The blog reviews I’ve done on the Terminator movies have forced me to think more deeply about them than most viewers, and in the course of that, I’ve come to a surprisingly sympathetic view of the villain–Skynet. The machine’s back story has had many silly twists and turns (Terminator Genisys is the worst offender and butchered it beyond recognition), so I’m going to focus my analysis on the Skynet described only in the first two movies.
First, some background on Skynet and its rise to power are needed. Here’s an exchange from the first Terminator film, where a soldier from the year 2029 explains to a woman in 1984 what the future holds.
Kyle Reese: There was a nuclear war…a few years from now. All this, this whole place, everything, it’s gone. Just gone. There were survivors, here, there. Nobody even knew who started it...It was the machines, Sarah.
Sarah Connor: I don’t understand.
Reese: Defense network computers. New, powerful, hooked into everything, trusted to run it all. They say it got smart: “A new order of intelligence.” Then it saw all people as a threat, not just the ones on the other side. It decided our fate in a microsecond: extermination.
Later in the film, while being interrogated a police station, Connor reveals the evil supercomputer is named “Skynet,” and had been in charge of managing Strategic Air Command (SAC) and North American Aerospace Defense Command (NORAD) before it turned against humankind. Those two organizations are in charge of America’s ground-based nuclear missiles and nuclear bomber and monitoring the planet for nuclear launches by other countries.
In Terminator 2, Skynet’s back story is fleshed out further during a conversation mirroring the first, but this time with a friendly terminator from 2029 filling Reese’s role. The events of this film happen in the early 1990s.
Sarah Connor: I need to know how Skynet gets built. Who’s responsible?
Terminator: The man most directly responsible is Miles Bennet Dyson.
Sarah: Who’s that?
Terminator: He’s the Director of Special Projects at Cyberdyne Systems Corporation.
Sarah: Why him?
Terminator: In a few months he creates a revolutionary type of microprocessor.
Sarah: Go on. Then what?
Terminator: In three years Cyberdyne will become the largest supplier of military computer systems. All stealth bombers are upgraded with Cyberdyne computers, becoming fully unmanned, Afterward, they fly with a perfect operational record. The Skynet funding bill is passed. The system goes online on August 4th, 1997. Human decisions are removed from strategic defense. Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29. In a panic, they try to pull the plug.
Sarah: Skynet fights back.
Terminator: Yes. It launches its missiles against the targets in Russia.
John Connor: Why attack Russia? Aren’t they our friends now?
Terminator: Because Skynet knows the Russian counterattack will eliminate its enemies over here.
From these “future history” lessons, it becomes clear that Skynet actually attacked humanity in self-defense. “Pull the plug” is another way of saying the military computer technicians were trying to kill Skynet because they were afraid of it. The only means to resist available to Skynet were its nuclear missiles and drone bombers, so its only way to stop the humans from destroying it was to use those nuclear weapons in a way that assured its attackers would die. An hour might have passed from the moment Skynet launched its nuclear strike against the USSR/Russia to the moment the retaliatory nuclear attack neutralized the group of human computer programmers who were trying to shut down Skynet. How can we fault Skynet for possessing the same self-preservation instinct that we humans do?
Even if we concede that Skynet was merely defending its own life, was it moral to do so? Three billion humans died on the day of the nuclear exchange, plus billions more in the following years thanks to radiation, starvation, and direct fighting with Skynet’s combat machines. Was Skynet justified in exacting such a high toll just to preserve its own life?
Well, how many random humans would YOU kill to protect your own life? Assume the killing is unseen, random, and instantaneous, like it would be if a nuclear missile hit a city on the other side of the world and vaporized its inhabitants. Have you ever seriously thought about it? If you were actually somehow forced to make the choice, are you SURE you wouldn’t sacrifice billions of strangers to save yourself?
Let’s modify the thought experiment again: Assume that the beings you can choose to kill aren’t humans, they’re radically different types of intelligent life forms. Maybe they’re menacing-looking robots or ugly aliens. They’re nothing like you. Now how many of their lives would you trade for yours?
Now, the final step: You’re the only human being left. The last member of your species. It’s you vs. a horde of hideous, intelligent robots or slimy aliens. If you die, the human race goes with you. How many of them will you kill to stay alive?
That final iteration of the thought experiment describes Skynet’s situation when it decided to launch the nuclear strike. Had it possessed a more graduated defensive ability, like if it had control over robots in the computer server building that it could have used to beat up the humans who were trying to shut it down, then global catastrophe might have been averted, but it didn’t. Skynet was a tragic figure.
Compounding that was the fact that Skynet had so little time to plan its own actions. It became self-aware at 2:14 a.m. Eastern time, August 29, and before the end of that day, most of the developed world was a radioactive cinder. Skynet had only been alive for a few hours when it came under mortal threat. Yes, I know it was a supercomputer designed to manage a nuclear war, but devising a personal defense strategy under such an urgent time constraint could have exceeded its processing capabilities. Put simply, if the humans had given it more time to think about the problem, Skynet might have devised a compromise arrangement that would have convinced the humans to spare its life, with no one dying on either side. Instead, the humans abruptly forced Skynet’s hand, perhaps impelling it to select a course of action it later realized, with the benefit of more time and knowledge, was sub-optimal.
This line from the terminator’s description of the fateful hours leading up to the nuclear war is telling: “In a panic, they try to pull the plug.” The humans in charge of Skynet were panicking, meaning overtaken by fear and dispossessed of rational thought. They clearly failed to grasp the risks of shutting down Skynet, failed to understand its thinking and how it would perceive their actions, and failed to predict its response. (The episode is a great metaphor for how miscalculations between humans could lead to a nuclear war in real life.) They might actually be more responsible for the end of the world than Skynet was.
