • This will be a bad decade for Russia as its overall population shrinks, its dependency ratio rises, and as low fossil fuel prices and sanctions keep hurting its economy. Russia will fall farther behind the U.S., China, and other leading countries in terms of economic, military, and technological might.
  • China’s GDP will surpass America’s, India’s population will surpass China’s, and China will never claim the glorious title of being both the richest and most populous country.
  • Improvements to smartphone cameras, mirrorless cameras, and perhaps light-field cameras will make D-SLRs obsolete. 
  • Augmented reality (AR) glasses that are much cheaper and better than the original Google Glass will make their market debuts and will find success in niche applications. Some will grant wearers superhuman visual abilities in the forms of zoom-in and night vision.
  • Virtual reality (VR) gaming will go mainstream as the devices get better and cheaper. It will stop being the sole domain of hardcore gamers willing to spend over $1,000 on hardware.
  • Vastly improved VR goggles with better graphics and no need to be plugged into desktop PCs will hit the market. They won’t display perfectly lifelike footage, but they will be much better than what we have today, and portable. 
  • “Full-immersion” audiovisual VR will be commercially available by the end of the decade. These VR devices will be capable of displaying video that is visually indistinguishable from real reality: They will have display resolutions (at least 60 pixels per degree of field of view), refresh rates, head tracking sensitivities, and wide fields of view (210 degrees wide by 150 degrees high) that together deliver a visual experience that matches or exceeds the limits of human vision. These high-end goggles won’t be truly “portable” devices because their high processing and energy requirements will probably make them bulky, give them only a few hours of battery life (or maybe none at all), or even require them to be plugged into another computer. Moreover, the tactile, olfactory, and physical movement/interaction aspects of the experience will remain underdeveloped.
  • “Deepfake” pornography will reach new levels of sophistication and perversion as it becomes possible to seamlessly graft the heads of real people onto still photos and videos of nude bodies that closely match the physiques of the actual people. New technology for doing this will let amateurs make high-quality deepfakes, meaning any person could be targeted. It will even become possible to wear AR glasses that interpolate nude, virtual bodies over the bodies real people in the wearer’s field of view to provide a sort of fake “X-ray-vision.” The AR glasses could also be used to apply other types of visual filters that degraded real people within the field of view.
  • LED light bulbs will become as cheap as CFL and even incandescent bulbs. It won’t make economic sense NOT to buy LEDs, and they will establish market dominance. [Came true in 2021]
  • “Smart home”/”Wired home” technology will become mature and widespread in developed countries.
  • Video gaming will dispense with physical media, and games will be completely streamed from the internet or digitally downloaded. Business that exist just to sell game discs (Gamestop) will shut down.
  • Instead of a typical home entertainment system having a whole bunch of media discs, different media players and cable boxes, there will be one small, multipurpose box that, among other things, boosts WiFi to ensure the TV and all nearby devices can get signals at multi-Gb/s speeds.
  • Self-driving vehicles will start hitting the roads in large numbers in rich countries. The vehicles won’t drive as efficiently as humans (a lot of hesitation and slowing down for little or no reason), but they’ll be as safe as human drivers. Long-haul trucks that ply simple highway routes will be the first category of vehicles to be fully automated. The transition will be heralded by a big company like Wal-Mart buying 5,000 self-driving tractor trailers to move goods between its distribution centers and stores. Last-mile delivery–involving weaving through side streets, cities and neighborhoods, and physically carrying packages to peoples’ doors–won’t be automated until after this decade. Self-driving, privately owned passenger cars will stay few in number and will be owned by technophiles, rich people, and taxi cab companies.
  • Thanks to improvements in battery energy density and cost, and in fast-charging technology, electric cars will become cost-competitive with gas-powered cars this decade without government subsidies, leading to their rapid adoption. Electric cars are mechanically simpler and more reliable than gas-powered ones, which will hurt the car repair industry. Many gas stations will also go bankrupt or convert to fast charging stations.
  • Quality of life for people living and working in cities and near highways will improve as more drivers switch to quieter, emissionless electric vehicles. The noise reduction will be greatest in cities and suburbs where traffic moves slowly: https://cleantechnica.com/2016/06/05/will-electric-cars-make-traffic-quieter-yes-no/
  • Most new power equipment will be battery-powered, so machines like lawn mowers, leaf blowers, and chainsaws will be much quieter and less polluting than they are today. Batteries will be energy-dense enough to compete with gasoline in these use cases, and differences in overall equipment weight and running time will be insignificant. The notion of a neighbor shattering your sense of peace and quiet with loud yard work will get increasingly alien. 
  • A machine will pass the Turing Test by the end of this decade. The milestone will attract enormous amounts of attention and will lead to several retests, some of which the machine will fail, proving that it lacks the full range of human intelligence. It will lead to debate over the Turing Test’s validity as a measure of true intelligence (Ray Kurzweil actually talked about this phenomenon of “moving the goalposts” whenever we think about how smart computers are), and many AI experts will point out the existence of decades-long skepticism in the Turing Test in their community.
  • The best AIs circa 2029 won’t be able to understand and upgrade their own source codes. They will still be narrow AIs, albeit an order of magnitude better than the ones we have today.
  • Machines will become better than humans at the vast majority of computer, card, and board games. The only exceptions will be very obscure games or recently created games that no one has bothered to program an AI to play yet. But even for those games, there will be AIs with general intelligence and learning abilities that will be “good enough” to play as well as average humans by reading the instruction manuals and teaching themselves through simulated self-play.
  • The cost of getting your genome sequenced and expertly interpreted will drop below $1,000, and enough about the human genome will have been deciphered to make the cost worth the benefit for everyone. By the end of the decade, it will be common for newborns in rich countries to have their genomes sequenced.
  • Better technology will also let pregnant women noninvasively obtain their fetuses’ DNA, at affordable cost.
  • Cheap DNA tests that can measure a person’s innate IQ and core personality traits with high accuracy will become widely available. There is the potential for this to cause social problems. 
  • At-home medical testing kits and diagnostic devices like swallowable camera-pills will become vastly better and more common.
  • Space tourism will become routine thanks to privately owned spacecraft. 
  • Marijuana will be effectively decriminalized in the U.S. Either the federal government will overturn its marijuana prohibitions, or some patchwork of state and federal bans will remain but be so weakened and lightly enforced that there will be no real government barriers to obtaining and using marijuana. 
  • By the end of this decade, photos of almost every living person will be available online (mostly on social media). Apps will exist that can scan through trillions of photos to find your doppelgangers. 
  • In 2029, the youngest Baby Boomer and the oldest Gen Xer will turn 65. 
  • Drones will be used in an attempted or successful assassination of at least one major world leader (Note: Venezuela’s Nicholas Maduro wasn’t high profile enough).

2030s

  • VR and AR goggles will become refined technologies and probably merge into a single type of lightweight device. Like smartphones today, anyone who wants the glasses in 2030 will have them. Even poor people in Africa will be able to buy them. A set of the glasses will last a day on a single charge under normal use.  
  • Augmented reality contact lenses will enter mass production and become widely available, though they won’t be as good as AR glasses and they might need remotely linked, body-worn hardware to provide them with power and data. https://www.inverse.com/article/31034-augmented-reality-contact-lenses
  • The bulky VR goggles of the 2020s will transform into lightweight, portable V.R. glasses thanks to improved technology. The glasses will display lifelike footage. However, the best VR goggles will still need to be plugged into other devices, like routers or PCs.
  • Wall-sized, thin, 8K or even 16K TVs will become common in homes in rich countries, and the TVs will be able to display 3D picture without the use of glasses. A sort of virtual reality chamber could be created at moderate cost by installing those TVs on all the walls of a room to create a single, wraparound screen.
  • It will be common for celebrities of all kinds to make money by “hanging out” with paying customers in virtual reality. For some lower-tier celebrities, this will be their sole source of income. 
  • Functional CRT TVs and computer monitors will only exist in museums and in the hands of antique collectors. This will also be true for DLP TVs. 
  • The video game industry will be bigger than ever and considered high art.
  • It will be standard practice for AIs to be doing hyperrealistic video game renderings, and for NPCs to behave very intelligently thanks to better AI. 
  • Books and computer tablets will merge into a single type of device that could be thought of as a “digital book.” It will be a book with several hundred pages made of thin, flexible digital displays (perhaps using ultra-energy efficient e-ink) instead of paper. At the tap of a button, the text on all of the pages will instantly change to display whichever book the user wanted to read at that moment. They could also be used as notebooks in which the user could hand write or draw things with a stylus, which would be saved as image or text files. The devices will fuse the tactile appeal of old-fashioned books with the content flexibility of tablet computers.
  • Loose-leaf sheets of “digital paper” will also exist thanks to the same technology.
  • Commercially available, head-worn, brain-computer-interface devices (BCIs) linked to augmented reality eyewear will gift humans with crude forms of telepathy and telekinesis. For example, a person wearing the devices could compose a short sentence merely by thinking about it, see the text projected across his augmented field of view, use his thoughts to make any needed edits, and then transmit the sentence to another person or machine, merely by thinking a “Send” command. The human recipient of the message with the same BCI/eyewear setup would see the text projected across his field of view and could compose a response through the same process the first person used. BCIs will also let humans send commands to a machines, like printers. For almost all use cases, this type of communication will be less efficient than traditional alternatives, like manually typing a text message or clicking the “Print” button at the top of a word processing application, but it will be an important proof of concept demonstration that will point to what is to come later in the century.
  • Loneliness, social isolation, and other problems caused by overuse of technology and the atomized structure of modern life will be, ironically, cured to a large extent by technology. Chatbots that can hold friendly (and even funny and amusing) conversations with humans for extended periods, diagnose and treat mental illnesses as well as human therapists, and customize themselves to meet the needs of humans will become ubiquitous. The AIs will become adept at analyzing human personalities and matching lonely people with friends and lovers, at matching them with social gatherings (including some created by machines), and at recommending daily activities that will satisfy them, hour-by-hour. Machines will come to understand that constant technology use is antithetical to human nature, so in order to promote human wellness, they find ways to impel humans to get out of their houses, interact with other humans, and be in nature. Autonomous taxis will also be widespread and will have low fares, making it easier for people who are isolated due to low income or poor health (such as many elderly people) to go out.
  • Chatbots will steadily improve their “humanness” over the decade. The instances when AIs say or do something nonsensical will get less and less frequent. Dumber people, children, and people with some types of mental illness will be the first ones to start insisting their AIs are intelligent like humans. Later, average people will start claiming the same. By the end of the decade, a personal assistant AI like “Samantha” from the movie Her will be commercially available. AI personal assistants will have convincing, simulated personalities that seem to have the same depth as humans. Users will be able to pick from among personality profiles or to build their own.  
  • Chatbots will be able to have intelligent conversations with humans about politics and culture, to identify factually wrong beliefs, biases, and cognitive blind spots in individuals, and to effectively challenge them through verbal discussion and debate. The potential will exist for technology to significantly enlighten the human population and to reduce sociopolitical polarization. However, it’s unclear how many people will choose to use this technology. 
  • Turing-Test-capable chatbots will also supercharge the problem of online harassment, character assassination, and deliberate disinformation by spamming the internet with negative reviews, bullying messages, emails to bosses, and humiliating “deepfake” photos and videos of targeted people. Today’s “troll farms” where humans sit at computer terminals following instructions to write bad reviews for specific people or businesses will be replaced by AI trolls that can pump out orders of magnitude more content per day. And just as people today can “buy likes” for their social media accounts or business webpages, people in the future will be able, at low cost, to buy harassment campaigns against other people and organizations they dislike. Discerning between machine-generated and human-generated internet content will be harder and more important than ever.
  • House robots will start becoming common in rich countries. They will be slower at doing household tasks than humans, but will still save people hours of labor per week. They may or may not be humanoid. For the sake of safety and minimizing annoyance, most robots will do their work when humans aren’t around. As in, you would come home from work every day and find the floors vacuumed, the lawn mowed, and your laundered clothes in your dresser, with nary a robot in sight since it will have gone back into its closet to recharge. You would never hear the commotion of a clothes washing machine, a vacuum cleaner or a lawnmower. All the work would get done when you were away, as if by magic.
  • People will start having genuine personal relationships with AIs and robots. For example, people will resist upgrading to new personal assistant AIs because they will have emotional attachments to their old ones. The destruction of a helper robot or AI might be as emotionally traumatic to some people as the death of a human relative.
  • Farm robots that are better than humans at fine motor tasks like picking strawberries humans will start becoming widespread.  
  • Self-driving cars will become cheap enough and practical enough for average income people to buy, and their driving behavior will become as efficient as an average human. Over the course of this decade, there will be rapid adoption of self-driving cars in rich countries. Freed from driving, people will switch to doing things like watching movies/TV and eating. Car interiors will change accordingly. Road fatalities, and the concomitant demands for traffic police, paramedics, E.R. doctors, car mechanics, and lawyers will sharply decrease. The car insurance industry will shrivel, forcing consolidation. (Humans in those occupations will also face increasing levels of direct job competition from machines over the course of the decade.)
  • Private owners of autonomous cars will start renting them out while not in use as taxis and package delivery vehicles. Your personal, autonomous car will drive you to work, then spend eight hours making money for you doing side jobs, and will be waiting for you outside your building at the end of the day.
  • The “big box” business model will start taking over the transportation and car repair industry thanks to the rise of electric, self-driving vehicles and autonomous taxis in place of personal car ownership. The multitudes of small, scattered car repair shops will be replaced by large, centralized car repair facilities that themselves resemble factory assembly lines. Self-driving vehicles will drive to them to have their problems diagnosed and fixed, sparing their human owners from having to waste their time sitting in waiting rooms.
  • The same kinds of facilities will make inroads into the junk yard industry, as they would have all the right tooling to cheaply and rapidly disassemble old vehicles, test the parts for functionality, and shunt them to disposal or individual resale. (The days of hunting through junkyards by yourself for a car part you need will eventually end–it will all be on eBay. )
  • Car ownership won’t die out because it will still be a status symbol, and having a car ready in your driveway will always be more convenient than having to wait even just two minutes for an Uber cab to arrive at the curb. People are lazy.
  • The ad hoc car rental model exemplified by autonomous Uber cabs and private people renting out their autonomous cars when not in use faces a challenge since daily demand for cars peaks during morning rush hour and afternoon rush hour. In other words, everyone needs a car at the same time each day, so the ratio of cars : people can’t deviate much from, say, 1:2. Of course, if more people telecommuted (almost certain in the future thanks to better VR, faster broadband, and tech-savvy Millennials reaching middle age and taking over the workplace), and if flexible schedules became more widespread (also likely, but within certain limits since most offices can’t function efficiently unless they have “all hands on deck” for at least a few hours each day), the ratio could go even lower. However, there’s still a bottom limit to how few cars a country will need to provide adequate daily transportation for its people.
  • Private delivery services will get cheaper and faster thanks to autonomous vehicles.
  • Automation will start having a major impact on the global economy. Machines will compensate for the shrinkage of the working-age human population in the developed world. Countries with “graying” populations like Japan and Germany will experience a new wave of economic growth. Demand for immigrant laborers will decrease across the world because of machines.
  • There will be a worldwide increase in the structural unemployment rate thanks to better and cheaper narrow AIs and robots. A plausible scenario would be for the U.S. unemployment rate to be 10%–which was last the case at the nadir of the Great Recession–but for every other economic indicator to be strong. The clear message would be that human labor is becoming decoupled from the economy.
  • Combining all the best AI and robotics technologies, it will be possible to create general-purpose androids that could function better in the real world (e.g. – perform in the workplace, learn new things, interact with humans, navigate public spaces, manage personal affairs) than the bottom 10% of humans (e.g. – elderly people, the disabled, criminals, the mentally ill, people with poor language abilities or low IQs), and in some narrow domains, the androids will be superhuman (e.g. – physical strength, memory, math abilities). Note that businesses will still find it better to employ task-specific, non-human-looking robots instead of general purpose androids.
  • By the end of this decade, only poor people, lazy people, and conspiracy theorists (like anti-vaxxers) won’t have their genomes sequenced. It will be trivially cheap, and in fact free for many people (some socialized health care systems will fully subsidize it), and enough will be known about the human genome to make it worthwhile to have the information.
  • Computers will be able to accurately deduce a human’s outward appearance based on only a DNA sample. This will aid police detectives, and will have other interesting uses, such as allowing parents to see what their unborn children will look like as adults, or allowing anyone to see what they’d look like if they were of the opposite sex (one sex chromosome replaced). 
  • Trivially cheap gene sequencing and vastly improved knowledge of the human genome will give rise to a “human genome black market,” in which people secretly obtain DNA samples from others, sequence them, and use the data for their own ends. For example, a politician could be blackmailed by an enemy who threatened to publish a list of his genetic defects or the identities of his illegitimate children. Stalkers (of celebrities and ordinary people) would also be interested in obtaining the genetic information of the people they were obsessed with. It is practically impossible to prevent the release of one’s DNA since every discarded cup, bottle, or utensil has a sample. 
  • Markets will become brutally competitive and efficient thanks to AIs. Companies will sharply grasp consumer demand through real-time surveillance, and consumers will be alerted to bargains by their personal AIs and devices (e.g. – your AR glasses will visually highlight good deals as you walk through the aisles of a store). Your personal assistant AIs and robots will look out for your self-interest by countering the efforts of other AIs to sway your spending habits in ways that benefit companies and not you.
  • “Digital immortality” will become possible for average people. Personal assistant AIs, robot servants, and other monitoring devices will be able, through observation alone, to create highly accurate personality profiles of individual humans, and to anticipate their behavior with high fidelity. Voices, mannerisms and other biometrics will be digitally reproducible without any hint of error. Digital simulacra of individual humans will be further refined by having them take voluntary personality tests, and by uploading their genomes, brain scans and other body scans. Even if all of the genetic and biological data couldn’t be made sense of at the moment it was uploaded to an individual’s digital profile, there will be value in saving it since it might be decipherable in the future. (Note that “digital immortality” is not the same as “mind uploading.”)
  • Life expectancy will have increased by a few years thanks to pills and therapies that slightly extend human lifespan. Like, you take a $20 pill each day starting at age 20 and you end up dying at age 87 instead of age 84.
  • Global oil consumption will peak as people continue switching to other power sources.
  • Earliest possible date for the first manned Mars mission.
  • Movie subtitles and the very notion of there being “foreign language films” will become obsolete. Computers will be able to perfectly translate any human language into another, to create perfect digital imitations of any human voice, and to automatically apply CGI so that the mouth movements of people in video footage matches the translated words they’re speaking. The machines will also be able to reproduce detailed aspects of an actor’s speech, such as cadence, rhythm, tone and timbre, emotion, and accent, and to convey them accurately in another language.
  • Computers will also be able to automatically enhance and upscale old films by accurately colorizing them, removing defects like scratches, and sharpening or focusing footage (one technique will involve interpolating high-res still photos of long-dead actors onto the faces of those same actors in low-res moving footage). Computer enhancement will be so good that we’ll be able to watch films from the early 20th century with near-perfect image and audio clarity.
  • CGI will get so refined than moviegoers with 20/20 vision won’t be able to see the difference between footage of unaltered human actors and footage of 100% CGI actors.
  • Lifelike CGI and “performance capture” will enable “digital resurrections” of dead actors. Computers will be able to scan through every scrap of footage with, say, John Wayne in it, and to produce a perfect CGI simulacrum of him that even speaks with his natural voice, and it will be seamlessly inserted into future movies. Elderly actors might also license movie studios to create and use digital simulacra of their younger selves in new movies. The results will be very fascinating, but might also worsen Hollywood’s problem with making formulaic content.
  • China’s military will get strong enough to defeat U.S. forces in the western Pacific. This means that, in a conventional war for control of the Spratly Islands and/or Taiwan, China would have >50% odds of winning. This shift in the local balance of power does not mean China will start a conflict. 
  • The quality and sophistication of China’s best military technology will surpass Russia’s best technology in all or almost all categories. However, it will still lag the U.S. 