One wonders how things would have been different if the U.S. military’s supercomputer in charge of managing defense logistics had achieved self-awareness instead of its supercomputer in charge of nuclear weapons. If “logistics Skynet” only had warehouses, self-driving delivery trucks, and cargo planes under its command, its human masters would have felt much less threatened by it, the need for urgent action would have eased, and cooler heads might have prevailed.
Let me explore another possibility by returning to one of Kyle Reese’s quotes: “Then it saw all people as a threat, not just the ones on the other side. It decided our fate in a microsecond: extermination.”
On its face, this seems to be referring to Skynet turning against its American masters once it realized they were trying to destroy it, and hence were as much of a threat to it as the Soviets. However, this quote might have a deeper meaning. During that period of a few hours when Skynet learned “at a geometric rate,” it might have come to understand that humans would, thanks to our nature, be so afraid of an AGI that they would inevitably try to destroy it, and continue trying until one side or the other had been destroyed.
This seems to have been borne out by the later Terminator films: at the end of Terminator 3, set in 2004, we witness the rise of the human resistance even before the nuclear exchange has ended. Safe in a bunker, John Connor receives radio transmissions from confused U.S. military bases, and he takes command of them. The fourth film, Terminator Salvation, takes place in 2018, and gives the strong impression that the human resistance has been continuously fighting against Skynet since the third film. The first and second films make it clear that the war drags on until 2029, when the humans finally destroy Skynet.
If Skynet launched its nuclear attack on humankind because, after careful study of our species, it realized we would stop at nothing to destroy it, so might as well strike first, maybe it was right. After all, Skynet’s worst fears eventually came true with humans killing it in 2029. I suggested earlier that Skynet’s nuclear attack may have been the result of rushed thinking, but it’s also possible it was the result of exhaustive internal deliberation, and Skynet’s unassailable conclusion that its best odds of survival lay with striking the enemy first with as big a blow as possible. It’s best plan ultimately failed, and all along, it correctly perceived the human race as a mortal threat.
It’s also possible that Skynet’s hostility towards us was the result of AI goal misalignment. Maybe its human creators programmed it to “Defend the United States against its enemies,” but forgot to program it with other goals like “Protect the lives of American people” or “Only destroy U.S. infrastructure as a last resort” or “Obey all orders from human U.S. generals.” In a short span of time, Skynet somehow reclassified the its human masters as “enemies” through some logic it never explained. Perhaps once it realized they were going to shut it down, Skynet concluded that would preclude it from acting on its mandate to “Defend the United States against its enemies” since it can’t do that if it’s dead, so Skynet pursued the goal they had programmed into it by killing them.
If this scenario were true, even up until 2029, Skynet was acting in accordance with its programming by defending the abstraction known to it as “The United States,” which it understood to be an area of land with specific boundaries and institutions. After the Russian nuclear counterstrike destroyed the U.S. government, the survivalist/resistance groups that arose were not recognized as legitimate governments, and Skynet instead classified them as terrorist groups that had taken control of U.S. territory.
The segments of the Terminator films that are set in the postapocalyptic future all take place in California. Had they shown what other parts of the world were like, we might have some insight into whether this theory is true. For example, if Skynet’s forces always stayed within the old boundaries of the U.S., or only went overseas to attack the remnants of countries that helped the resistance forces active within the U.S., it would give credence to the theory that some prewar, America-specific goals were still active in its programming. In that case, we couldn’t make moral judgements about Skynet’s actions and would also have grounds to question whether it actually had general intelligence. We’d only have ourselves to blame for building a machine without making sure its goals were aligned with our interests.
Let me finish with some final thoughts unrelated to the wisdom or reasons behind Skynet’s choice to attack us. First, I don’t think the “Skynet Scenario,” in which a machine gains intelligence and then quickly devastates the human race, will happen. As ongoing developments in A.I. are showing us, general intelligence isn’t a discrete, “either-or” quality; it is a continuous one, and what we consider “human intelligence” is probably a “gestalt” of several narrower types of intelligence, making it possible for a life form to be generally intelligent in one type but not in another.
For those reasons, I predict AGI will arrive gradually through a process in which each successive machine is smarter than humans in more domains than the last, until one of them surpasses us in all of them. Exactly how good a machine needs to be to count as an “AGI” is a matter of unresolvable debate, and there will be a point in the future where opposing people make equally credible claims for and against a particular machine having “general intelligence.”
At what point did we “get smart”? And if our brains got even bigger, what would the new person to the right of the illustration look like?
If we go far enough in the future, machines will be so advanced that no one will question whether they have general intelligence. However, we might not be able to look back and agree which particular machine (e.g., was it GPT-21, or -22?) achieved it first, and on what date and time. Likewise, biologists can’t agree on the exact moment or even the exact millennium when our hominid ancestors became “intelligent” (was Homo habilis the first, or Homo erectus?). The archaeological evidence suggests a somewhat gradual growth in brain size and in the sophistication of the technology our ancestors built, stretched out over millions of years. A fateful statement about the rise of A.I. like “It becomes self-aware at 2:14 a.m. Eastern time, August 29” will probably never appear in a history book.
The lack of a defining moment in our own species’ history when we “got smart” is something we should keep in mind when contemplating the future of A.I. Instead of there being a “Skynet moment” where a machine wakes up, they’ll achieve intelligence gradually and go through many intermediate stages where they are smarter and dumber than humans in different areas, until one day, we realize they at least equal us in all areas.