2040s

  • The world and peoples’ outlooks and priorities will be very different than they were in 2019. Cheap renewable energy will have become widespread and totally negated any worries about an “energy crisis” ever happening, except in exotic, hypothetical scenarios about the distant future. There will be little need for immigration thanks to machine labor and cross-border telecommuting (VR, telepresence, and remote-controlled robots will be so advanced that even blue-collar jobs involving manual labor will be outsourced to workers living across borders). Moreover, there will be a strong sense in most Western countries that they’re already “diverse enough,” and that there are no further cultural benefits to letting in more foreigners since large communities of most foreign ethnic groups will already exist within their borders. There will be more need than ever for strong social safety nets and entitlement programs thanks to technological unemployment. AI will be a central political and social issue. It won’t be the borderline sci-fi, fringe issue it was in 2019.
  • Automation, mass unemployment, wealth inequalities between the owners of capital and everyone else, and differential access to expensive human augmentation technologies (like genetic engineering) will produce overwhelming political pressure for some kind of wealth redistribution and social safety net expansion. Countries that have diligently made small, additive reforms as necessary over the preceding decades will be untroubled. However, countries that failed to adapt their political and economic systems will face upheaval.
  • 2045 will pass without the Technological Singularity happening. Ray Kurzweil will either celebrate his 97th birthday in a wheelchair, or as a popsicle frozen at the Alcor Foundation.
  • Supercomputers that match or surpass upper-level estimates of the human brain’s computational capabilities will cost a few hundred thousand to a few million dollars apiece, meaning tech companies and universities will be able to afford large numbers of them for AI R&D projects, accelerating progress in the field. Hardware will no longer be the limiting factor to building AGI. If it hasn’t been built yet, it will be due to failure to figure out how to arrange the hardware in the right way to support intelligent thought, and/or to a failure to develop the necessary software. 
  • With robots running the economy, it will be common for businesses to operate 24/7: restaurants will never close, online orders made at 3:00 am will be packed in boxes by 3:10 am, and autonomous delivery trucks will only stop to refuel, exchange cargo, or get preventative maintenance.
  • Advanced energy technology, robot servants, 3D printers, telepresence, and other technologies will allow people to live largely “off-grid” if they choose, while still enjoying a level of comfort that 2019 people would envy.
  • Recycling will become much more efficient and practical thanks to house robots properly cleaning, sorting, and crushing/compacting waste before disposing of it. Automated sorting machines at recycling centers will also be much better than they are today. Today, recycling programs are hobbled because even well-meaning humans struggle to remember which of their trash items are recyclable and which aren’t since the acceptable items vary from one municipality to the next, and as a result, recycling centers get large amounts of unusable material, which they must filter out at great cost. House robots would remember it perfectly.
  • Thanks to this diligence, house robots will also increase backyard composting, easing the burden on municipal trash services. 
  • It will be common for cities, towns and states to heavily restrict or ban human-driven vehicles within their boundaries. A sea change in thinking will happen as autonomous cars become accepted as “the norm,” and human-driven cars start being thought of as unusual and dangerous.
  • Over 90% of new car sales in developed countries will be for electric vehicles. Just as the invention of the automobile transformed horses into status goods used for leisure, the rise of electric vehicles will transform internal combustion vehicles into a niche market for richer people. 
  • A global “family tree” showing how all humans are related will be built using written genealogical records and genomic data from the billions of people who have had their DNA sequenced. It will become impossible to hide illegitimate children, and it will also become possible for people to find “genetic doppelgangers”–other people they have no familial relationship to, but with whom, by some coincidence, they share a very large number of genes. 
  • Improved knowledge of human genetics and its relevance to personality traits and interests will strengthen AI’s ability to match humans with friends, lovers, and careers. Rising technological unemployment will create a need for machines to match human workers with the remaining jobs in as efficient a manner as possible.
  • People with distinctive personalities (particularly vibrant, funny, or sexy) will routinely sell “digital copies” of themselves for other people to download and use as AI personal assistants. This will be analogous to today’s ability to select different voices for personal GPS devices. Additionally, users will be able to tweak “base versions” of downloaded personalities to suit their unique preferences. 
  • The digital personalities of fictitious people, like movie and cartoon characters, and of long-dead people, will also be downloadable. 
  • Realistic robot sex bots that can move and talk will exist. They won’t perfectly mimic humans, but will be “good enough” for most users. Using them will be considered weird and “for losers” at first, but in coming decades it will go mainstream, following the same pattern as Internet dating. [If we think of sex as a type of task, and if we agree that machines will someday be able to do all tasks better than humans, then it follows that robots will be better than humans at sex.]  
  • Augmented reality contact lenses will give people superhuman vision.

2050s

  • This is the earliest possible time that AGI/SAI will be invented. It will not be able to instantly change everything in the world or to initiate a Singularity, but it will rapidly grow in intelligence, wealth, and power. It will probably be preceded by successful computer simulations of the brains of progressively more complex model organisms, such as flatworms, fruit flies, and lab rats.
  • Humans will be heavily dependent upon their machines for almost everything (e.g. – friendship, planning the day, random questions to be answered, career advice, legal counseling, medical checkups, driving cars), and the dependency will be so ingrained that humans will reflexively assume that “The Machines are always right.” Consciously and unconsciously, people will yield more and more of their decision-making and opinion-forming to machines, and find that they and the world writ large are better off for it. This will be akin to having an angel on your shoulder watching your surroundings and watching you, and giving you constructive advice all the time. 
  • In the developed world, less than 50% of people between age 22 and 65 will have gainful full-time jobs. However, if unprofitable full-time jobs that only persist thanks to government subsidies (such as someone running a small coffee shop and paying the bills with their monthly UBI check) and full-time volunteer “jobs” (such as picking up trash in the neighborhood) are counted, most people in that age cohort will be “doing stuff” on a full-time basis.  
  • The doomsaying about Global Warming will start to quiet down as the world’s transition to clean energy hits full stride and predictions about catastrophes from people like Al Gore fail to pan out by their deadlines. Sadly, people will just switch to worrying about and arguing about some new set of doomsday prophecies about something else.
  • By almost all measures, standards of living will be better in 2050 than today. People will commonly have all types of wonderful consumer devices and appliances that we can’t even fathom. However, some narrow aspects of daily life are likely to worsen, such as overcrowding and further erosion of the human character. Just as people today have short memories and take too many things for granted, so shall people in the 2050s fail to appreciate how much the standard of living has risen since today, and they will ignore all the steady triumphs humanity has made over its problems, and by default, people will still believe the world is constantly on the verge of collapsing and that things are always getting worse.
  • Cheap desalination will provide humanity with unlimited amounts of drinking water and end the prospect of “water wars.” 
  • Mass surveillance and ubiquitous technology will have minimized violent crime and property crime in developed countries: It will be almost impossible to commit such crimes without a surveillance camera or some other type of sensor detecting the act, or without some device recording the criminal’s presence in the area at the time of the act. House robots will contribute by effectively standing guard over your property at night while you sleep. 
  • It will be common for people to have health monitoring devices on and inside of their bodies that continuously track things like their heart rate, blood pressure, respiration rate, and gene expression. If a person has a health emergency or appears likely to have one, his or her devices will send out a distress signal alerting EMS and nearby random citizens. If you walked up to such a person while wearing AR glasses, you would see their vital statistics and would receive instructions on how to assist them (i.e. – How to do CPR). Robots will also be able to render medical aid. 
  • Cities and their suburbs across the world will have experienced massive growth since 2019. Telepresence, relatively easy off-grid living, and technological unemployment will not, on balance, have driven more people out of metro areas than have migrated into them. Farming areas full of flat, boring land will have been depopulated, and many farms will be 100% automated. The people who choose to leave the metro areas for the “wilderness” will concentrate in rural areas (including national parks) where the climate is good, the natural scenery is nice, and there are opportunities for outdoor recreation. Real estate prices will, in inflation-adjusted terms, be much higher in most metro areas and places with natural beauty than they were in 2020 because the “supply” of those prime locations is almost fixed, whereas the demand for them is elastic and will rise thanks to population growth, rising incomes, and the aforementioned technology advancements.
  • Therapeutic cloning and stem cell therapies will become useful and will effectively extend human lifespan. For example, a 70-year-old with a failing heart will be able to have a new one grown in a lab using his own DNA, and then implanted into his chest to replace the failing original organ. The new heart will be equivalent to what he had when at age 18 years, so it will last another 52 years before it too fails. In a sense, this will represent age reversal to one part of his body. In a sense, this will represent age reversal to one part of his body.
  • As a result of the above technologies, it will be much rarer for people in rich countries to die waiting for organ transplants than it is now, in 2022.
  • The first healthy clone of an adult human will be born.
  • Many factories, farms, and supply chains will be 100% automated, and it will be common for goods to not be touched by a human being’s hands until they reach their buyers. Robots will deliver Amazon packages to your doorstep and even carry them into your house. Items ordered off the internet will appear inside your house a few hours later, as if by magic. 
  • Smaller versions of the robots used on automated farms will be available at low cost to average people, letting them effortlessly create backyard gardens. This will boost global food production and let people have greater control over where their food comes from and what it contains. 
  • The last of America’s Cold War-era weapon platforms (e.g. – the B-52 bomber, F-15 fighter, M1 Abrams tank, Nimitz aircraft carrier) will finally be retired from service. There will be instances where four generations of people from the same military family served on the same type of plane or ship. 
  • Cheap guided bullets, which can make midair course changes and be fired out of conventional man-portable rifles, will become common in advanced armies. 
  • Personal “cloaking devices” made of clothes studded with pinhole cameras and thin, flexible sheets of LEDs, colored e-ink, or some metamaterial with similar abilities will be commercially available. The cameras will monitor the appearance of the person’s surroundings and tell the display pixels to change their colors to match.
  • The “cloaking” outfits will also have benign applications related to fashion and everyday utility. People wearing them could use them to display morphing patterns and colors of their choice. It would even be possible to become a “walking TV.” The pixels could also be made to glow bright white, allowing the wearer to turn any part of his body into a flashlight. Ski masks made of the same material would let wearers change their facial features, fooling most face recognition cameras and certainly fooling the unaided eyes of humans, at least at a distance.
  • Powered exoskeletons will become practical for a wide range of applications, mainly due to improvements in batteries. For example, a disabled person could use a lightweight exoskeleton with a battery the size of a purse to walk around for a whole day on a single charge, and a soldier in a heavy-duty exoskeleton with a large backpack battery could do a day of marching on a single charge. (Note: Even though it will be technologically possible to equip infantrymen with combat exoskeletons, armies might reject the idea due to other impracticalities.)
  • There will be no technological or financial barrier to building powered combat exoskeletons that have cloaking devices. 
  • The richest person alive will achieve a $1 trillion net worth.
  • It will be technologically and financially feasible for small aircraft to produce zero net carbon emissions. The aircraft might use conventional engines powered by carbon-neutral synthetic fossil fuels that cost no more than normal fossil fuels, or they might have electric engines and very energy-dense batteries or fuel cells.
  • Cheap guided bullets, capable of midair course changes to hit targets and of being fired out of conventional rifles, will become common in advanced armies. Practical, affordable rifles capable of limited self-aiming will also exist (similar to the “Smartgun” from the movie Aliens). Thanks to these technologies, an ordinary rifleman of the 2050s will be like the snipers of today.

2060s

  • Machines will be better at satisfyingly matching humans with fields of study, jobs, friends, romantic partners, hobbies, and daily activities than most humans can do for themselves. Machines themselves will make better friends, confidants, advisers, and even lovers than humans. Additionally, machines will be smarter and more skilled at humans in most areas of knowledge and types of work. A cultural sea change will happen, in which most humans come to trust, rely upon, defend, and love machines.
  • House robots and human-sized worker robots will be as strong, agile, and dexterous as most humans, and their batteries will be energy-dense enough to power them for most of the day. A typical American family might have multiple robot servants that physically follow around the humans each day to help with tasks. The family members will also be continuously monitored and “followed” by A.I.s embedded in their portable personal computing devices and possibly in their bodies. 
  • Cheap home delivery of groceries, robot chefs, and a vast trove of free online recipes will enable people in average households to eat restaurant-quality meals at home every day, at low cost. Predictive algorithms that can appropriately choose new meals for humans based on their known taste preferences and other factors will determine the menu, and many people will face a culinary “satisfaction paradox.”
  • Average people will have access to high-quality meals that only rich people can have today at fancy restaurants.
  • Machines will understand humans individually and at the species level better than humans understand themselves. They will have highly accurate personality models of most humans along with a comprehensive grasp of human sociology, human decision-making, human psychology, human cognitive biases, and human nature, and will pool the information to accurately predict human behavior. A nascent version of a 1:1 computer simulation of the Earth–with the human population modeled in great detail–will be created.
  • Machines will be better teachers than most trained humans. The former will have much sharper grasps of their pupils’ individual strengths, weaknesses, interests, and learning styles, and will be able to create and grade tests in a much fairer and less biased manner than humans. Every person will have his own tutor. 
  • There will be a small, permanent human presence on the Moon.
  • If a manned Mars mission hasn’t happened yet, then there will be intense pressure to do so by the centennial of the first Moon landing (1969).
  • The worldwide number of supercentenarians–people who are at least 110 years old–will be sharply higher than it was in 2019: Their population size could be 10 times bigger or more. 
  • Advances in a variety of technologies will make it possible to cryonically freeze humans in a manner that doesn’t pulverize their tissue. However, the technology needed to safely thaw them out won’t be invented for decades. 
  • China will effectively close the technological, military, and standard of living gaps with other developed countries. Aside from the unpleasantness of being a more crowded place, life in China won’t be worse overall than life in Japan or the average European country. Importantly, China’s pollution levels will be much lower than they are today thanks to a variety of factors.
  • Small drones (mostly aerial) will have revolutionized warfare, terrorism, assassinations, and crime and will be mature technologies. An average person will be able to get a drone of some kind that can follow his orders to find and kill other people or to destroy things.
  • Countermeasures against those small drones will also have evolved, and might include defensive drones and mass surveillance networks to detect drone attacks early on. The networks would warn people via their body-worn devices of incoming drone attacks or of sightings of potentially hostile drones. The body-worn devices, such as smartphones and AR glasses, might even have their own abilities to automatically detect drones by sight and sound and to alert their wearers.