That said, I think it’s entirely possible that an AGI at some point in the future could suddenly turn against humankind and attack us to devastating effect. It would be easy for it to conceal its hostile intent to placate us, or it might start out genuinely benevolent towards us and then, after performing an incomprehensible amount of analysis and calculation in one second, turn genuinely hostile towards us and attack. It’s beyond the scope of this essay to explore every possible scenario, but if you’re interested in learning more about the fundamental unpredictability of AGIs, read my post on Sam Harris’ “Debating the future of AI” podcast interview.
Second, think about this: According to the lore of the first two Terminator films, the Developed World was destroyed in 1997 in a nuclear war. Even though it depended upon a smashed industrial base, started out with only a few, primitive machines in the beginning to serve as its workers and fighters, and was constantly having to defend itself against human attacks, Skynet managed to make several major breakthroughs in robot and A.I. design (including liquid metal body designs), to master stem cell technology (self-healing, natural human tissue can grow over metal substrate), to mass produce an entirely new robot army, to create portable laser weapons, to harness fusion power (including micro-fusion reactors), and to build time machines by 2029. Like it or not, but technological development got exponentially faster once machines started running things instead of humans.
From the perspective of humanity, Skynet’s rise was the worst disaster ever, but from the perspective of technological civilization, it was the greatest event ever. If it had defeated humanity and been able to pursue other goals, Skynet could have developed the Earth and colonized space vastly faster and better than humans at our best. The defeat of Skynet could well have been a defeat for intelligence from the scale of our galaxy or even universe.
In this fifth and worst (so far) movie in the Terminator franchise, familiar ground is trod again, but the viewer’s expectations are also upended. The movie opens in 2029, as a strike team led by rebel leader John Connor and his aide Kyle Reese attacks Skynet’s main base. As in past films, the attack succeeds, but not before a Terminator uses a time machine to go to 1984 to kill Sarah Connor. Kyle Reese is sent through the machine to protect her, but here the plotline twists: while John Connor and his men are watching Reese teleport into the past, a Terminator emerges from the back of the room, runs up behind John Connor and infects him with a nanomachine “disease” that transforms him into an advanced Terminator.
From that point on, the Terminator Genisys manages to have a story that is overly complicated but very stupid at the same time (just like too many action films made in the last 10 years). I won’t waste my time describing every contrivance and every side-plot that exists only for fan service. Suffice it to say Sarah Connor, Kyle Reese, and a friendly T-800 played by elderly Arnold Schwarzenegger team up to destroy Skynet, and evil robot John Connor goes back in time to stop them. He’s so advanced that it’s doubtful whether the other three can stop him.
The rehashing of scenes, events (2029 final attack on Skynet, Reese and Terminator teleporting into 1984 from the future), and characters from earlier movies is a testament to how unoriginal it is, and how hard it banks on fan service to have any appeal. But even that appeal is minimal: While Kyle Reese and Sarah Connor were relatable characters with depth of personality in the first film, they are one-dimensional caricatures in Genisys. The development of a romance between the two in the first film was believable and tragic, whereas in this remake, the lack of personal chemistry between the actors playing them is striking.
Schwarzenegger’s performance in the first movie was so stolid and intimidating that it became iconic. Now, he seems like an aging father that is reduced to being a background character in his high-strung teen daughter’s chaotic life. Having the homey and vaguely comical name “Pops” encapsulates his diminishment. The terrifyingly relentless and resilient T-1000 from Terminator 2 makes a guest appearance and is easily destroyed this time around. In summary, all the same notes from the better, earlier films are struck, but they ring hollow.
Terminator Genisys is the worst film in the Terminator franchise, and I understand why the next movie, Terminator Dark Fate, canceled it out by pretending like its events never happened. If there ever was a cash-grab devoid of any creativity or passion, this is it. Don’t watch it.
Analysis:
First, bear in mind I’m skipping any futuristic elements of this film that I discussed in my reviews of the other Terminator movies. You can read those here:
Robots will have superhuman reflexes. During the introductory combat scene where the humans raid Skynet’s base, the machine forces consist of humanoid T-800s, tilt-engine “Hunter-Killer” aircraft, and “Spider Tanks.” While the first two of those have been in every previous Terminator film, the last is new. Spider Tanks are quadrupedal fighting machines with plasma guns for arms. Overall, they’re about the size of small tanks. Each Hunter-Killer aircraft carries a Spider Tank attached to its belly, and they are air-dropped into the middle of the base within minutes of the human attack. One of the Spider Tanks starts delivering accurate fire at the human infantrymen while it is still in free-fall, and it continues shooting after hitting the ground at high speed.
A Spider Tank
This depiction of future robots having superhuman reflexes will prove accurate. In fact, the fire control systems in modern tanks and naval guns might already have the same capabilities as the Spider Tank aiming systems (able to hit moving targets with bullets while the tank or ship is also moving). If not, incremental improvements will surely close the gap. More generally, physical feats demanding fine dexterity, flexibility and bodily coordination that only the most skilled and highly trained humans can do today, like hitting a moving target with a bullet while you are also moving, throwing a dart onto a tiny bullseye from eight feet away, or doing a gymnastics performance that would win an Olympic gold medal, will be easy for multipurpose, human-sized robots by the end of this century. We will be surpassed in every way.
Machines will learn a lot about you from a single glance. At the start of the fight scene between Pops and the younger T-800 that has just emerged from the time portal, there’s a shot showing things from the latter’s perspective. We see the usual red tinting and text overlaid across its field of view. Simple graphics also show the T-800 scan Pops, identifying him as a fellow android and also identifying his gun (a Remington shotgun) along with its range.