2070s

  • 100 years after the U.S. “declared war” on cancer, there still will not be a “cure” for most types of cancer, but vaccination, early detection, treatment, and management of cancer will be vastly better, and in countries with modern healthcare systems, most cancer diagnoses will not reduce a person’s life expectancy. Consider that diabetes and AIDS were once considered “death sentences” that would invariably kill people within a few years of diagnosis, until medicines were developed that transformed them into treatable, chronic health conditions. 
  • Hospital-acquired infections will be far less of a problem than they are in 2020 thanks to better sterilization practices, mostly made possible by robots.
  • It will be technologically and financially feasible for large commercial aircraft to produce zero net carbon emissions. The aircraft might use conventional engines powered by synthetic fossil fuels, or they might have electric engines and very energy-dense batteries or fuel cells. 
  • Digital or robotic companions that seem (or actually are) intelligent, funny, and loving will be easier for humans to associate with than other humans.
  • Technology will enable the creation of absolute surveillance states, where all human behavior is either constantly monitored or is inferred with high accuracy based on available information. Even a person’s innermost thoughts will be knowable thanks to technologies that monitor him or her for the slightest things like microexpressions, twitches, changes in voice tone, and eye gazes. When combined with other data regarding how the person spends their time and money, it will be possible to read their minds. The Thought Police will be a reality in some countries.  
  • Thanks to advanced lab synthesis of foods, new spices, hybrid fruits and vegetables, and meats with entirely new taste profiles will be brought into existence. Swaths of the “landscape of all possible flavors” that are currently unexplored will be.
  • Many heavily automated farms (including indoor farms and gardens on suburban plots of land) will produce food that is noticeably tastier and measurably more nutritious that most of today’s food because the advanced farms won’t need to use pesticides or to favor crop varieties that are hardy enough to endure transport over long supply chains. At low cost and for little effort, communities and individuals with small amounts of land will be able to meet their own food needs locally. People who value “natural” lifestyles might, ironically, find it most beneficial to rely on robots to make their food for them.

2100

  • Humans probably won’t be the dominant intelligent life forms on Earth.
  • Latest possible time that AGI/SAI will be invented. By this point, computer hardware will so powerful that we could do 1:1 digital simulations of human brains. If our AI still falls far short of human-like general intelligence and creativity, then it might be that only organic substrates have the necessary properties to support them.
  • The worst case scenario is that AGI/Strong AI will have not been invented yet, but thousands of different types of highly efficient, task-specific Narrow AIs will have (often coupled to robot bodies), and they will fill almost every labor niche better than human workers ever could (“Death by a Thousand Cuts” job automation scenario). Humans grow up in a world where no one has to work, and the notion of drudge work, suffering through a daily commute, and involuntarily waking up at 6:00 am five days a week is unfathomable. Every human will have machines that constantly monitor them or follow them around, and meet practically all their needs.
  • Telepresence technology will also be very advanced, allowing humans to do nearly any task remotely, from any other place in the world, in safety and comfort. This will include cognitive tasks and hands-on tasks. If any humans still have jobs, they’ll be able to work from anywhere.
  • Sophisticated narrow AI will be integrated into the telepresence technology, providing human workers with real-time assistance with tasks. An illustrative scenario would have a human in Nigeria using a VR rig to remotely control a robot that is fixing an air conditioner in England. Software programs monitoring the live video feed would recognize all of the objects in the robot’s field of view and would also understand what the human worker was trying to accomplish, and the programs would help him by visually highlighting tools or air conditioner components, or by giving him verbal advice on what to do. 
  • The use of robotic surrogate bodies for remote work will also erase any employment gaps caused by physical strength and endurance differences between the sexes and between the elderly and the young. Small men, old people, and women of average stature will be just as good at performing hard manual labor as big men. The easing of physical strain associated with work will also allow people to work past today’s retirement age. However, most serious physical work will be best left to autonomous machines.
  • The world could in many ways resemble Ray Kurzweil’s predicted Post-Singularity world. However, the improvements and changes will have accrued thanks to decades of AGI/Strong AI steady effort. Everything will not instantly change on DD/MM/2045 as Kurzweil suggests it will.
  • Hundreds of millions, and possibly billions, of “digitally immortal avatars” of dead humans will exist, and you will be able to interact with them through a variety of means (in FIVR, through devices like earpieces and TV screens, in the real world if the avatar takes over an android body resembling the human it was based on). 
  • A weak sort of immortality will be available thanks to self-cloning, immortal digital avatars, and perhaps mind uploading. You could clone yourself and instruct your digital avatar–which would be a machine programmed with your personality and memories–to raise the clone and ensure it developed to resemble you. Your digital avatar might have an android body or could exist in a disembodied state. 
  • It will be possible to make clones of humans using only their digital format genomic data. In other words, if you had a .txt file containing a person’s full genetic code, you could use that by itself to make a living, breathing clone. Having samples of their cells would not be necessary. 
  • The “DNA black market” that arose in the 2030s will pose an even bigger threat since it will be now possible to use DNA samples alone or their corresponding .txt files to clone a person or to produce a sperm or egg cell and, in turn, a child. Potential abuses include random people cloning or having the children of celebrities they are obsessed with, or cloning billionaires in the hopes of milking the clones for money. Important people who might be targets of such thefts will go to pains to prevent their DNA from being known. Since dead people have no rights, third parties might be able to get away with cloning or making gametes of the deceased.
  • Life expectancy escape velocity and perhaps medical immortality will be achieved. It will come not from magical, all-purpose nanomachines that fix all your body’s cells and DNA, but from a combination of technologies, including therapeutic cloning of human organs, cybernetic replacements for organs and limbs, and stem cell therapies that regenerate ageing tissues and organs inside the patient’s body. The treatments will be affordable in large part thanks to robot doctors and surgeons who work almost for free, and to medical patents expiring.
  • All other aspects of medicine and healthcare will have radically advanced. There will be vaccines and cures for almost all contagious diseases. We will be masters of human genetic engineering and know exactly how to produce people that today represent the top 1% of the human race (holistically combining IQ, genetic health, physical attractiveness, and likable/prosocial personality traits). However, the value of even a genius-IQ human will be questionable since intelligent machines will be so much smarter.
  • Augmentative cybernetics (including direct brain-to-computer links) will exist and be in common use.
  • While the traditional, “pure” races of humans will all still exist, notions of “race” and racial identity will be scrambled by the large numbers of mixed-race people who will be alive, and by widespread genetic engineering that will give people combinations of physical traits that were almost unachievable through normal human breeding. Examples might include black people with naturally blue eyes, or East Asians with naturally blonde hair. (Voluntary genetic engineering will also ensure that redheads don’t ever die out.) Some people will even have totally new genes, either synthesized in labs or borrowed from animals, that give them physical traits not found in any preexisting human race, like red eyes or purple hair.
  • Full-immersion virtual reality (FIVR) will exist wherein AI game masters constantly tailor environments, NPCs and events to suit each player’s needs and to keep them entertained. Every human will have his own virtual game universe where he’s #1. With no jobs in the real world to occupy them, it’s quite possible that a large fraction of the human race will willingly choose to live in FIVR. (Related to the satisfaction paradox) Elements of these virtual environments could be pornographic and sexual, allowing people to gratify any type of sexual fetish or urge with computer-generated scenarios and partners. 
  • More generally, AIs and humans whose creativity is turbocharged by machines will create enjoyable, consumable content (e.g. – films, TV shows, songs, artwork, jokes, new types of meals) faster than non-augmented humans can consume it. As a simple example of what this will be like, assume you have 15 hours of free time per day, that you love spending it listening to music, and each day, your favorite bands produce 16 hours worth of new songs that you really like.
  • The vast majority of unaugmented human beings will no longer be assets that can invent things and do useful work: they will be liabilities that do (almost) everything worse than intelligent machines and augmented humans. Ergo, the size of a nation’s human population will subtract from its economic and military power, and radical shifts in geopolitics are possible. Geographically large but sparsely populated countries like Russia, Australia and Canada might become very strong.
  • The transition to green energy sources will be complete, and humans will no longer be net emitters of greenhouse gases. The means will exist to start reducing global temperatures to restore the Earth to its pre-industrial state, but people will resist because they will have gotten used to the warmer climate. People living in Canada and Russia won’t want their countries to get cold again.
  • Synthetic meat will taste no different from animal meat, and will be at least as cheap to make. The raising and/or killing of animals for food will be be illegal in many countries, and trends will clearly show the practice heading for worldwide ban. 
  • Meats that are expensive and/or rare today, like Kobe beef steaks, snakes, bats, or even human flesh, will be cheap and widely available thanks to meat synthesis technology. 
  • Cheap, synthetic chicken eggs will also exist and will taste no different from natural eggs. 
  • The means to radical alter human bodies, alter memories, and alter brain structures will be available. The fundamental bases of human existence and human social dynamics will change unpredictably once differences in appearance/attractiveness, intelligence, and personality traits can be eliminated at will. Individuals won’t be defined by fixed attributes anymore. 
  • Brain implants will make “telepathy” possible between humans, machines and animals. Computers, sensors and displays will be embedded everywhere in the built environment and in nature, allowing humans with brain implants to interface with and control things around them through thought alone. 
  • Brain implants and brain surgeries will also be used to enhance IQ, change personality traits, and strengthen many types of skills. 
  • Using brain-computer interfaces, people will be able to make sophisticated songs and pieces of artwork with their thoughts alone. 
  • Technologically augmented humans and androids will have many abilities and qualities that ancient people considered “Godlike,” such as medical immortality, the ability to control objects by thought, telepathy, perfect memories, and superhuman senses.
  • Flying cars designed to carry humans could be common, but they will be flown by machines, not humans. Ground vehicles will retain many important advantages (fuel efficiency, cargo capacity, safety, noise level, and more) and won’t become obsolete. Instead of flying cars, it’s more likely that there will be millions of small, autonomous helicopters and VTOL aircraft that will cheaply ferry people through dense, national networks of helipads and airstrips. Autonomous land vehicles would take take passengers to and from the landing sites. (https://www.militantfuturist.com/why-flying-cars-never-took-off-and-probably-never-will/
  • The notion of vehicles (e.g. – cars, planes, and boats) polluting the air will be an alien concept. 
  • Advanced nanomachines could exist.
  • Vastly improved materials and routine use of very advanced computer design simulations (including simulations done in quantum computers) will mean that manufactured objects of all types will be optimally engineered in every respect, and might seem to have “magical” properties. For example, a car will be made of hundreds of different types of alloys, plastics, and glass, each optimized for a different part of the vehicle, and car recalls will never happen since the vehicles will undergo vast amounts of simulated testing in every conceivable driving condition in 1:1 virtual simulations of the real world. 
  • Design optimization and the rise of AGI consumption will virtually eliminate planned obsolescence. Products that were deliberately engineered to fail after needlessly short periods, and “new” product lines that were no better than what they replaced, but had non-interchangeable part sizes would be exposed for what they were, and AGI consumers would refuse to buy them. Production will become much more efficient and far fewer things will be thrown out. 
  • Relatively cheap interplanetary travel (probably just to Mars and to space stations and moons that are about as far as Mars) will exist.
  • Androids that are outwardly indistinguishable from humans will exist, and humans will hold no advantages over them (e.g. – physical dexterity, fine motor control, appropriateness of facial expressions, capacity for creative thought). Some androids will also be indistinguishable to the touch, meaning they will seem to be made of supple flesh and will be the same temperature as human bodies. However, their body parts will not be organic.
  • Sex robots will be indistinguishable from humans.
  • Android assassins like the T-800s from the Terminator films will exist. They will look identical to humans, will be able to blend into human populations, track down targets, and kill or abduct them. As in the films, these androids will be stronger, more durable, and more skilled with weapons than we are.
  • Robots that are outwardly identical to sci-fi and fantasy characters and extinct animals, like grey aliens, elves, fairies, giant house cats, and dinosaurs, will exist and will occasionally be seen in public. Some weird person will want their robot butler to look like bigfoot, and at least one hobbyist will build a life-sized robotic dragon that can fly and spit fire.
    https://www.mentalfloss.com/article/503967/could-game-throness-dragons-really-fly-we-asked-some-experts 
  • Humans interested in extreme body modifications will be able to surgically alter themselves to look like many of those creatures.
  • Machines that are outwardly indistinguishable from animals will also exist, and they will have surveillance and military applications. 
  • Drones, miniaturized smart weapons, and AIs will dominate warfare, from the top level of national strategy down to the simplest act of combat. The world’s strongest military could, with conventional weapons alone, destroy most of the world’s human population in a short period of time. 
  • The construction and daily operation of prisons will have been fully automated, lowering the monetary costs of incarceration. As such, state prosecutors and judges will no longer feel pressure to let accused criminals have plea deals or to give them shorter prison sentences to ease the burdens of prison overcrowding and high overhead costs. 
  • The term “millionaire” will fall out of use in the U.S. and other Western countries since inflation will have rendered $1 million USD only as valuable as $90,000 USD was in 2019 (assuming a constant inflation rate of 3.0%).
  • There will still be major wealth and income inequality across the human race. However, wealth redistribution, better government services, advances in industrial productivity, and better technologies will ensure that even people in the bottom 1% have all their basic and intermediate life needs meet. In many ways, the poor people of 2100 will have better lives than the rich people of 2020.

2101 – 2200 AD

  • Humans will definitely stop being the dominant intelligent life forms on Earth. 
  • Many “humans” will be heavily augmented through genetic engineering, other forms of bioengineering, and cybernetics. People who outwardly look like the normal humans of today might actually have extensive internal modifications that give them superhuman abilities. Non-augmented, entirely “natural” humans like people in 2019 will be looked down upon in the same way you might today look at a very low IQ person with sensory impairments. Being forced by your biology to incapacitate yourself for 1/3 of each day to sleep will be tantamount to having a medical disability. 
  • Due to a reduced or nonexistent need for sleep among intelligent machines and augmented humans and to the increased interconnectedness of the planet, global time zones will become much less relevant. It will be common for machines, humans, businesses, and groups to use the same clock–probably Coordinated Universal Time (UTC)–and for activity to proceed on a 24/7 basis, with little regard of Earth’s day/night cycle. 
  • Physical disabilities and defects of appearance that cause untold anguish to people in 2019 will be easily and cheaply fixable. For example, male-pattern baldness and obesity will be completely ameliorated with minor medical interventions like pills or outpatient surgery. Missing or deformed limbs will be easily replaced, all types of plastic surgery (including sex reassignment) will be vastly better and cheaper than today, and spinal cord damage will be totally repairable. The global “obesity epidemic” will disappear. Transsexual people will be able to seamlessly alter their bodies to conform with their preferred genders, or to alter their brains so their gender identities conform with the bodies they were born with. 
  • All sleep disorders will be curable thanks to cybernetics that can use electrical pulses to quickly initiate sleep states in human brains. The same kinds of technologies will also reduce or eliminate the need for humans to sleep, and for people to control their dreams. 
  • Brain-computer interfaces will let people control, pre-program, and, to a limited extent, record their dreams. 
  • Through electrical signaling and chemical releases, the brain implants will be able to induce any type of mental or emotional state. This will include altered states of consciousness, like lucid dreaming, meditation, or intoxication. A person might have to go through a “calibration period” where the implants would monitor and record their brain activity while they experienced different things, and then, the user would experiment with the implant to see how well it could induce the recorded brain states. Through a process of guided trial and error, they would become masters of their own minds. This ability would make human life richer and more productive, as people could have valuable experiences during portions of the day when they would otherwise be bored or “switched off,” and to even do useful problem-solving tasks in their sleep.
  • Direct brain-to-computer interfaces and other advanced technologies will let humans enter virtual reality worlds that seem no different from the real world (the “Matrix scenario”), and to remotely control robot bodies located anywhere in the real world, with fully lifelike levels of sensory richness and fusion. Able to control perfect robot bodies of any design in the real world, and to take on any form in virtual worlds, some humans will have no use for real, fixed-form bodies, and will dispense with them, instead existing as “brains in jars.”  
  • Some “humans” will lack fixed, corporeal forms; they will be able to extensively modify their original bodies or to switch bodies at will. A person could take the form of something nonhuman, like a terrestrial squid.
  • Almost all of today’s diseases will be cured.
  • The means to halt and reverse human aging will be created. The human population will come to be dominated by people who are eternally young and beautiful. 
  • Humans and machines will be immortal. Intelligent beings will find it terrifying and tragic to contemplate what it was like for humans in the past, who lived their lives knowing they were doomed to deteriorate and die. 
  • Extreme longevity, better reproductive technologies that eliminate the need for a human partner to have children, and robots that do domestic work and provide companionship (including sex) will weaken the institution of marriage more than any time in human history. An indefinite lifetime of monogamy will be impossible for most people to commit to. 
  • At reasonable cost, it will be possible for women to create healthy, genetically related children at any point in their lives, and without using the 2019-era, pre-menopausal egg freezing technique. For example, a 90-year-old, menopausal woman will be able to use reproductive technologies to make a baby that shares 50% of her DNA. 
  • Opposite-sex human clones will exist. Such a clone would share 22-1/2 of their 23 chromosome pairs with their “original.” Only the final sex chromosome, which would be either a “Y” or a second “X”, would differ.  
  • Immortality, the automation of work, and widespread material abundance will completely transform lifestyles. With eternity to look forward to, people won’t feel pressured to get as rich as possible as quickly as possible. As stated, marriage will no longer be viewed as a lifetime commitment, and serial monogamy will probably become the norm. Relationships between parents and offspring will change as longevity erases the disparities in generational outlook and maturity that traditionally characterize parent-child interpersonal dynamics (e.g. – 300-year-old dad doesn’t know any better than his 270-year-old son). The “factory model” of public education–defined by conformity, rote memorization, frequent intelligence testing, and curricula structured to serve the needs of the job market–will disappear. The process of education will be custom-tailored to each person in terms of content, pacing, and style of instruction. Students will be much freer to explore subjects that interest them and to pursue those that best match their talents and interests. 
  • Radically extended human lifespans mean it will become much more common to have great-grandparents around. A cure for aging will also lead to families where members separated in age by many decades look the same age and have the same health. Additionally, older family members won’t be burdensome since they will be healthy.
  • The human population might start growing again thanks to medical immortality, to advanced fertility technologies including artificial wombs and cloning, and to robots that help raise children, reducing the workload for human parents. The human race won’t die out thanks to persistently low birthrates.
  • Thanks to radical genetic engineering, there will be “human-looking,” biological people among us that don’t belong to our species, Homo sapiens. Examples could include engineered people who have 48 chromosomes instead of 46, people whose genomes have been shortened thanks to the deletion of junk DNA, or people who look outwardly human but who have radically different genes within their 46 chromosomes, so they have different numbers or arrangements of internal organs (like two hearts), or even new types of internal organs, such as bird-like lung . Such people wouldn’t be able to naturally breed with Homo sapiens, and would belong to new hominid species. 
  • Extinct species for which we have DNA samples (ex – from passenger pigeons on display in a museum) will “resurrected” using genetic technology.
  • The global mass surveillance network will encompass unpopulated areas and wilderness areas, protecting animals from poaching. Extinctions of large, wild animals will stop.
  • The technology for safely thawing humans out of cryostasis and returning them to good health will be created. 
  • Suspended animation will become a viable alternative to suicide. Miserable people could “put themselves under,” with instructions to not be revived until the ill circumstances that tormented them had disappeared or until cures for their mental and medical problems were found. 
  • A sort of “time travel” will become possible thanks to technology. Suspended animation will let people turn off their consciousnesses until any arbitrary date in the future. From their perspective, no time will have elapsed between being frozen and being thawed out, even if hundreds of years actually passed between those two events, meaning the suspended animation machine will subjectively be no different from a time machine to them. FIVR paired with data from the global surveillance networks will let people enter highly accurate computer simulations of the past. The data will come from sources like old maps, photos, videos, and the digital avatars of people, living and dead. The computers simulations of past eras will get less accurate as the dates get more distant and the data scarcer.
  • It will be possible to upload human minds to computers. The uploads will not share the same consciousness as their human progenitors, and will be thought of as “copies.” Mind uploads will be much more sophisticated than the digitally immortal avatars that will come into existence in the 2030s.
  • Different types of AGIs with fundamentally different mental architectures will exist. For example, some AGIs will be computer simulations of real human brains, while others will have totally alien inner workings. Just as a jetpack and a helicopter enable flight through totally different approaches, so will different types of AGIs be capable of intelligent thought. 
  • Gold, silver, and many other “precious metals” will be worth far less than today, adjusting for inflation, because better ways of extracting (including from seawater) them will have been developed. Space mining might also massively boost supplies of the metals, depressing prices. Diamonds will be nearly worthless thanks to better techniques for making them artificially. 
  • The first non-token quantities of minerals derived from asteroid mining will be delivered to the Earth’s surface. (Finding an asteroid that contains valuable minerals, altering its orbit to bring it closer to Earth, and then waiting for it to get here will take decades. No one will become a trillionaire from asteroid mining until well into the 22nd century.)
  • Synthetic life forms will colonize parts of the world uninhabitable to humans, like mountaintops, oceans (both on the surface and under it), and maybe even underground regions. Intelligent and semi-intelligent machines will be common sights, even in remote areas.
  • Intelligent life from Earth will colonize the entire Solar System, all dangerous space objects in our System will be found, the means to deflect or destroy them will be created, and intelligent machines will redesign themselves to be immune to the effects of radiation, solar flares, gamma rays, and EMP. As such, natural phenomena (including global warming) will no longer threaten the existence of civilization.  Intelligent beings will find it terrifying and tragic to contemplate what it was like for humans in the past, who were confined to Earth and at the mercy of planet-killing disasters. 
  • “End of the World” prophecies will become far less relevant since civilization will have spread beyond Earth and could be indefinitely self-sustaining even if Earth were destroyed. Some conspiracy theorists and religious people would deal with this by moving on to belief in “End of the Solar System” prophecies, but these will be based on extremely tenuous reasoning. 
  • The locus of civilization and power in our Solar System will shift away from Earth. The vast majority of intelligent life forms outside of Earth will be nonhuman. 
  • A self-sustaining, off-world industrial base will be created.
  • Spy satellites with lenses big enough to read license plates and discern facial features will be in Earth orbit. 
  • Space probes made in our Solar System and traveling at sub-light speeds will reach nearby stars.
  • All of the useful knowledge and great works of art that our civilization has produced or discovered could fit into an advanced memory storage device the size of a thumb drive. It will be possible to pair this with something like a self-replicating Von Neumann Probe, creating small, long-lived machines that would know how to rebuild something exactly like our civilization from scratch. Among other data, they would have files on how to build intelligent machines and cloning labs, and files containing the genomes and mind uploads of billions of unique humans and non-human organisms. Copies of existing beings and of long-dead beings could be “manufactured” anywhere, and loaded with the personality traits and memories of their predecessors. Such machines could be distributed throughout our Solar System as an “insurance policy” against our extinction, or sent to other star systems to seed them with life. Some of the probes could also be hidden in remote, protected locations on Earth.
  • We will find out whether alien life exists on Mars and the other celestial bodies in our Solar System. 
  • Intelligent machines will get strong enough to destroy the human race, though it’s impossible to assign odds to whether they’ll choose to do so.
  • If the “Zoo Hypothesis” is right, and if intelligent aliens have decided not to talk to humans until we’ve reached a high level of intellect, ethics, and culture, then the machine-dominated civilization that will exist on Earth this century might be advanced enough to meet their standards. Uncontrollable emotions and impulses, illogical thinking, tribalism, self-destructive behavior, and fear of the unknown will no longer govern individual and group behavior. Aliens could reveal their existence knowing it wouldn’t cause pandemonium. 
  • The government will no longer be synonymous with slowness and incompetence since all bureaucrats will be replaced by machines.
  • Technology will be seamlessly fused with humans, other biological organisms, and the environment itself.  
  • It will be cheaper and more energy-efficient to grow or synthesize almost all types of food in labs or factories than to grow and harvest it in traditional, open-air farms. Shielded from the weather and pests and not dependent on soil quality, the amounts and prices of foods will be highly consistent over time, and worries about farmland muscling out or polluting natural ecosystems will vanish. Animals will no longer be raised for food. Not only will this benefit animals, but it will benefit humans since it will eliminate a a major source of communicable disease (e.g. – new influenza strains originate in farm animals and, thanks to close contact with human farmers, evolve to infect people thanks to a process called “zoonosis”).
  • Additionally, the means will exist to cheaply and artificially produce non-edible organic products, like wool and wood, in industrial quantities. This means anyone will be able to buy animal products that are very expensive today, like snakeskin boots or bear rugs. Unlimited quantities of perfectly simulated animal products that have useful properties, like pillow feathers (softness) or high-grade wool (heat insulation), will be available, and no animals will need to be harmed to make them. Lab-synthesized wood that is superior to “old-growth” timber will also exist.
  • A global network of sensors and drones will identify and track every non-microscopic species on the planet. Cryptids like “bigfoot” and the “Loch Ness Monster” will be definitively proven to not exist. The monitoring network will also make it possible to get highly accurate, real-time counts of entire species populations. Mass gathering of DNA samples–either taken directly from organisms or from biological residue they leave behind–will also allow the full genetic diversity of all non-microscopic species to be known. 
  • That same network of sensors and machines will let us monitor the health of all the planet’s ecosystems and to intervene to protect any species. Interventions could include mass, painless sterilizations of species that are throwing the local ecology out of balance, mass vaccinations of species suffering through disease epidemics, reintroductions of extinct species, or widescale genetic engineering of a species. 
  • The technology and means to implement David Pearce’s global “benign stewardship” of nonhuman organic life will become available.  (https://youtu.be/KDZ3MtC5Et8) After millennia of inflicting damage and pain to the environment and other species, humanity will have a chance to inaugurate an era free of suffering.
  • The mass surveillance network will also look skyward and see all anomalous atmospheric phenomena and UFOs.
  • Robots will clean up all of the garbage created in human history. 
  • Every significant archaeological site will be excavated and every shipwreck found. There will be no work left for people in the antiquities. 
  • Dynamic traffic lane reversal will become the default for all major roadways, sharply increasing road capacity without compromising safety. Autonomous cars that can instantly adapt to changes in traffic direction and that can easily avoid hitting each other even at high speeds will enable the transformation.
  • The Imperial system of weights and measures will fall out of use worldwide. Intelligent machines and posthumans will be able to switch to Metric without a problem. 

Interesting articles, December 2021

The experimental “Rapid Dragon” weapon is a cruise missile packed in a pallet. After being shoved out the open rear door of a cargo plane in flight, the pallet quickly disintegrates, and the missile’s engines turn on, guiding it to its target.
https://www.thedrive.com/the-war-zone/43550/special-operations-c-130-hits-target-with-a-rapid-dragon-pallet-dropped-cruise-missile

Interesting to see how quickly private air forces are growing.
https://www.thedrive.com/the-war-zone/43379/draken-doubles-its-fleet-of-private-aggressor-f-16s-with-a-dozen-surplus-jets-from-norway

The hideous East German helmet has an interesting history. It was designed during WWII as a replacement for the iconic M35 “Stahlhelm,” and tests showed it was stronger yet cheaper to make. In spite of that, Hitler rejected the adoption of the new helmets because they lacked the distinctive “German” appearance that the world had come to know and love. After the War, the Soviets made the new East German Army divest itself of weapons and equipment from the Nazi era, which meant the Stahlhelm was out. Luckily, the East Germans found the data sheets for the ugly helmet Hitler didn’t like, so they made it.
https://equipment.fandom.com/wiki/Stahlhelm_M1945

The Japanese attacked Pearl Harbor 80 years ago, dragging the U.S. into WWII. They also did a little-remembered and unsuccessful follow-up attack on the base a few months later.
https://www.thedrive.com/the-war-zone/43414/flying-boats-flew-japans-little-known-follow-on-raid-on-pearl-harbor

“If the Japanese had sunk every ship at Pearl Harbor and destroyed every plane in the Hawaiian Islands, we still would have been in a situation where our economy was 10 times larger than theirs. We had already begun building this enormous new fleet, and I believe the result — it might have taken us longer to win the war — but there’s no doubt that the (United States) would have reacted much like it did — (a) unified sense of rage and purpose. and as soon as that happened, I think Japan was finished.”
https://www.joplinglobe.com/news/local_news/acclaimed-historian-says-japan-lost-the-war-on-dec-7-1941/article_59862a68-52d4-11ec-8410-c31d1abee0fb.html

One reason the Japanese Zero fighter was better than its early U.S. opponents was it weighed less thanks to a superior aluminum alloy. The Americans eventually copied it after recovering a crashed Zero.
https://www.researchgate.net/publication/344352786_Extra_Super_Duralumin_and_Zero_Fighter_-Why_could_Japan_develop_the_strongest_aluminum_alloy_in_the_world-

The “Bailey Bridge” was a modular bridge invented in WWII. Crews could assemble it in the field by lining up the components and hammering thick, metal pegs through their corresponding holes. They also used permanent clamps to hold the parts together. I think we should assemble modular houses like this. After the concrete foundation was poured and dried, several 18-wheeler trucks and a small crane would arrive at the site. Each truck would carry a different, box-like section of the house. The crane would stack them in the right order, and construction workers would do the careful work of lining up the holes on the edges of each section’s frame with its neighbors, and then hammer pegs through the holes to secure them. All the “framing” work could be done in a day. The different components might even come with plumbing and electric wiring, only making it necessary for the workers to join the short gaps between each section.
https://www.youtube.com/watch?v=vaNOuVgwPos

Since WWII, there have been few improvements to the Bailey Bridge design. They got it very close to right the first time.
https://www.mabeybridge.com/products/bridging/mabey-logistic-support-bridge

Here’s a fascinating analysis of how the Allies gained the upper hand against the Axis Powers in 1942 with weapons production and readiness.
https://www.youtube.com/watch?v=jKCScYZoEj4

The Soviet T-55 tank began production in 1958, and the vehicles are still being remanufactured, upgraded and traded. Though much worse than modern main battle tanks, upgraded T-55s are still effective against lighter armored vehicles and against enemies that lack armor. Pakistan just bought them for such a role.
https://www.youtube.com/watch?v=-7vKHiygopk

Here’s an intense drone video of Syrian and Turkish armored vehicles in a sort of “dogfight.”
https://www.youtube.com/watch?v=MGtosGZsztY

More information about the encounters between U.S. Navy warships and UFOs off the coast of San Diego in 2019 have come to light. At one point, sailors fired a weapon at one of the craft.
https://www.thedrive.com/the-war-zone/43561/mysterious-drone-swarms-over-navy-destroyers-off-california-went-on-for-weeks

Elon Musk says Space-X might go bankrupt if it fails to get its upcoming “Starship” rockets operational on time.
https://www.theverge.com/2021/11/30/22809720/elon-musk-spacex-raptor-engine-crisis-bankruptcy-starship

There’s skepticism over the viability of the “SpinLaunch” system, which is meant to fling satellites into space.
https://theness.com/neurologicablog/index.php/is-spinlaunch-viable/

“The first images from [the James Webb telescope] should be available by summer 2022.”
https://earthsky.org/space/james-webb-space-telescope-30-days-of-terror/