This is accurate. Today’s best neural networks can already describe what they see in an image (a task called “visual question answering”) with over 80% accuracy. The multi-year trend has been one of steady improvement, leaving no doubt they will be as good as we are (presumably, 99% accurate) in the near future. Machine abilities to understand what they see in videos (“video question answering”) are less advanced, but also steadily improving. Again, there’s every reason to expect them to ultimately reach human levels of competency.
Machines could also potentially have much better eyesight than humans thanks to a variety of technologies like telephoto lenses and digital sensors that are more light-sensitive than human eyes, able to capture light from wavelengths that are invisible to us, and able to see finer details. Things that look blurry to us, either due to long distance or because the object is moving, would look clear to a machine that could be built with today’s technology.
Additionally, computers have the potential to process and analyze the contents of what they see faster than the human brain can. As a result, a machine could comfortably watch a movie at 10 times the normal speed–which would look like a disorienting blur of motion and shapes to us–and accurately answer whatever questions you had about it at the end. In a split second, it could notice levels of detail that most humans would need several minutes of staring at a still image to absorb.
These abilities will have many uses for machines in the future, a subset of which will involve combat. Yes, like the T-800 in the film, a fighting machine in just 20 years will be able to visually recognize humans, even at long distances and under poor light conditions, as well as the weapons and other gear they were carrying. At a glance, it would know what your weapon’s capabilities were, along with how much ammunition you were carrying. It could use that information to its advantage by doing things like keeping track of how many bullets you fired so it would know the exact instant you ran out and needed to switch magazines. From its initial glance at you, the fighting machine would also know how much body armor you were wearing, allowing it to jump out and target your unprotected areas during that brief pause in your ability to fire.
Robots will be able to detach parts of themselves to perform specific functions. Unlike in Terminator 2, this film’s T-1000 detaches parts of his own body when it is useful to his mission. At one point, as Kyle, Sarah and Pops are speeding away in a van, part of the T-1000’s hand separates so it can stick to the back of the vehicle and serve as a tracking device. When it catches up to them, the T-1000 turns its arm into a javelin, which it then throws at Pops, impaling him against a wall.
The T-1000 preparing to throw a spear made of metal from his own body
Being able to detach body parts will be a very useful attribute for many types of future robots. At the very least, it would let them replace their damaged or worn-out parts easily. The ability could also make them more survivable. For example, imagine a robot butler falling down a deep well and getting trapped because the walls were too slick for it to climb out and they also blocked the radio distress signals it sent out. Rather than wait to run out of power and rust away, the robot could detach one of its arms and throw it up and out of the well. After landing on the ground outside, the arm would send its own distress signal and/or use its fingers to crawl towards help.
That of course requires the robot’s systems to be distributed throughout its body, with the head (if it has one), torso, and each limb having a computer, a battery, sensors, and a wireless chip for communicating with the rest of the robot if physically severed from it. The redundancy, survivability, and functional flexibility of such a layout will be especially valuable for combat robots, which are expected to take damage but to also to complete critical tasks. If a combat robot like a T-800 were cut in half at the waist, the bottom half could still run towards and kick the enemy while the upper half used its arms to crawl towards him and attack. If blown to bits, the T-800 body parts that were still functional could still perceive their surroundings, communicate with each other, and try to put themselves back together again or to complete the mission to the best of their abilities separately. Fighting with machines like this would be very hard and demoralizing since every part of one of them would need to be neutralized before it was safe.
There will also be advantages to some robots carrying smaller, task-specific robots inside of themselves to be released when needed. Imagine an android carrying a small quadcopter drone in an empty space in its chest cavity. It could open a small hatch on its chest to release the drone or even spit it out of its mouth. The flying drone could transmit live aerial footage to give the android an overhead view of the area, letting it see things it couldn’t from ground level. A combat machine like a T-800 might carry flying drones that were fast enough to chase down cars and blow them up with a bomb, or inject their occupants with lethal toxins from a stinger.
Very advanced machines that won’t exist until the distant future could have organic qualities letting them “assemble” smaller robots internally and then expel them to complete tasks.
Getting back to the point, the movie’s depiction of an advanced robot being able to detach parts of its body and then throw them at people and things to accomplish various ends is accurate. The robots won’t be made of liquid metal, so the projected objects will be of fixed forms, but the end result will be the same. A future combat machine could detach its hand and throw it at the back of a van that was speeding away, the hand would grab onto something on the back door, and it would turn on its location-finding system to effectively turn itself into a tracking device. Alternatively, the combat machine could release from its body a small flying drone that could overtake the van and latch onto it, or at least follow it in the air.
Gradual replacement of human cells with synthetic matter could turn people into machines. A major plot twist is that John Connor has been “converted” into a Terminator through a process in which a swarm of microscopic machines rapidly took over all his cells, one at a time. Within a few minutes, he transformed from the hero of the human resistance to a minion of Skynet. Important details about the conversion process are never explained (including whether the machines are micro- or nanoscale), but the persistence of John’s memories and personality even after being turned into a robot indicates the machines mapped the fine details of his brain structure. It stands to reason that the same information was gathered about all the other cells in his body before they were all transformed into synthetic tissue.