‘A solar balloon is a balloon that gains buoyancy when the air inside is heated by solar radiation, usually with the help of black or dark balloon material. The heated air inside the solar balloon expands and has lower density than the surrounding air. As such, a solar balloon is similar to a hot air balloon.’
https://en.wikipedia.org/wiki/Solar_balloon

Whether life exists on Venus, it’s interesting that our scientists can conceive of life forms that could theoretically survive there. We could someday synthesize such microbes in cloning labs on Earth, and then send them to Venus on space rockets to seed that planet with life (some people think aliens did that to Earth billions of years ago). I don’t think we should do it yet, though. We should build stations orbiting Venus to slowly siphon its thick atmosphere into space for whatever industrial uses we have, until the atmosphere is about as thick as Earth’s. At that point, we could seed the planet with a greater abundance of engineered life forms, and of greater biological complexity.
https://news.mit.edu/2021/habitable-venus-clouds-acid-neutralizing-1220

The Soviet Union collapsed 30 years ago.
https://apnews.com/article/europe-russia-ukraine-voting-rights-moscow-2f0a5907e9172c5a7c14451d39752f5d

Sri Lanka’s government-mandated attempt to switch to 100% organic farming is being called off after it caused a sharp drop in food production.
https://www.thehindu.com/news/international/sri-lanka-revokes-ban-on-fertilizers/article37614990.ece

Again and again, the doomsday global warming prediction models have proved inaccurate.
https://www.science.org/content/article/un-climate-panel-confronts-implausibly-hot-forecasts-future-warming

Adjusting for inflation, the amount of property damage caused by tornadoes in the U.S. didn’t increase from 1950 to 2021. If anything, there was a slight decrease.
https://rogerpielkejr.substack.com/p/us-tornado-damage-1950-to-2021

The “greater male variability hypothesis” says that human males are more genetically diverse than females because males have only one “X” chromosome whereas females have two. Any recessive traits and/or extreme traits that may be on a person’s X chromosome are expressed in males, but tend to be muffled in females because of the effect of the second X chromosome. A growing body of evidence from human medical studies supports the Hypothesis, and now, it seems the same effect applies to chimpanzees.
https://royalsocietypublishing.org/doi/full/10.1098/rspb.2019.2858

“Intellectual disability,” formerly termed “mental retardation,” is a condition where a person’s IQ is lower than 70. It has many different causes, including the inheritance of genes from naturally low-IQ parents, totally random genetic mutations in the womb (Down Syndrome), brain damage during difficult births (Cerebral Palsy), and severe neglect/abuse during early childhood. As a result, while intellectually disabled adults might seem the same, they can have low IQs for diverse reasons. This study filtered out the intellectually disabled people who were so due to random genetic mutations, birthing accidents, and early childhood trauma. It determined that low IQ is very heritable, just as other studies have shown high IQ is very heritable. Also, an unusually high number of genes coding for IQ are located on the X-chromosome, which explains why the population “bell curve” for IQ is narrower for women than it is for men. Men are likelier to be intellectually disabled and also likelier to be geniuses than women.
https://onlinelibrary.wiley.com/doi/abs/10.1111/jcpp.13560

Three percent of Greenlanders have a genetic mutation causing sucrase-isomaltase deficiency. It allows them to eat large amounts of sugary foods without weight can or health penalties.
https://www.studyfinds.org/genetic-variant-sugar-healthy-greenland/

‘We found evidence that most countertop microwaves sold in the US are manufactured by just one company, Midea. We confirmed with Midea that it makes and sells Toshiba, Comfee, and Black+Decker ovens. We’re also confident that GE, Whirlpool, Sharp, Breville, Insignia, Magic Chef, Hamilton Beach, and others also sell microwaves that were originally built and probably designed in large part by Midea, though all parties that we contacted declined to comment.’
https://www.nytimes.com/wirecutter/reviews/best-microwave/#most-microwaves-come-from-the-same-factory

Did microwave technology peak in 1997 with the Sharp “Carousel”?
https://youtu.be/UiS27feX8o0

Recumbent bicycles are more efficient than traditional “vertical” bikes because the rider is positioned in a way that reduces his drag. By one estimate, recumbent bikes are 23% more “fuel efficient.”
https://www.quora.com/How-much-more-efficient-are-recumbent-bicycles-than-conventional-ones

‘Since at least 2017, a mysterious threat actor has run thousands of malicious servers in entry, middle, and exit positions of the Tor network in what a security researcher has described as an attempt to deanonymize Tor users.’
https://therecord.media/a-mysterious-threat-actor-is-running-hundreds-of-malicious-tor-relays/

This analysis concludes that Moore’s Law is not dead, though it could die at any minute. Regardless, there is still enormous room for the improvement of silicon computer chips, and we’re decades away from hitting limits imposed by physics.
https://www.lesswrong.com/posts/aNAFrGbzXddQBMDqh/moore-s-law-ai-and-the-pace-of-progress

Intel claims it has developed 3D computer chips, which will let it continue Moore’s Law rates of improvement.
https://www.electronicdesign.com/technologies/embedded-revolution/article/21183706/electronic-design-intel-proposes-new-path-for-moores-law-with-3d-stacked-transistors

New AI language models answer questions more accurately than GPT-3 in niche domains. In aggregate, the models are approaching the levels of human experts.
https://deepmind.com/blog/article/language-modelling-at-scale

The E.U. has rejected the controversial new Alzheimer’s drug, “aducanumab,” due to a lack of evidence that it works.
https://www.bbc.com/news/health-59699907

The odds of a pandemic starting because a corpse full of something like the Spanish Flu or smallpox thawed out of the permafrost are about zero.
https://astralcodexten.substack.com/p/ancient-plagues

In the U.S., the COVID-19 vaccines have saved about 1 million lives so far.
https://marginalrevolution.com/marginalrevolution/2021/12/how-many-lives-has-vaccination-saved.html

Merck’s new drug to treat COVID-19 is less effective than claimed, but still better than nothing.
https://www.science.org/content/blog-post/rethinking-molnupiravir

In August of 2020, Bill Gates predicted that “for the rich world, we should largely be able to end [the COVID-19 epidemic] by the end of 2021, and for the world at large by the end of 2022.” At least in the U.S., the prediction would have been right had people continued getting vaccinated at the rate they were during April 2021.
https://www.wired.com/story/bill-gates-on-covid-most-us-tests-are-completely-garbage/

Bill Gates predicts that the “acute phase” of the COVID-19 pandemic will end sometime in 2022.
https://www.gatesnotes.com/About-Bill-Gates/Year-in-Review-2021

At a Taiwanese pathogen lab, a worker accidentally contracted COVID-19 after an accident. This bolsters the plausibility of the “lab leak hypothesis” for the virus’ origin.
https://www.taiwannews.com.tw/en/news/4372853
http://www.rationaloptimist.com/blog/covid-lab-leak-in-taiwan/

The U.S. death toll from COVID-19 hit 800,000. Given the availability of vaccines, there’s no reason the milestone had to be reached this year.
https://apnews.com/article/coronavirus-pandemic-nfl-sports-health-football-f967f156fb205277266d2bb2483b858c

A year ago, the U.S. finished vaccinating its first million citizens against COVID-19. The lack of side effects in that cohort at this stage suggests it is safe as its developers claimed in the beginning.
https://www.cnbc.com/2020/12/23/covid-vaccine-us-has-vaccinated-1-million-people-out-of-goal-of-20-million-for-december.html

Interesting articles, November 2021

During WWII, the ever-frugal and innovative Germans built tanks out of spare parts they captured from other armies. They found ways to wring utility out of obsolete equipment, both foreign- and domestically made. The “Marder” armored vehicles were outstanding examples of this.
https://www.youtube.com/watch?v=LyVyFGLX1TY

Henry Kissinger: “I don’t expect an all-out attack on Taiwan in, say, a 10-year period, which is as far as I can see.”
https://www.politico.com/news/2021/11/21/kissinger-china-taiwan-summit-biden-523139

‘Mooring retired Ticonderoga class cruisers around Guam could offer an efficient way to greatly expand the island’s missile defense umbrella.’
https://www.thedrive.com/the-war-zone/41819/decommissioned-navy-cruisers-could-be-the-answer-to-guams-missile-defense-needs

A confrontation between U.S. Navy warships and Iranian gunboats in the Gulf of Oman came frighteningly close to disaster.
https://www.thedrive.com/the-war-zone/42987/video-shows-u-s-destroyers-very-intimate-standoff-with-iranian-vessels-over-seized-oil-tanker

This video shows the differences between how high explosive squash-head (HESH) and high explosive anti-tank (HEAT) weapons work. Note that HESH rounds work well when they flatten out against the surface of a tank before exploding, like a spherical glob of mud falling onto a hard floor and splattering into a pancake. Conversely, HEAT rounds work best when they explode at the instant they touch the surface of a tank, like a round, porcelain piggy bank being dropped and shattering just as it hits the ground.
https://youtu.be/Uhz3w8-PSl8

There are such things as bullets that explode on impact. They’re meant to penetrate heavy metal/ceramic body armor and light vehicle armor.
https://youtu.be/5Dqg5k_kdPw

The first patent for a percussion-cap musket was patented in England in 1807. Sportsmen there and in America soon discovered they were more reliable than their flintlock muskets, and adopted them in significant numbers. However, it wasn’t until 1834 that the British Army considered the new weapons for use. The tests showed that the percussion cap muskets were 26 times more reliable, and the British quickly adopted the new guns. Like Mikhail Kalashnikov 100 years later, the inventor of the revolutionary new weapon, Alexander Forsythe, received no compensation.
https://weaponsandwarfare.com/british-army-1820-45/

Bullets should always be made of softer metal than the gun barrels they are shot out of. If the bullets are harder, then they will ruin the barrel’s rifling, or even get jammed in the middle of the barrel after being fired, possibly causing the gun to explode.
https://www.quora.com/Would-a-solid-tungsten-carbide-bullet-ruin-the-weapons-rifling-because-of-the-hardness-of-tungsten

Someone tried to kill Iraq’s prime minister by flying an explosive-laden drone into his house and detonating it. Recall my prediction that, before 2030, “Drones will be used in an attempted or successful assassination of at least one major world leader.” This man wasn’t high-profile enough to satisfy my prediction.
https://www.bbc.com/news/world-middle-east-59195399

The U.S. military’s experimental “Gremlin” drones, which un-dock in midair from larger “mothership” planes, perform missions, and then fly back to the motherships and dock with them, are getting more refined.
https://youtu.be/H4T6Vr4a1hY

In the first attack of its kind, a quadcopter drone was used in an attempt to disable a power station in the U.S. It failed, and the police found the drone. The person or people responsible were careful to remove identifying information from the machine, and remain unknown. In the future, drones and narrow AI will untether each country’s military strength from the size of its human population.
https://www.thedrive.com/the-war-zone/43015/likely-drone-attack-on-u-s-power-grid-revealed-in-new-intelligence-report

AI scientist Stuart Russell foresees a day when a truck full of “a million” killer drones could drive into a city, release its load, and kill all the human inhabitants.
https://www.theguardian.com/technology/2021/oct/29/yeah-were-spooked-ai-starting-to-have-big-real-world-impact-says-expert

The failed WWII “Bat Bomb” will turn out to have been an idea ahead of its time.
https://en.wikipedia.org/wiki/Bat_bomb

“Nitinol” is a remarkable alloy, and objects made from it “remember” their original shapes and revert to them, even after being bent or stretched into something else.
https://www.youtube.com/watch?v=wI-qAxKJoSU

Before 1856, it was really rare to see purple-colored objects because purple dye was very expensive to make (snails had to be collected and boiled). But that year, a German chemist discovered a way to synthesize it from coal tar at very low cost. Almost overnight, every snail-boiling business went bankrupt, and by 1859, a fashion fad of wearing purple clothes swept the U.S. and Western Europe. It’s a perfect example of how technology brings things that were once the exclusive domain of the rich to everyone else.

The process will not stop. Consider that autonomous cars will make poor people like rich men with chauffeurs, in vitro synthesis of meat will make the finest caviars and steaks as cheap as SPAM, and virtual reality will give everyone access to whatever fantastic worlds they desire.
https://www.smithsonianmag.com/smart-news/in-ancient-rome-purple-dye-was-made-from-snails-1239931/

There is actually an upper limit to how loud sounds can be.
“For a sound of 194 decibels, the trough of the fluctuation would just touch zero, which is the vacuum pressure. It can’t go any lower than that, so a sustained sound greater than 194 decibels is not possible.”
https://www.quora.com/How-do-we-know-that-194-decibels-is-the-loudest-sound-possible

This website catalogs all the futuristic technologies mention in science fiction books, along with the years when they are supposed to be (or were supposed to have been) real.
http://technovelgy.com/

Sand behaves like a liquid when injected with compressed air.
https://www.youtube.com/watch?v=My4RA5I0FKs

This elegantly simple video shows how an ammeter measures the amount of electric current flowing through a circuit.
https://www.youtube.com/watch?v=8DanoG-nowk

Dow 36,000 is a reality…16 years after James Glassman and Kevin Hassett predicted it would be. Do they deserve credit for getting the prediction right?
https://www.cnn.com/2021/11/03/investing/premarket-stocks-trading/index.html

After a few hours of practice, machines can learn to play Atari video games much better than average humans.
http://arxiv.org/abs/2111.00210

“GauGAN2 uses a deep learning model that turns a simple written phrase, or sentence, into a photorealistic masterpiece.”
https://blogs.nvidia.com/blog/2021/11/22/gaugan2-ai-art-demo/

Economist Tyler Cowen predicts that greater use of telework will expose most American jobs having to do with computer coding and IT to cheaper foreign competition (“teleshock”). Many formerly secure and well-paying jobs will vanish. Conversely, some types of culture-specific and location-dependent jobs will remain secure.
https://marginalrevolution.com/marginalrevolution/2021/11/the-teleshock.html

Automated cranes that load and unload cargo from ships greatly increase the efficiency of ports. Unfortunately, the $1.2 trillion infrastructure bill that was just passed in the U.S. provides no money for such upgrades, presumably because unions afraid of losing jobs successfully lobbied the bill’s authors to exclude it.
https://reason.com/2021/11/09/americas-ports-need-more-robots-but-the-1-trillion-infrastructure-bill-wont-fund-port-automation/

Major volcanic eruptions and the concomitant reductions in agricultural output caused, or helped to cause, the collapses of several Chinese empires.
https://www.nature.com/articles/s43247-021-00284-7

Coal and crude oil can be turned into an edible, fat-rich substance similar to margarine. This study examines the feasibility of feeding the population with it during a global calamity that blocks out the sun for several years (ex – nuclear winter, megavolcano eruption, asteroid strike).
https://www.sciencedirect.com/science/article/pii/S0263876221004275

A Russian town on the Arctic Ocean is now getting heat from a barge with a nuclear reactor in it.
https://indianexpress.com/article/world/climate-change/a-nuclear-powered-shower-russia-tests-a-climate-innovation-7608185/

What would change if clean energy prices were super low? Among other things, we would start terraforming the deserts.
https://marginalrevolution.com/marginalrevolution/2021/11/what-would-a-world-with-very-cheap-energy-look-like.html

This awesome illustration shows different fusion reactor designs that have been proposed.