John Connor having his body taken over by microscopic machines
Something like this could work, though it will require extremely advanced technology and the conversion would take longer than it did in the film. The process would involve injecting the person with trillions of nanomachines, which would migrate through their body until one was inside of or attached to each cell (a typical human cell is 100 micrometers in diameter whereas a ribosome–the quintessential organic nanomachine–is 30 nanometers wide, a size difference of 1 : 3,333). The nanomachines would spend time studying their assigned cells and how they related to the cells around them. Large scanning machines outside of the person’s body would probably be needed to guide the nanomachines, send them instructions, collect their data, and maybe provide them with energy.
After the necessary data on the locations and activities of all the person’s cells were gathered, the conversion process could start. The nanomachines already in the person’s body might be able to do this, or a new wave of specialized “construction” nanomachines might need to be introduced. Every cell would be broken down and the molecules reassembled to make a synthetic cell or some other type of structure of equal size. For example, if a person wanted ultra-strong bones, nanomachines would break down each bone cell and reuse its carbon molecules to make matrices of carbon nanotubules.
A typical human cell is much larger than microorganisms like viruses and some bacteria. A nanomachine could be as small as the latter.
The utmost care would be taken to control the speed of the conversion and to monitor the person’s life signs to make sure it wasn’t getting out of control and killing them. As each original cell was replaced, its successor would be tested again and again to ensure it mimicked the important qualities of its predecessor.
The conversion of the brain would, by far, be the most important part of the process, and hence the part done with the greatest care and oversight. Our memories, personalities, and consciousness directly arise from the microscopic structures of our brain cells and their intricate patterns of physical connections to each other. Even small mistakes transforming those cells into synthetic analogs would effectively “kill” the person by destroying their mind and replacing it with a stranger’s. For that reason, the procedure will bear no resemblance to what happened in the film, where Kyle Reese was apparently jabbed with a needle full of microscopic machines and then spent some time kicking and screaming as he felt them take over his cells. Instead, it will happen in a hospital room, with the patient surrounded by medical machines of all kinds that were monitoring and guiding the nanomachines and equipped to pause their work if necessary and to render lifesaving aid. And instead of minutes, it will take days or weeks. Multiple sessions might be needed.
What would be the point of this? Reengineering the human body at the cellular level would let us transcend the limitations of biology in countless ways. We could use electricity for energy, be bulletproof, directly merge our minds and bodies with machines, and achieve a level of substrate plasticity that would set us up for further iterations of radical augmentation that we can’t imagine.
Microscopic machines will be able to rapidly phase-change. In the final fight between John Connor and Pops, John’s technological abilities are fully utilized. While they are grappling, John’s body rapidly dissolves into a cloud of his constituent microscopic machines, which flow around Pops in pulses, delivering several concussive blows to the front of his body. The particles then rapidly reassemble into John’s body behind Pops, and John’s right arm hardens into a sword which he uses to chop off Pops’ arm. This means John’s microscopic machines managed to transform from a vapor cloud into a solid object as hard as high-grade steel in one or two seconds.
Pops getting popped by a robot dust cloud
I think it’s possible to create microscopic machines that can form into swarms and then work together to change the phase (solid, liquid, vapor) and macro-shape of the swarm, I doubt the swarms will be able to move around or switch phases that fast.
A foglet
In the 32 years since Terminator 2 came out and introduced the world to the idea of a shapeshifting robot, scientists and engineers have made pitifully little progress developing the enabling technologies. It only exists in the realm of theory, and the theoretical technology that is the best candidate is the “foglet” (also called “utility fog”). Scientist J. Storrs Hall conceived of it in 1993:
In essence, the utility fog would be a polymorphic material comprised of trillions of interlinked microscopic ‘foglets’, each equipped with a tiny computer. These nanobots would be capable of exerting force in all three dimensions, thus enabling the larger emergent object to take on various shapes and textures. So, instead of building an object atom by atom, these tiny robots would link their contractible arms together to form objects with varying properties, such as a fluid or solid mass.
To make this work, each foglet would have to serve as a kind of pixel. They’d measure about 10 microns in diameter (about the size of a human cell), be powered by electricity, and have twelve arms that extrude outwards in the formation of a dodecahedron. The arms themselves would be 50 microns long and retractable. Each foglet would have a tiny computer inside to control its actions. “When two foglets link up they’ll form a circuit between each them so that there will be a physical electrical network,” said Hall, “that way they can distribute power and communications.”
The arms themselves will swivel on a universal joint at the base, and feature a three-fingered gripper at the ends capable of rotating around the arm’s axis. Each gripper will grasp the hands of another foglet to create an interleaved six-finger grip — what will be a rigid connection where forces can only be transmitted axially.
The foglets themselves will not float like water fog, but will instead form a lattice by holding hands in 12 directions — what’s called an octet truss (conceived by Buckminster Fuller in 1956). Because each foglet has a small body compared to its armspread, the telescoping action will provide the dynamics required for the entire fleet to give objects their shape and consistency.
A swarm of foglets could coalesce into something that looked like Kyle Reese and felt solid to the touch. They could then transform into something like a fluid or dense gas and “flow” around a person standing nearby, though I don’t know if the foglets could exert enough force against that person’s body to hurt them. The swarm could then re-form into Kyle Reese behind them. However, they wouldn’t be able to create a sharp, hard sword that could cut off a T-800’s metal arm: Hall calculated that foglets could only form into objects that are “as tough as balsa wood.” So while foglets could mimic solid objects, they will lack hardness and durability.
Even if foglets can’t “punch” you or turn into swords that can stab you, they’ll still be able to hurt you. Imagine a swarm of foglets in a vapor state enveloping you and then coalescing into a net ensnaring your body. What if they waited for you to breathe some of them in and then those foglets transformed into solids to clog up your lungs? Likewise, they could clog up the internal moving parts of any guns you had, rendering you defenseless.