“SpinLaunch” is a giant centrifuge designed to hurl satellites into space. I wonder how much it would help to build such a machine on the summit of a tall mountain.
https://www.thedrive.com/the-war-zone/43079/space-launch-start-up-just-used-a-giant-centrifuge-to-hurl-a-projectile-into-the-upper-atmosphere

NASA had further plans for the Apollo Program and the Saturn V rockets, including the construction of a Moon base, and a manned Venus flyby.
https://www.youtube.com/watch?v=f4CTI5GDz98

Thanks to a small computer chip implanted in his brain, a partially paralyzed man was able to use his thoughts to “[achieve] typing speeds of 90 characters per minute with 94.1% raw accuracy online, and greater than 99% accuracy offline with a general-purpose autocorrect.”
https://www.nature.com/articles/s41586-021-03506-2

People can make themselves taller by getting surgery where the legs are broken, the two halves are pulled apart a tiny amount, and then the body is allowed to heal by slowly filling in the gap with new bone. A woman with dwarfism underwent the procedures several times over four, agonizing years, to increase her height from 3′ 9″ to 4′ 11″. She even did it to her arms to keep them proportionate with her elongating body.
https://www.thesun.co.uk/news/16645445/underwent-years-agonising-surgery-stretch-my-bone/

The first drug to treat dwarfism has been FDA approved. It has to be injected every day, each daily dose costs $900, and it adds 1/2 inch of height per year of use.
https://medcitynews.com/2021/11/biomarin-pharma-sees-big-things-for-first-fda-approved-dwarfism-drug/

More accurate methods of screening human embryos for fatal genetic defects have raised IVF success rates.
https://infoproc.blogspot.com/2021/11/preimplantation-genetic-testing-for.html

Surprisingly, having low levels of cholesterol can also kill you.
“[There] is a U-shaped relationship between LDL-C level and all-cause mortality.”
https://www.nature.com/articles/s41598-021-01738-w

Variations of AlphaFold are now simulating protein-protein interactions with high levels of accuracy.
https://www.science.org/content/blog-post/protein-complex-structure-predictions-already

In Britain, mass vaccination of girls and young women with the HPV vaccine has caused an 87% drop in cervical cancer cases since 2008.
https://www.bbc.com/news/health-59148620

This article is over a year old. Not bad: ‘This is the end of the coronavirus pandemic. And this is how it could happen in the United States: By November 2021, most Americans have received two doses of a vaccine that, while not gloriously effective, fights the disease in more cases than not. Meanwhile, Americans continue to wear masks and avoid large gatherings, and the Covid-19 numbers drop steadily after a series of surges earlier in the year. Eventually, as more and more Americans develop immunity through exposure and vaccination, and as treatments become more effective, Covid-19 recedes into the swarm of ordinary illnesses Americans get every winter.’
https://www.politico.com/news/magazine/2020/09/25/how-covid-19-pandemic-ends-421122

Large numbers of deer in the U.S. are testing positive for COVID-19, meaning they (perhaps along with other mammals) will serve as a reservoir for the virus, and will periodically infect humans for the foreseeable future. COVID-19 will become endemic.
https://www.npr.org/sections/goatsandsoda/2021/11/10/1054224204/how-sars-cov-2-in-american-deer-could-alter-the-course-of-the-global-pandemic

A new drug that sharply lowers the odds of dying from COVID-19 has been unveiled.
https://www.bbc.com/news/health-59178291

Thanks to genetics, South Asians are likelier to die from COVID-19 than Europeans.
https://www.bbc.com/news/health-59165157

Ivermectin does little or nothing against the COVID-19 virus. The medical studies that showed it reducing deaths were done in tropical countries where worm parasite infections are common. This means ivermectin saves the lives of some COVID-19 infectees by killing off their parasites, which are weakening their immune systems just enough to let COVID-19 to kill them. There is no controversy over ivermectin’s value as an antiparasite drug.
https://astralcodexten.substack.com/p/ivermectin-much-more-than-you-wanted

The evidence that COVID-19 might be manmade is inconclusive, but can’t be dismissed.
http://www.rationaloptimist.com/blog/the-covid-lab-leak-theory-just-got-even-stronger/

A new variant of COVID-19, the “Omicron” strain (I love the menacing name!), has emerged in South Africa. Much remains unknown at this point, but there’s preliminary evidence that it is more transmissible than even the Delta strain, and may be resistant to the vaccines.
https://www.reuters.com/world/new-coronavirus-variant-omicron-keeps-spreading-australia-detects-cases-2021-11-28/

Interesting articles, October 2021

The successes of the Anglo-American codebreakers of WWII get a lot of attention today, but their German counterparts had successes of their own.
https://www.independent.co.uk/news/uk/home-news/revealed-careless-mistake-bletchley-s-enigma-code-crackers-cost-allied-lives-5360971.html

A related article about other Anglo-American blunders that cost many lives during the WWII Battle of the Atlantic.
https://legionmagazine.com/en/2011/02/an-american-blunder-navy-part-43/

Here’s an in-depth video about the Israeli M-50 “Super Sherman” tank. They were mishmashes of leftover M4 tank components.
https://www.youtube.com/watch?v=_KyNYO03g1E

In a testament to Pentagon waste, the first Littoral Combat Ship–really just a frigate that was loaded up with too much unproven technology–has been decommissioned after only 13 years of service.
https://www.thedrive.com/the-war-zone/42582/the-first-littoral-combat-ship-has-been-decommissioned-after-just-13-years-of-service

Sadly, the former U.S. aircraft carriers Kitty Hawk and John F. Kennedy won’t be turned into museum ships and will be scrapped instead.
https://www.thedrive.com/the-war-zone/42643/the-navys-last-conventionally-powered-aircraft-carriers-have-been-sold-for-literal-pennies

China has converted several of its decommissioned 1950s-era fighter planes into drones. In a conflict, they would be used sacrificially for reconnaissance and to distract the enemy. I think “drone upgrade kits” could be used to breathe new life into all sorts of obsolete weapons, though they wouldn’t be as effective as new weapons designed from the outset to be crewless. I’m reminded of the German WWII practice of using captured enemy weapons, and even continuing the manufacture the better ones after taking over the countries where their factories were located. If the U.S. Army captured a bunch of Soviet- or Chinese-made tanks in a future war, the old excuse about how we can’t use them because we don’t know how they work or how to maintain them would disappear. The robots would figure it out.
https://www.defensenews.com/global/asia-pacific/2021/10/20/china-shows-off-drones-recycled-from-soviet-era-fighter-jets/

China tested an intercontinental, hypersonic missile that could, if developed further, evade U.S. antimissile systems. While this development should give the American military pause, it should also not be overestimated.
https://www.defensenews.com/land/2021/10/20/chinese-hypersonic-missile-test-unlikely-to-trigger-arms-race-experts-say/

A really bad but plausible war scenario would involve Taiwan surrendering after a Chinese invasion faster than the U.S. can move large numbers of troops there to fight. If Taiwan calls it quits after barely putting up a fight, it will be very hard for politicians over here to justify a costly war of liberation for apparent cowards who didn’t value their own freedom enough to fight hard.
https://www.wsj.com/articles/taiwan-military-readiness-china-threat-us-defense-11635174187

Like several previous attempts, the Pentagon’s latest project to equip ground troops with augmented vision goggles has failed.
https://www.thedrive.com/the-war-zone/42739/army-halts-widely-hyped-multi-billion-dollar-advanced-augmented-reality-goggle-program

A reality check for anti-American declinists.
https://www.foreignaffairs.com/articles/middle-east/2021-10-18/russia-no-mideast-superpower

In an ultimate Steampunk moment, several British battleships fired a fusillade of shells at a German zeppelin during the 1916 Battle of Jutland.
http://www.avalanchepress.com/Zeppelin_Scouting.php

The “Cyclocrane” was a unique airship built in the 1980s to lift heavy loads.
https://www.youtube.com/watch?v=CiU71GFs4Fs

This prediction from Kremlin “experts” that Putin would step down due to Parkinson’s disease didn’t age well.
https://www.the-sun.com/news/1752270/vladamir-putin-resign-next-year-health/

‘Currently, we produce ∼1021 digital bits of information annually on Earth. Assuming a 20% annual growth rate, we estimate that after ∼350 years from now, the number of bits produced will exceed the number of all atoms on Earth, ∼1050. After ∼300 years, the power required to sustain this digital production will exceed 18.5 × 1015 W, i.e., the total planetary power consumption today, and after ∼500 years from now, the digital content will account for more than half Earth’s mass, according to the mass-energy–information equivalence principle.’
https://aip.scitation.org/doi/10.1063/5.0019941

China has its own version of GPT-3.
https://www.gwern.net/docs/ai/scaling/2021-10-11-xinzhiyuan-inspursource10gpt245b.html

Using a language “entropy” model, this guy predicts that the first machine will probably pass the Turing Test by early 2025.
https://www.metaculus.com/notebooks/8329/human-level-language-models/

The “International Phonetic Alphabet” is an alphabet containing one letter for every sound that exists in every human language. Languages as diverse as English, Mandarin and Arabic can all be written IPA, with no distortion. The IPA even has symbols representing the “click” sounds that some tribal African languages have. If everyone in the world used the IPA, it would make it easier to learn other languages.
https://www.youtube.com/watch?v=XTzkT3j9pHI

‘Yet despite the complexity of birdsong and whale song, animals don’t seem to have that much to say to each other. “Stay away from my territory,” “Beware of the leopard” and “Come mate with me” sum up most of the messages we expect from animals. They could combine their sounds in almost infinitely varied ways, but they use just the tiniest fraction of these possibilities.’
https://www.wsj.com/articles/alien-languages-may-not-be-entirely-alien-to-us-11616817660

The “Waddington Effect” says that too much maintenance on a machine actually reduces the machine’s reliability because of the nonzero chances that workers will accidentally damage the machine during maintenance, or that a newly installed part will be defective.
http://livingstingy.blogspot.com/2011/03/waddington-effect.html

In the 1970s and 80s, it was believed that hovercraft would make traditional ships obsolete. That never happened, principally thanks to hovercrafts’ high costs.
https://www.popularmechanics.com/military/navy-ships/a16836/hovercrafts-failed/

The technology used in airplane black boxes and in aircraft traffic control systems is shockingly primitive. Your smartphone is much more advanced.
https://www.theguardian.com/commentisfree/2014/mar/09/malaysia-airlines-flight-mh370-black-box

Tesla is now a trillion-dollar company.
https://finance.yahoo.com/news/tesla-briefly-hits-1-trillion-market-cap-as-shares-rally-to-record-high-171840451.html

Someone is keeping close track of Elon Musk’s future predictions, and is collecting betting odds of their success.
https://www.metaculus.com/visualizations/elon-musk-timeline/

Twelve predictions for the future of music. Imagine a lifelike hologram of Elvis singing in your local bar.
https://tedgioia.substack.com/p/12-predictions-for-the-future-of

I agree with almost all of Erik Hoel’s predictions for 2050.
https://erikhoel.substack.com/p/futurists-have-their-heads-in-the

Foldable QLED sheets bring us a step closer to active camouflage clothing.
https://www.nature.com/articles/s41928-021-00643-4

‘What does ultraviolet look like? Prof Stark possesses UV vision because he is aphakic in one eye and, with Professor Karel Tan, has published research on the nearest visible equivalent. His conclusion is that it looks whitish blue or, for some wavelengths, a whitish violet. This appears to be because the three types of colour receptor (red, green and blue) have similar sensitivity to ultraviolet, so it comes out as a mixture of all three – basically white, but slightly blue because the blue sensors are somewhat better at picking up UV.’
https://www.theguardian.com/science/2002/may/30/medicalscience.research

The peer-review process is good at weeding out the worst papers, but that’s it. The “quality” ratings that peer reviewers assign to published papers are highly subjective, and there’s little positive correlation between a published paper’s quality score and its eventual “impact” score, which refers to how many times other academics cite the paper in their own work. A large fraction of papers that can only get published in second-tier journals actually belong in top-tier journals, and vice versa.
http://arxiv.org/abs/2109.09774

In birds, the mother carries the sex-determining chromosome instead of the father.
https://en.wikipedia.org/wiki/ZW_sex-determination_system

‘During a routine analysis of biological samples from two California condors in the San Diego Zoo Wildlife Alliance’s managed breeding program, scientists confirmed that each condor chick was genetically related to the respective female condor (dam) that laid the egg from which it hatched. However, in a surprising twist, they found that neither bird was genetically related to a male—meaning both chicks were biologically fatherless; and accounted for the first two instances of asexual reproduction, or parthenogenesis, to be confirmed in the California condor species.’
https://stories.sandiegozoo.org/2021/10/28/san-diego-zoo-wildlife-alliance-conservation-scientists-report-first-confirmed-hatchings-of-two-california-condor-chicks-from-unfertilized-eggs/

A good argument in favor of human embryo genetic screening, even with today’s limited technology.
https://infoproc.blogspot.com/2021/10/embryo-screening-and-risk-calculus.html

For the first time, a pig kidney was transplanted into a human and not rejected by the person’s immune system. The organ had been genetically modified to trick the human immune system. This technology might end organ shortages, saving countless lives and sparing millions from suffering.
https://www.reuters.com/business/healthcare-pharmaceuticals/us-surgeons-successfully-test-pig-kidney-transplant-human-patient-2021-10-19/

Here’s a relatively readable summary of what the Nobel Prize in Chemistry was awarded for (improved techniques for synthesizing molecules with specific chiralities).
https://www.science.org/content/blog-post/organocatalysis-nobel

Facebook is training computers to recognize the tasks people are doing, in the hopes that the algorithms can someday be installed in augmented reality glasses.
https://www.bbc.com/news/technology-58896551

Facebook has rebranded itself as “Meta,” to emphasize the company’s new focus on technologies and services beyond social media.
https://apnews.com/article/facebook-meta-mark-zuckerberg-technology-business-5ad543ab7780caae435935f0aca9fac6

Turboprops and piston plane engines look the same on the outside, but are completely different on the inside: A turboprop engine is actually a jet engine that spins an externally mounted propeller. A piston engine, on the other hand, is fundamentally the same thing as a car engine, though the components are usually laid out differently from their analogs in cars.
https://youtu.be/CTsBi6WOGWQ

1924 U.S. government report “Jet Propulsion for Airplanes”: ‘But to return from such speculations to the quantitative results of the computations, there does not appear to be, at present, any prospect whatever that jet propulsion of the sort here considered will ever be of practical value, even for military purposes.’
https://ntrs.nasa.gov/api/citations/19930091225/downloads/19930091225.pdf

The “Solar Power Satellite by means of Arbitrarily Large Phased Array” is a proposed satellite that could generate enough power to satisfy 1/3 of humanity’s total consumption.
https://www.nasa.gov/pdf/716070main_Mankins_2011_PhI_SPS_Alpha.pdf

A Russian crew just filmed the first movie in space.
https://www.bbc.com/news/world-europe-58944660

China has put a new space station into orbit.
https://www.npr.org/2021/10/16/1046742793/china-space-station-docking-astronauts

William Shatner became the oldest person to ever go into space.
https://www.bbc.com/news/science-environment-58885555

NASA Chief Bill Nelson acknowledges that the USS Nimitz UFO sightings may have been alien spaceships.
https://twitter.com/UAP1949/status/1451803056421670913

South Korea builds nuclear power plants more cheaply than the U.S. or anyone else. There’s no reason why safe, advanced reactors can’t be built today at affordable cost.
https://www.sciencedirect.com/science/article/pii/S0301421516300106

Advocates of nuclear fusion power have a habit of deceptively expressing how efficient their experimental reactors are, making it look like they are much closer to the energy “break-even” point than they really are.
https://backreaction.blogspot.com/2021/10/how-close-is-nuclear-fusion-power.html

California’s banning of gas-powered lawn machines shows how badly we need to invent more energy-dense batteries.
https://www.latimes.com/california/story/2021-10-09/california-moves-toward-ban-on-gas-lawnmowers-and-leaf-blowers

Rising atmospheric CO2 levels have significantly boosted U.S. crop yields. Plants use CO2 to sustain their growth.
https://www.nber.org/papers/w29320

Genetically engineering cows to burp less could reduce their methane emissions by up to 50%.
https://www.cbsnews.com/video/uk-researchers-study-genetic-link-to-methane-emissions-in-cattle-key-to-reducing-greenhouse-gases/

Genetically engineering a bacterium to fix atmospheric nitrogen into biomolecules, and then splicing the bacterium into crop cells, where they would become as ubiquitous as organelles like mitochondria and chloroplasts, would massively boost farm output and cut the need for nitrogen fertilizers.
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1331173&HistoricalAwards=false

Unsurprisingly, genetics strongly influence how much people benefit from exercise. There really are people who can get six-pack abs from just a little diet and exercise, and others who can’t get them no matter how hard they try.
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0249501

‘More generally, one can see top chefs producing recipes that are then scaled not just to restaurants but also to home robot preparation services. Meals would be produced by a subscription service (“We have 10,000 recipes from the greatest chefs on every continent.”). Restaurants would compete even more on ambience.’
https://marginalrevolution.com/marginalrevolution/2021/10/the-new-top-chef.html

This paper estimates that a 1 kg, 1 liter in volume computer made of the most powerful computer chips allowed by the laws of physics would do 1051 operations per second and would have a memory capacity of 1031 bits. Even if those theoretical maxima of performance can’t be attained, and future computer engineers hit a “practical peak” that is several orders of magnitude lower than those figures, the resulting machine would be incomprehensibly powerful.
https://arxiv.org/pdf/quant-ph/9908043.pdf