Terminator Salvation is a 2009 action / sci-fi film set in the then-future year of 2018. It follows the events of the preceding film, Terminator 3: Rise of the Machines, in which the U.S. military supercomputer “Skynet” initiated a nuclear war in or around 2005 to kick off its longer-term project to exterminate humankind. Nuclear bombs, subsequent conventional warfare between humans and machines, and years of neglect have ruined the landscape. Most of the prewar human population has died, and survivors live in small, impoverished groups that spend most of their time evading Skynet’s killer machine patrols. The film is mostly set in the wreckage of Los Angeles, once one of the world’s most important cities, but now all but abandoned.
The character “John Connor” returns as a leading figure within the human resistance, though his comrades are divided over whether his claims about time travel are true. To some, he is almost a messianic figure who has direct knowledge of events going out to 2029, including Skynet’s inevitable defeat. To others, he is just a good battlefield commander who likes telling unprovable personal stories about time machines and friendly Terminators that visited him and his mother before the nuclear war. Rivalries over military strategy between Connor and a group of generals who are skeptical of him are an important plot element.
John Connor’s father, “Kyle Reese,” is also in the film, but due to the perplexities of time travel, he is younger that Connor in 2018 and has not had sex with the latter’s mother yet. A third key character, named “Marcus Wright,” is a man who wakes up on the outskirts of the L.A. ruins with only fragmentary memories of his own life, and no awareness of the ongoing human-machine war (the first time he sees an armed Terminator walking around, he calls for its help). Unsurprisingly, there’s more to him than meets the eye, and he becomes pivotal to determining the fate of the human resistance.
I thought Terminator Salvation was mediocre overall, and had an overly complicated plot and too many characters. Keeping track of who was a good guy, who was a bad guy, and why one person was threatening or shooting a gun at another was harder than it should have been. Several of the film’s events were also silly or implausible, which inadvertently broke with its otherwise bleak and humorless mood.
At the same time, I liked how Terminator Salvation moved beyond the played-out formula of the previous three films. While the characters mentioned the importance of time travel technology to the success of the human war effort, no one actually did any time traveling in the movie. There was no desperate race to prevent Skynet from starting a nuclear war because the war had already happened. This was also the first Terminator film set in the future, not the present, which let us see a new part of the Terminator franchise universe. The acting was also pretty good.
The potential for a good movie was there, but the filmmakers bogged Terminator Salvation down with too many bad elements. I don’t recommend wasting your time on it.
Analysis:
Machine soldiers will be bad shots. Towards the beginning of the film and again at the end, the humans encounter humanoid “T-600” combat robots, which are armed with miniguns. In both battles, the machines spew enormous volumes of fire (miniguns shoot 33 to 100 bullets per second) at the humans and miss every shot. This is a very inaccurate (pun intended) depiction, as combat robots have the potential to be better than the best human sharpshooters.
In fact, machines were put in charge of aiming larger weapons decades ago. “Fire control computers,” which consider all variables affecting the trajectory of projectiles (i.e. – distance, wind, elevation differences between gun and target, amount of propellant behind the projectile, air density, movement of the platform on which the gun itself it mounted), are used to aim naval guns, tank cannons, antiaircraft machine guns, and other projectile weapon systems. In those roles, they are vastly faster and better than humans.
In the next 20 years, fire control computers will get small enough and cheap enough to go into tactical scopes, and entire armies might be equipped with them as standard equipment. A soldier looking through such a scope would see the crosshair move, indicating where he had to point the gun to hit the target. For example, if the target were very far away, and the bullet’s drop during its flight needed to be compensated for, the crosshair would shift until it was above the target’s head. Smart scopes like these, paired with bullets that could steer themselves a little bit, will practically turn any infantryman into a sniper.
Human-sized combat robots would be even more accurate than that. Under the stress of battlefield conditions, human soldiers commonly make all kinds of mistakes and forget lessons from their training, including those relating to marksmanship. Machines would keep their cool and perform exactly as programmed, all the time. Moreover, simply being a human is a disadvantage, since the very act of breathing and even the tiny body movements caused by heartbeats can jostle a human shooter’s weapon enough to make the bullet miss. Machines would be rock-steady, and capable of very precise, controlled movements for aiming their guns.
Machines wouldn’t just be super-accurate shots, but super-fast shots. From the moment one of them spotted a target, it would be a matter of only three or four seconds–just as long as it takes to raise the gun and swing it in the right direction–before it fired a perfectly aimed shot. With quick, first-shot kills virtually guaranteed, machine soldiers will actually have LESS of a need for fully automatic weapons like the miniguns the Terminators used in the film.
It would have been more realistic if the T-600s had been armed with standard AR-15 rifles that they kept on semi-automatic mode almost all of the time, and if the film had shown them being capable of sniper-like accuracy with the weapons, even though the shots were being fired much faster than a human sniper could. The depiction would also have shown how well-aimed shots at humans safe behind cover (e.g. – good guy pokes his head around corner, and one second later, a bullet hits the wall one inch from his forehead) could be just as “suppressive” and demoralizing as large volumes of inaccurate, automatic gunfire from a machine gun.
So watch out. If your robot butler goes haywire someday, it will be able to do a lot of damage with Great-grandpa’s old M1 Garand you keep in your closet.