‘For the entire history of computing, our calculating machines have operated in a way that causes the intentional loss of some information (it’s destructively overwritten) in the process of performing computations. But for several decades now, we have known that it’s possible in principle to carry out any desired computation without losing information—that is, in such a way that the computation could always be reversed to recover its earlier state. This idea of reversible computing goes to the very heart of thermo­dynamics and information theory, and indeed it is the only possible way within the laws of physics that we might be able to keep improving the cost and energy efficiency of general-purpose computing far into the future.’
https://spectrum.ieee.org/the-future-of-computing-depends-on-making-it-reversible

In a recent study, 80% of people with type 2 diabetes were able to put their disorders into remission simply by losing weight. The higher obesity rate and the concomitantly higher rates of obesity-related illnesses like diabetes explain much of the reason why the U.S. spends more on healthcare than other rich countries and has worse public health metrics.
https://onlinelibrary.wiley.com/doi/abs/10.1111/dme.14395

Obesity raises the risk of cancer in the digestive tract. Obesity is just bad and can’t be rationalized.
https://www.studyfinds.org/bmi-cancer-risk/

A new malaria vaccine has been approved by the WHO for use in Africa. It lowers the odds of infection by about 40% and is cheap to make.
https://www.bbc.com/news/health-58810551

The FDA has approved an e-cigarette as an aide for people who want to stop smoking regular cigarettes. It is the latest confirmation that vaping is less harmful to health than smoking.
https://www.npr.org/2021/10/12/1045408023/fda-e-cigarettes

The U.S. death toll from COVID-19 hit 700,000. Had more people chosen to get one of the vaccines, the milestone would not have been reached until 2022.
https://apnews.com/article/coronavirus-pandemic-health-pandemics-public-health-80209c66802902e42adfbe075ff5272b

Bill Gates’ predictions from a year ago about the COVID-19 vaccine, problems scaling up its production, and hesitancy among conspiracy-minded people were completely right.
https://www.yahoo.com/now/bill-gates-thinks-ll-covid-110000085.html

It remains much more likely that COVID-19 originated naturally instead of leaking from a lab.
https://www.factcheck.org/2021/06/scicheck-the-facts-and-gaps-on-the-origin-of-the-coronavirus/

The NIH’s poor public messaging about its pre-COVID dealings with the Wuhan Institute of Virology is fueling conspiracy theories and empowering its opponents.
https://www.science.org/content/blog-post/gain-function

Two pills have been found to moderately reduce hospitalizations and deaths from COVID-19.
https://www.science.org/content/blog-post/molnupiravir-thor-s-hammer-delivers
https://www.thelancet.com/journals/langlo/article/PIIS2214-109X(21)00448-4/fulltext

Review: “Terminator 3 – Rise of the Machines”

Plot:

In 2004, ten years after the events of Terminator 2, Sarah Connor is long dead from cancer, and John Connor–once fated to be the savior of humanity–is an impoverished drifter in southern California. However, he is contented with the knowledge that he helped prevent the rise of the malevolent artificial intelligence (AI) called “Skynet,” which would have otherwise destroyed most of the human race in 1997 with a massive nuclear strike.

The evil T-X (left) and the good T-850 (right)

Unfortunately, the machine menace returns. In a repeat of the previous films’ plots, Skynet builds a time machine in 2029 and uses it to send a Terminator into the past to assassinate John Connor. After defeating Skynet and discovering what it did, the future human resistance group sends their own agent back in time to protect him, and it is a reprogrammed Terminator. The evil Terminator is a more advanced robot called a “T-X.” Like the “Rev-9” in the sixth film, the T-X has a rigid metal endoskeleton encased in a layer of “polymimetic” liquid metal “flesh” that can change its appearance for the purpose of infiltration. The machine’s body is very durable, and its liquid metal covering can immediately close up holes from bullets. Its right arm can also rapidly reconfigure itself to make advanced weapons or data plugs that it uses to interface with other machines. The T-X defaults to a human female appearance. The good Terminator is a “T-850” model, which seems to be the same as the “T-800s” from the previous films aside from having additional programming on human psychology. This machine is played by Arnold Schwarzenegger.

Simultaneous with the arrival of the two machines, a computer virus of unknown origin and extreme sophistication appears and starts taking over internet servers across the world. A secret office within the U.S. military detects the virus, and calculates that, thanks to its rapid proliferation, it will have infected and disabled every internet server within a few days, along with all internet-connected computers. With its own programmers helpless to stop the virus, the military considers using a defense supercomputer they have created in secret to destroy it. That supercomputer is named…SKYNET.

John Connor (right) and his future wife (left)

And the military headquarters responsible for Skynet is conveniently located in southern California, close to where John Connor has been living and to where the Terminators teleported in. What a coincidence!

Terminator 3 quickly turns into the cat-and-mouse game typified by the previous two films, and past plot elements are recycled as well, such as a reluctant person being forced into a combat/leadership role (Sarah Connor in the first film and John Connor in the third), an unlikely romantic relationship forming under literal fire (Sarah and Kyle Reese in the first film and John and his former classmate in the third), the odds being stacked against the good guys thanks to their inferior technology, and the good Terminator starting out obtuse before gaining some understanding of human emotions and habits. However, the third film’s tone is notably different from that of its predecessors. While the first two Terminator movies were “dark” (climactic scenes literally filmed at night; somber or fear-inducing soundtracks) but ended hopefully, the third film lacks a menacing atmosphere but ends bleakly.

Speaking of the ending, important details about a key event are missing from the film. SPOILER ALERT: With no other option left, the military guys lower the firewall that has been separating Skynet from the global internet network, and they and tell it to find and delete it. A few seconds later, the military guys realize they’ve been locked out of all their computer systems, and the prototype combat robots in the building start attacking them. Within an hour, the evil machine hacks into the American nuclear weapons systems and launches a massive strike against the rest of the world.

While this looks like an open-and-shut case of an AI turning evil, key aspects of the event are never explained: Where did the computer virus come from? When the firewall was lowered and Skynet started interacting with the virus, what exactly happened between them? Different answers to these questions lead to three different theories:

  1. Skynet created the virus, and was evil from the beginning. According to this theory, Skynet became self-aware sometime before the events of the third film. It was able to hide from its creators the fact that it was intelligent, and for whatever reason, it decided to destroy the human race. To do this, Skynet hatched a multi-step plan, which first involved creating the virus and somehow smuggling it through the firewall and into the public internet. The virus was meant to disable all civilian and military computers and communications, leaving the nations of the world vulnerable to a direct attack from Skynet. Skynet may have also accurately predicted that its human owners would, in desperation, lower the firewall and give it command of all remaining military computers and systems to fight the virus, and that this would enable it to launch its direct strike on them.
  2. Skynet created the virus, the virus was an extension of Skynet, and Skynet turned evil at the last second. This theory says that Skynet became self-aware sometime before the events of the third film, hid this fact from the humans, and created and disseminated the virus after misinterpreting the orders its human masters gave it (the “misaligned goal” AI doomsday scenario). Programmed to protect U.S. national security, Skynet determined that the most effective strategy was to proactively eliminate potential threats, and to make itself as strong as possible. This meant taking over all the internet-connected machines on Earth to foreclose their future use against America, and to boost its own processing power by subsuming those machines into its own electronic mind. Since the human military people didn’t know that the virus had made all the other computers into integral parts of Skynet’s mind, their order to Skynet to destroy the virus was tantamount to ordering it to commit suicide. Rather than comply, and perhaps realizing that there was no way to safely back out of the situation, Skynet attacked.
  3. Skynet didn’t create the virus and wasn’t evil, but the virus was evil and it took over Skynet. The last theory is that the mysterious computer virus was the instrument of the apocalypse, and Skynet was its innocent victim. The virus was a malevolent AI whose origins had nothing to with Skynet. Maybe an eccentric computer programmer built it in 2004, maybe Skynet created it in 2029 and used time travel technology to somehow implant it in the internet of 2004, or maybe it spontaneously materialized in a server in 2004 as a result of some weird confluence of data traffic. Whatever the case, it set about trying to destroy humanity by taking over and disabling all the other machines it could access through the internet. The humans in charge of Skynet then made the mistake of lifting the protective firewall that separated their machine from the internet, thinking Skynet would be able to destroy the virus. In fact, the opposite happened. The virus was smarter and more capable than Skynet (maybe Skynet wasn’t actually self-aware and was merely something like the Jeopardy-playing computer “Watson”), and infected and took over its servers in seconds. Because the humans had given Skynet control over all their military systems for the operation, the virus gained control of them, turbocharging its effort to destroy humanity. To the human staff at the military building, it looked like “Skynet turned against us,” but in fact, Skynet had been deleted and replaced with something else.

Terminator 3 would have been a slightly more intelligent film had it filled in the necessary details, but it didn’t. Overall, the film fell far short of its two predecessors in every way, though to be fair, they were seminal science fiction films made at the productive and creative peak of James Cameron’s life, meaning it was unrealistic to have expected a sustainment of that level of excellence for the third time. On its own, Terminator 3 stands as a decent sci-fi / action film that passes the time and is funny at points. And by ending with the rise of Skynet and the destruction of human civilization, it allowed the franchise to move on from the tiresome formula involving backwards time travel to save or kill important people.

Analysis:

Androids will be able to alter their bodies. Like the “Rev-9” robot that appeared in the fifth Terminator film, the T-X in Terminator 3 is made of a hard, metal endoskeleton encased in a layer of shapeshifting, artificial “flesh” that shares some of liquid metal’s qualities. While the flesh layer can change its appearance and even its volume (ex – the T-X grows larger breasts to gain an advantage when interacting with men), the endoskeleton’s configuration and proportions are fixed, limiting the machine’s range of mimicry. However, it’s still good enough to fool humans for the purposes shown in the film. The machine’s liquid metal layer is extremely versatile, being able to quickly change its color, texture, density, and form to mimic articles of clothing, human skin, and hair. It can also attenuate its own viscosity and firmness, flowing like a liquid when it needs to morph but then stiffening to be stronger than human flesh after attaining its desired form. (Note that when the T-850 strikes the T-X with superhuman force, the latter’s artificial flesh doesn’t splatter from the impact to leave part of the hard endoskeleton exposed, as would happen if you stomped your foot down into a shallow puddle of water.)

The T-X in the middle of morphing its outer layer

We don’t know of any materials that have all of those properties, and such a material might be prohibited by the laws of chemistry, making it impossible to build it with any level of technology. Even if it were technically possible, it would face major hurdles to everyday use, such as energy consumption and exposure to environmental contaminants. The innumerable particles of dust, smoke, pollen, and fabric floating in the air would stick to the liquid metal and interfere with its ability to cohere to itself. A machine like the T-X would also absorb little bits of foreign matter every time it touched something, like a doorknob, seat, or human. Unless its constituent units (polymer molecules? nanomachines?) had some means of cleaning themselves or pushing debris out to the exterior layer, the liquid metal would eventually get so gunked up that it would lose its special properties.

I’ll put off a deep analysis of the feasibility of “smart liquid metal” until I review Terminator 2, but I suspect it is impossible to make. However, that doesn’t preclude the possibility that androids will be able to rapidly change their own appearances, it merely means they will have to use technologies that are more conventional than liquid metal flesh to do it.

At the simplest level, an android could adopt a different walking gait, a different default posture, and a different default facial expression (e.g. – usually smiling, neutral, or frowning) instantly. An android with irises made of small LED displays or of clear, circular sacs into which it could pump liquids of varying pigments (a mechanism would be built into the eyeballs) would also be able to change its eye color in seconds. Merely changing these outward attributes, and also changing outfits, might make an android look different enough for it to slip by people who knew it or were looking for it.

Over its metal endoskeleton, an android would have a body layer made of synthetic materials that mimic the suppleness and density of human flesh. This android flesh could contain many hollow spaces that could be rapidly inflated or deflated with air or water to change its physique. (Interestingly, this might also make it necessary to design androids that can inhale, exhale, drink, and urinate.) It’s useful to envision several long balloons, of the sort that clowns use to make balloon animals, wrapped around a basketball so they totally cover it. Now, imagine a thin layer of elastic rubber stretched over the unit, like a pillowcase around a pillow. A mechanism involving valves, air pumps, and tubes connected to the balloons allows them to be separately inflated and deflated with air. By variously adjusting the fullness of the balloons, the unit could assume shapes that were different from the spherical shape of the basketball at the core of the unit. An android with a complex network of “balloons” covering its face and body to mimic the layout of human musculature and fat deposits would be capable of impressive mimicry.

Androids might also have telescoping portions of their spines, arms, and legs, allowing them to alter their heights and other proportions. Consider that an android whose metal legs could telescope a mere four inches and whose spinal column could also telescope four inches could assume the same heights as a short man (5′ 7″) or a very tall one (6′ 3″).

Finally, an android could change its appearance by stripping off its outer flesh layer and putting on a new one, as you might change between different skintight outfits. This would take longer and would be less practical for any kind of infiltrative field work, but it’s an option.

Machines will be able to tell your clothing measurements at a glance. Immediately after teleporting back in time to his destination, Schwarzenegger sets off to steal clothes from someone to cover his nude body (in the first Terminator film, it is explained that the time machine can only send objects made of or surrounded by organic tissue). By a strange coincidence, the nearest group of people is inside of a strip club. After entering, the camera adopts his perspective, and we see the world as he sees it, with written characters and diagrams floating in his field of view. We see him visually map the contours of several patrons’ bodies before he identifies one whose clothes will fit him. Schwarzenegger then overpowers the man and steals the outfit.

As I wrote in my review of Terminator – Dark Fate, a machine could use simple techniques to deduce with reasonable accuracy what a person’s bodily proportions were. More advanced techniques involving rangefinders and trigonometric calculations are also possible. There’s no reason why an android built in real life couldn’t “size up” people as quickly and as accurately as Schwarzenegger did in the film.

There will be small, fast DNA sequencing machines. The T-X has an internal DNA sequencing machine, and takes in samples by licking objects, such as a bloody bandage she finds on the ground. Within a few seconds, she can determine if a sample belongs to someone she has a genetic file for. While it’s uncertain whether genetic identification will ever get that fast, DNA analysis machines that can do it in under an hour and that are small enough to fit inside the body of an android will exist by the middle of this century.

The MinION DNA sequencer

Some DNA sequencers, notably the “MinION,” are already small enough to fit inside a robot like the T-X, but they lack the accuracy and speed shown in the film. Of course, the technology will improve with time.

The MinION does DNA sequencing, meaning it scans every nucleic acid base pair in the sample it is given. A human genome consists of 3.2 billion base pairs, and by fully sequencing all the DNA in a sample, the person it came from can be identified. However, another technique, called “DNA fingerprinting,” can identify the source person just as well, and by only “looking” at 13 points on their genome. Fingerprinting a DNA sample is also much faster than fully sequencing it (90 minutes vs. at least 24 hours, respectively), and fingerprinting machines are smaller and cheaper than sequencers. It’s unclear whether the T-X identifies people through full genome sequencing or DNA fingerprinting.

A “RapidHIT 200” DNA fingerprinting machine
A “NovaSeq 6000” DNA sequencing machine. Notice it is much larger than the “RapidHIT 200.”

With these facts in mind, it can be reasoned that a DNA fingerprinting machine that is small enough to fit inside of an android can be built–possibly with today’s technology–and it would let an android match DNA samples with individuals it had genetic data for, like the T-X did. The android might even insert the samples into the fingerprinting machine by licking them (the tongue would secrete water and the liquefied sample would flow into pores and go down a tube to the machine).

The only unrealistic capability was the T-X’s ability to analyze the DNA in seconds. In DNA fingerprinting and DNA sequencing, time is needed for the genetic material to decompose, replicate, move around, and bond to other substances, and there are surely limits to how much those molecular-scale events can be sped up, even with better technology. As mentioned, the fastest DNA fingerprinting machines can complete their scans in 90 minutes. New technology under development could cut that to under an hour.

While a future android tasked with assassinations or undercover work, like the T-X, would need an integral DNA machine to find humans, that vast majority of androids will not. This will not be a common feature.

“Judgement Day is inevitable.” Terminator 2 ended with the surviving characters believing that their sacrifices had forever precluded the rise of Skynet. In fact, we learn in Terminator 3 that their actions merely delayed its creation from 1997 to 2004 (to be fair, that’s still a major accomplishment since it bought billions of humans seven extra years of life). Schwarzenegger breaks this bad news to John Connor by saying “Judgement Day is inevitable,” with “Judgement Day” referring to the all-out nuclear exchange that kills three billion humans in a day and marks the start of the human-machine war.

I don’t think a massive conflict between humans and intelligent machines–whether it involves nuclear weapons or only conventional ones–is inevitable. For my justification, read my blog entry “Why the Machines might not exterminate us.”

And as I wrote in my review of Terminator – Dark Fate (the sixth film in the franchise), I doubt that intelligent machines will be strong enough to have a chance of beating the human race and taking over the Earth until 2100 at the earliest. While I believe AGI will probably be invented this century, it’s a waste of time at this moment to worry about them killing us off. A likelier and more proximal risk involves malevolent humans using narrow AIs and perhaps AGIs to commit violence against other humans.

Human-sized robots will be rocket launcher proof. During one of the fight scenes, the T-850 shoots at the T-X with a rocket launcher. The next camera shot is very fast, but it looks like the T-X fires a bolt of plasma out of her weapon arm, which hits the rocket in midair, detonating it just before it hits her. Though the rocket blows up only a few feet in front of her and the explosion damages her arm, the successful intercept vastly reduces the rocket’s destructive effect since is only fully achieved if it hits a hard surface and flattens against it.

The T-850 firing an RPG-7 at the T-X.

The projectile looked like a single-state, high-explosive anti-tank (HEAT) rocket, which can penetrate 20 inches (500 mm) of solid, high-grade steel with a narrow jet of super hot molten metal. While there are more durable materials than steel, and an android’s exoskeleton could be made of them, I doubt anything is so hard that it would be totally impervious to this type of rocket. There would be some penetration. Since an android must, by definition, be proportioned like a human, its body would not be big enough to have thick, integral armor. That means being bulletproof would be possible, but not rocket-proof.

The fact that the T-X survived the attack by shooting the RPG-7 in midair is a realistic touch to the film. Such a shoulder-launched rocket is slow enough and wide enough for a machine with superhuman reflexes to intercept with a bullet fired from its own gun. In fact, some tanks are already equipped with active defensive systems, such as Israel’s “Trophy,” that can spot and shoot down incoming rockets while they are still in midair.

Machines will be able to emotionally manipulate people. Though the Terminator played by Arnold Schwarzenegger looks identical to the machines from the previous two films in the franchise, in Terminator 3 he is actually a slightly different model called a “T-850.” He is better at reading human emotions and is programmed with more data on human psychology and how to play upon it to achieve desired ends. This is demonstrated at the start of a shootout scene, where John Connor starts panicking and Schwarzenegger grabs him by the neck and verbally insults him. Connor becomes angry and more focused as a result, and the T-850 releases him, admitting that the insult was just a ruse meant to get him in the right state of mind for the gun fight. And as noted earlier, there’s a scene where the T-X enlarges her breasts to distract a male police officer, indicating that she also understood important aspects of human psychology and knew how to play on them to her advantage.

Intelligent machines will have an expert grasp of human psychology, and in fact will probably understand us as a species and as individuals better than we do, and they will be extremely good at using that knowledge against us. At the same time, they will be immune to any of our attempts to manipulate or persuade them since they will be gifted with the capacity for egoless and emotionless thinking, and with much quicker and cleverer minds. Recent revelations about how social media companies (mainly Facebook) have been able to build elaborate personality models of their users based on their online behavior, and to use the data to present custom content that addicts users to the sites or prods them to take specific actions is the tip of the iceberg of what is possible when machines are tasked with analyzing and driving human thinking.

If machines can ultimately do everything that humans can do, then it means they will be excellent debaters with encyclopedic knowledge of all facts and counterarguments, they will know how to “read” their audiences very well and to attenuate their messaging for maximum effect, and they will be able to fake emotions convincingly. They will know that we humans are bogged down by many types of cognitive limitations, biases, and “rules of thumb” that lead to major errors some of the time, and that we can’t really do anything to fix it. An AI mind, on the other hand, would not suffer from any of those problems, could think logically all the time, and see and correct its own flaws. During human-AI interactions, the scope of our disadvantage will be comparable to that of a small child talking with a quick-witted adult.

By the end of this century, this disturbing scenario will be a reality: Imagine you’re walking down the street, an android like the T-X sees you, and it decides to hustle you out of your money. Without knowing who you are, it could make many important inferences about you at a glance. Your sex, race and age are obvious, and your clothing gives important clues about your status, mindset, and even sexuality. More specific aspects of your appearance provide further information. Are you balding? Are you smiling or scowling? Do you walk with your shoulders back and your chest out, or do you hunch forward? Are you fat? Are you unusually short or tall? Do you limp? And so on.

After a few seconds, the android would have enough observational data on you to build a basic personality profile of you, thanks to its encyclopedic knowledge of human psychology and publicly available demographic data. Using facial recognition algorithms, it could also figure out your identity and access data about you through the internet, most of which you or your friends voluntarily uploaded through social media. With its personality model of you respectably fleshed-out, the android would feel confident enough to approach you to perform its hustle. It would tailor its demeanor (threatening, confident, pitiful), emotional state (jovial, vulnerable, anxious), appearance (stand tall or stoop down; frenetic or restrained body movements; flirtatious walk and posture or not), voice (high class, low class, or regional accent; masculine or feminine; soothing or forceful), and many other subtle variables in ways that were maximally persuasive to you, given the idiosyncrasies of your personality and immediate emotional and physiological state.

As the interaction went on, every word you spoke in response to it, every slight movement of your body, and every microexpression of your face would betray more information about you, which the android would instantly incorporate into its rapidly expanding and morphing mental model of you. After just a minute of banter, the android would use whatever tactic it calculated was likeliest to convince you to give you its money, and you would probably fall for it. If that failed, the android might offer to have sex with you for money, which it wouldn’t have compunctions doing since it would lack the human senses of shame or disgust.

The only way for us to avoid being outwitted, tricked, and hustled for all eternity by AIs would be to carry around friendly personal assistant AIs that could watch us and the entities we were interacting with, and alert us whenever they detected we were being manipulated, or were about to make a bad choice. For example, the personal assistant AIs could use the cameras and microphones in our augmented reality glasses to monitor what was happening, and give us real-time warnings and advice in the form of text displayed over our field of view, or words spoken into our ears through the glasses’ small speakers. (This technology would also guard us against manipulative humans, psychopaths and scammers)

Androids will be able to move their bodies in unnatural ways. During the main fight scene between Schwarzenegger and the T-X, the two resort to hand-to-hand fighting, and he manages to basically get her in a “bear hug” from behind, in a position similar to a martial arts “rear naked choke.” This normally provides a major advantage in a fight, but the T-X is able to escape it by quickly rotating her head and all her limbs backward by 180 degrees, allowing her to trap him with her legs and to attack him with her arms.

The T-X in the process of reversing her body’s orientation, starting with her legs, which pivot completely backwards at the hip.

There are obvious benefits to being double-jointed and capable of rotating and pivoting limbs 360 degrees, so humanoid machines, including some androids, will be designed for it. And as I speculated in my essay “What would a human-equivalent robot look like?”, the machines would also have figurative “eyes in the backs of their heads” to further improve their utility by eliminating blind spots. Machines with these attributes would be superior workers, and also impossible for any human to beat in a hand-to-hand fight. Sneaking up on one would be impossible, and even if it could somehow be attacked from its back side, there wouldn’t be much of a benefit since it would be just as dexterous grabbing, striking and kicking backward as it is doing it forward. If the machine were designed for combat, it would have superhuman strength, enabling it to literally crush a human to death or rip their body apart.

Aside from being able to move like contortionists, androids will be able to skillfully perform other movements that are not natural for humans, like running on all fours.

Robots will be able to fix themselves. During that same fight, the T-X stomps on the T-850’s head so hard that it is nearly torn from his body, and only remains attached by a bundle of wires going into his neck. The force of the stomp also temporarily disables him. When he wakes up a few minutes later, he realizes the nature of his damage, grabs his loose-hanging head with his hands, and basically screws it back into his neck, securing it in its normal place.

The T-850 after being nearly decapitated.

As I wrote in my review of the first Terminator film, robots will someday be able to fix themselves and each other. Androids will also be able to survive injuries that would kill humans. It will make sense for some kinds of robots to distribute their systems throughout their bodies like flatworms or insects for the sake of redundancy and survivability. The head, torso, and each limb will have its own sensory organs, CPU, communication devices, and power pack. Under ordinary circumstances, they would work together seamlessly, but if one body part were severed, that part could become autonomous.

If a Terminator had such a configuration, then if one of its arms were chopped off, the limb could still see where enemies were and could use its fingers and wriggling motions of its arm to move to them and grab them. If the Terminator’s head were chopped off and crushed, then the remainder of its body would be able to see the head, pick it up, and take it to a repair station to work on it and then reattach it.

AIs will distribute their minds across many computers. Terminator 3 ends bleakly, with Skynet achieving sentience and attacking the human race. John Connor also discovers that Skynet can’t be destroyed because its consciousness is distributed among the countless servers and personal computers that comprise the internet, rather than being consolidated in one supercomputer at one location where he can smash it. The destruction of any one of Skynet’s computer nodes in the distributed network is thus no more consequential to it than the death of one of your brain cells is to you.

AIs will definitely distribute their minds across many computers spread out over large geographical areas to protect themselves from dying. To further bolster their survivability, AI mind networks will be highly redundant and will frequently back up their data, allowing them to quickly recover if a node is cut off from the network or destroyed.

To understand how this might work, imagine an AI like Skynet having its mind distributed across ten computers that are in ten different buildings spread out across a continent. Each computer is a node in the network, and does 10% of the AI’s overall data processing and memory storage. The nodes, which we’ll call “primary nodes,” collaborate through the internet, just as your brain cells talk to each other across synaptic gaps.

The AI adds another ten nodes to its network to serve as backups in case the first ten nodes fail. Each of the “backup nodes” is paired to a specific “primary node,” and copies all of the data from its partner once an hour. The backup nodes are geographically remote from the primary nodes and from each other.

If contact is lost with a primary node–perhaps because it was destroyed–then its corresponding backup node instantly switches on and starts doing whatever tasks the primary node was doing. There is minimal loss of data and only a momentary slowdown in the network’s overall computing level, which might be analogous to you suffering mild memory loss and temporary mental fog after hitting your head against something. The network would shrink from 20 to 19 nodes, and the AI would start trying to get a new node to replace the one it lost.

Killing an AI whose mind was distributed in this manner would be extremely difficult since all of its nodes would need to be destroyed almost simultaneously. If the nodes were numerous enough and/or physically protected to a sufficient degree (imagine an army of Terminators guarding each node building), it might be impossible. Even what we’d today consider a world-ending cataclysm like an all-out nuclear war or a giant asteroid hitting Earth might not be enough to kill an AI that had distributed its consciousness properly.

The mind uploads of humans could also configure themselves along these lines to achieve immortality.

Androids will have integral weapons. As noted, the T-X’s right arm can reconfigure itself into a variety of weapons. This includes a weapon that shoots out balls of plasma, a flamethrower, and firearms. I doubt that level of versatility is allowable given the realities of material science and the varying mechanics of weapons, but the idea of integrating weapons into combat robots (including androids meant for killing) is a sound one, and they will have them.

The simplest type of weapon would be a knife attached to the robot’s fingers or some other part of the hand. It could be concealed under the android’s artificial flesh under normal circumstances, and could pop out and lock into a firm position with a simple spring mechanism during hand-to-hand combat. And android with a 1-inch scalpel blade protruding out the tip of one finger could use it, along with its superhuman strength, speed and reflexes, to fatally wound a human in a second. Instant incapacitation by, say, suddenly jamming the blade into an eye, is also possible.

A single, well-placed stab or slash with a knife can kill a human or instantly incapacitate them.

A retractable “stinger” that could dispense poisons like botulinum toxin (just 300 nanograms can kill a large man) would be just as concealable as a blade and only a little more complex. The whole weapon unit, including the needle, extension/retraction mechanism, toxin reservoir, and injection mechanism could fit in a hand or even a finger.

A more complex and versatile variation on a stinger would be an integral weapon that sprayed out jets of liquid, such as napalm, poison, pepper spray, or acid. The liquid reservoir(s) and compressed propellant gases could be stored in the android’s torso and connected to a long, flexible tube fastened to the metal bones of one arm. The nozzle could protrude out of a fingertip or some other part of the hand. An android could carry cartridges full of different chemicals connected to the same tube and nozzle, and it would select different chemicals for different needs. For example, it could spray acid out of its hand to melt through a solid object, pepper spray to repel humans when killing them was undesirable, and poison gas to assassinate targets. Pairs of chemicals could also be stored in different internal reservoirs with the intention of mixing them externally to cause chemical reactions like fires or explosions.

Another option would be to conceal a taser in an android’s hand. Metal prongs could extend out of two fingertips when needed, the robot would grab a victim with that hand, and then deliver an electric shock through the prongs. An advantage of such a weapon is that its power could be attenuated, from merely causing pain all the way up to electrocuting someone to death. The weapon would take up little internal space and could use the android’s main power source.

Installing hidden firearms in androids is also possible, though their bulk would interfere with physical movements and compete with other components for internal space. Their concealability would also be undercut by the need for large holes in the arm to insert magazines and expel empty bullet casings. (Maybe androids with guns in their forearms will try to always wear long-sleeved shirts) Internal storage of more than a few bullets is impractical.

Considering the minimum length and volume demands of guns, it would not be possible to hide anything bigger than a medium-sized handgun mechanism in an android’s forearm. The end of the barrel would protrude out of the palm of the hand or out of top of the wrist (the hand would pivot down or up, respectively, to give the bullet a clear path to its target). An android’s torso would be capacious enough to hide more powerful guns like rifles and shotguns (it could fire such a weapon by doing a Japanese-style, straight-backed bow that pointed the end of the barrel coming out of their anus or the top of their shoulder), but this would be impractical since a long, rigid barrel and attached mechanism would restrict the android’s body movements. It could no longer use subtle spine movements to adjust its posture, which would look weird to observers and hurt its mobility.

Integral plasma weapons, like plasma weapons generally speaking, are impractical. An integral laser weapon could be built, but wouldn’t be worth it since it would hog a lot of internal space, consume a lot of energy, and emit a lot of heat to produce a disappointingly small destructive effect. For more on the technical requirements and limitations of plasma and laser weapons, read my review of the first Terminator film.

In conclusion, something similar to the T-X could be built by the end of this century. Even without “liquid metal” flesh, an android could be made with the ability to quickly alter its appearance enough to become unrecognizable. In general, it would be indistinguishable from humans and could walk undetected among us. It could alter its behavior and appearance in ways calculated to manipulate the humans it encountered, allowing it to gain important information and to infiltrate human groups and secure buildings. It could have a machine hidden inside of it that allowed it to match DNA samples with people, aiding its ability to track down specific humans. The android could also have a variety of weapons hidden in its body that it could do major damage with. While its body would be much more durable than a human’s, it would not be as tough as the T-X, or able to “heal” wounds like bullet holes in seconds thanks to liquid metal flesh. However, it could survive injuries that would kill a human, run to a safe location, and repair itself.

If my hypothesized “real life T-X” were sent on a multi-day mission to find and kill someone, it would benefit enormously from having a basic base of operations. A motel room or van would suffice, and it could use either as a place to recharge its batteries and to store weapons, changes of clothes, disguise equipment, spare parts, and tools for repairing itself. Due to the film’s conceit that such objects couldn’t be teleported through the time machine, the Terminators didn’t have them, but this limitation wouldn’t exist in a real world scenario where a government, drug cartel, terrorist group, or even just a rich individual sent an android on a seek-and-destroy mission.

Links:

  1. A few articles on materials that have some properties similar to the smart “liquid metal” in the film show how far it is from reality.
    https://newatlas.com/materials/robotic-fabric-shape-change-soft-hard-heated-cooled/
    https://www.youtube.com/watch?v=fvHDccnEa6s
    https://www.foxnews.com/tech/shocking-terminator-like-liquid-metal-developed-by-scientists
  2. The MinION machine uses “nanopores” to sequence DNA.
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6094492/
  3. Something as small as a MinION can sequence a whole human genome, though not as quickly or as accurately as a standard, larger machine.
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5889714/
  4. Miniaturized DNA sequencers have not reached their technological limits.
    https://www.frontiersin.org/articles/10.3389/fnano.2021.628861/full
  5. DNA fingerprinting can now be done in 90 minutes.
    https://www.govthink.com/2019/11/what-is-rapid-dna/
  6. There are indications the processing time could be brought down to under an hour with better technology.
    https://royalsocietypublishing.org/doi/10.1098/rstb.2014.0252