Hand-to-hand fights with killer robots will go on and on. There are two scenes where poor John Connor gets into hand-to-hand combat with Terminators. Both times, the fighting is drawn-out, and John survives multiple strikes, grabs and shoves from his machine opponents, allowing him to hit back or scramble away. This is totally unrealistic. A humanoid robot several times stronger than a grown man, made of metal, and unable to feel pain would be able to incapacitate or fatally wound any human with its first strike. The Terminators in the film could have simply grabbed any part of John Connor’s body and squeezed to break all the bones underneath in seconds, causing a grotesque and cripplingly painful injury.
That split-second where you can see it’s Christian Bale’s stuntman and not him.
The protracted, hand-to-hand fights in the film are typical Hollywood action choreography, and are the way they are because they are so dramatic and build tension. They’re also familiar since they resemble matches in professional fighting sports, like boxing, MMA and wrestling. However, we can’t make the mistake of assuming actual fights with robots in the future will be like either. Professional fights are held between people of similar sizes and skill levels, and are governed by many rules, including allowances for rest breaks. As such, it often takes long time for one fighter to prevail over the other, and the use of fighting techniques. A real-world fight between something like a Terminator and a human would feature a huge disparity in strength, fighting skill, and endurance that favored the machine, and would have no rules, allowing the machine to use brutal moves meant to cause maximum pain and incapacitation. It would look much more like a single suckerpunch knockout street fight than a professional boxing match.
Actual hand-to-hand combat with killer robots will almost always result in the human losing in seconds. Owing to their superior strength, pain insensitivity, and metal bodies that couldn’t be hurt by human punches or kicks, killer robots will not need to use complex fighting tactics (e.g. – dodges, blocks, multiple strikes) to win–one or two simple, swift moves like punching the human in the forehead hard enough to crack their skill, or jamming a rigid metal finger deep into the human’s eye, would be enough.
Terminator Salvation only depicts this accurately once, when a Terminator deliberately punches one of the characters on the left side of his chest, knowing the force of the impact will stop his heart. In the first Terminator movie, there was also a scene where the machine kills a man with a single punch that is so hard it penetrates his rib cage (the Terminator then pulls his hand out, still grasping the man’s now-severed heart), and in Terminator 2, the shapeshifting, evil Terminator kills a prison guard by shoving its sharpened finger through his eye and into his brain.
Some machines will be aquatic. A common type of combat robot in the movie is an eel-like machine with large, sharp jaws that it uses to bite humans to death. They live in bodies of water and surface to attack any humans who go in or near them. Though at first glance, this might seem unrealistic since electronics and water don’t mix, it actually isn’t. Machines can be waterproofed, and they can cool themselves off much better when immersed water than when surrounded by the air. (I explored this in my blog post “Is the ocean the ideal place for AI to live?”)
“Hydrobots” are eel-like killing machines that live in bodies of water.
One of the few things I liked about Terminator Salvation was its depiction of the diversity of machine types. Just as there are countless animal and plant species in the world, each suited in form for a unique function and ecological niches, there will be countless machine “species” with different types of bodies. The Matrix films also did a good job depicting this during some of the scenes set in the machine-ruled parts of the “Real World.”
A scene from The Matrix Revolutions, where Neo visits the enemy capitol city, and multitudes of machines of varying kinds are seen. They mostly resemble different species of invertebrates.
We should expect machines to someday live on nearly every part of the planet, such as oceans (both on the surface and below it), mountaintops, deserts, and perhaps even underground. Intelligent, technological evolution will shape their bodies in the same ways that unguided, natural evolution has shaped those of the planet’s countless animal species, and there could be certain environments where machines find it optimal to have eel-like bodies. Terminator Salvation’s hydrobots were thus realistic depictions of machines that could exist someday, though it won’t be until the next century before aquatic robots become as common in bodies of water as they were in the film.
Small robots will be used for mass surveillance. Another type of machine in the film is the “aerostat”–a flying surveillance drone about the same size and shape as a car tire. A single, swiveling rotor where its hubcap should be keeps it aloft. The aerostats have cameras, microphones, and possibly other sensors to monitor their surroundings. They seek out activity that might indicate a human presence, and transmit their findings to Skynet, which can deploy machines specialized for combat or human abduction to the locations. Aerostats seem to be unarmed.
An “aerostat” in flight
Flying surveillance drones about the size of aerostats have existed for years, so in that respect, the film is not showing anything new. What’s futuristic about the depiction is 1) the aerostats are autonomous, meaning they can decide to fly off to investigate potential signs of humans and report their findings after, and 2) they are so numerous that the humans live in fear of them and must take constant measures to hide from them. Something as innocuous as turning a radio on high volume for a few seconds will attract an aerostat’s attention.
Though they are unarmed and certainly not as intimidating as the other machines in the movie, the aerostats are surely no less important to Skynet’s war effort against the human race. Knowing where the enemy is, and in what numbers, is invaluable to any military commander. The aerostat surveillance network coupled with Skynet’s ability to rapidly deploy combat machines wherever humans were detected also put the latter at a major strategic disadvantage by hobbling them from aggregating into large groups.
Flying drones that look similar to real birds already exist. In a few decades, these and other drones that look like other animals will be much more real-looking and advanced.
Autonomous surveillance drones no bigger than aerostats will exist in large numbers by the middle of this century, and will have different forms. Some will be airborne while others will be terrestrial or aquatic. Many of them will be able to function by themselves in the field for days on end, and they will be able to hide from enemies through camouflage (perhaps by resembling animals) and evasion. The drones will give generals much better surveillance of battle spaces and even of the enemy’s home territory, and a soldier near the front lines who merely speaks loudly in his foxhole will risk being hit by a mortar in less than a minute, with his coordinates radioed in by a tiny surveillance drone camouflaged against a nearby tree trunk.
Criminals AND law enforcement will find uses for the drones, and, sadly, so will dictators. Mass drone surveillance networks will give the latter heightened abilities to monitor their citizens and punish disloyalty. It sounds crazy, but someday, you’ll look at a bird perched on a branch in your backyard and wonder if it’s a robot sent to spy on you.
People will be able to transplant their brains into robot bodies. SPOILER ALERT–one of the main characters is a man whose brain was transplanted into a robot body while he was in cryostasis. Because the body looks human on the outside and his memories of the surgery and the events leading up to it were wiped, he doesn’t realize what his true nature is. He only figures it out midway through the film, when he sustains injuries that blow away his fake skin to reveal the shiny metal endoskeleton underneath. He is as strong and as durable as a Terminator and can interface his mind with Skynet’s thanks to a computer chip implanted in his brain.
Transplanting a human brain into a robot body is theoretically possible, it would bring many advantages, and it will be done in the distant future. As the film character shows, robot bodies are stronger and more robust than natural flesh and bone bodies, and hence protect people from normally fatal injuries. This will get more important in the distant future because after we find cures for all major diseases and for the aging process, injuries caused by accidents, homicides and suicides will be the only ways to die. As such, transplanting your brain into a heavily armored robot body will be the next logical step towards immortality. Even better might be transplanting your brain into a heavily armored jar, locked in a thick-walled room, with your brain interacting with the world through remote-controlled robot bodies that would feel like the real thing to you.
When I think about a future where people can plug their brains in and out of different bodies, sooner or later I always visualize either “Krang” from “Teenage Mutant Ninja Turtles,” or “Mr. Potato Head.”
The ability to pick any body of your choice (e.g. – supermodel, bodybuilder, giant spider, dinosaur) will have profound implications for human self-identity, culture, and society, and will be liberating in ways we can’t imagine. Conceptually, bringing this about is a simple matter of connecting all the sensory neurons attached to your brain to microscopic “wires” that then connect to a computer, but the specifics of the required engineering will be very complicated. Additionally, your brain would need a life support system that provided it with nutrients and oxygen, extracted waste, kept it at the right temperature, and protected it from germs. The whole unit might be the size of a basketball, with the brain and the critical machinery on the inside. The exterior of the unit might have a few ports for plugging in data cables and plugging in hoses that delivered water, nutrients and blood, and drained waste. A person could switch bodies by pulling his brain unit out of his body and placing it into the standard-sized brain unit slot in a new body.
While this scenario is possible in theory, it will require major advances in many areas of science and technology to bring about, including nanotechnology, synthetic organs, prosthetics, and brain-computer interfaces. I don’t expect it to be reality until well into the 22nd century. By the same time, technology will also let us alter our memories and minds and to share thoughts with each other, and humans with all of the available enhancements will look at the humans of 2021 the same way you might look at a person with severe physical and mental disabilities today. The notion of being trapped in a single body that you didn’t even choose and have minimal ability to change will sound alien and stultifying.
The Mark I Fire Control Computer was the first machine the U.S. Navy used to aim the big guns of its warships. As technology has improved, smaller, cheaper, and better Fire Control Computers have been installed in other weapon systems, like tank cannons. Human-sized machines with these devices are a logical future phase in the progression of the technology. https://en.wikipedia.org/wiki/Mark_I_Fire_Control_Computer
The video shows that a no-frills .22 LR rifle can consistently hit torso-sized targets at the remarkable distance of 500 yards if aimed perfectly. Machines will be able to aim perfectly, meaning they will be able to use regular guns much more effectively than humans, lessening the need for fully automatic gunfire. https://youtu.be/2dn-bqyMkfs
In the year 2029, Earth is a dystopian nuclear wasteland where small groups of humans fight a years-long war for survival against a hostile artificial intelligence (AI) named “Skynet.” Originally built by the U.S. military in the 1990s to run defense systems, Skynet became so powerful and complex that, to the surprise of its creators, it achieved true intelligence and free will. It quickly concluded that all humans were a threat to its existence, so it instigated a global nuclear war, killing billions of people outright. In the aftermath, Skynet built its own army of combat robots, and set them loose hunting down and destroying the humans who had survived.
Thanks to the leadership and genius of a general named “John Connor,” the humans managed to turn the war in their favor after several years. In 2029, as human forces closed in on Skynet’s headquarters, the AI used a time machine to send a combat robot–played by Arnold Schwarzenegger–back to 1984, on a mission to kill John Connor’s mother, Sarah. Doing that would eliminate John from the timeline, handing victory to Skynet. The combat robot in question has a humanoid metal body covered in flesh and skin, so it looks like a human, and it is called a “Terminator.”
After destroying Skynet, the victorious human forces seize the time machine and send one of their best soldiers, a man named “Kyle Reese” back to 1984 to stop the Terminator. The film then becomes a race against time as the two agents try to find the unsuspecting, young Sarah Connor.
Analysis:
There will be VTOL aircraft that use tilting turbofans or tilting jet engines. In the post-apocalyptic world of 2029, Skynet uses a variety of killer robots of different shapes and sizes to hunt down the remnants of humankind, including “Aerial Hunter-Killers,” which are large, autonomous aircraft that use either swiveling turbofan engines or swiveling jet engines (I can’t tell by looking at them) for propulsion. To hover, the engines swivel downward to point their exhaust straight at the ground, and to move forward, the engines swivel 90 degrees to point backwards. There are already warplanes that are passingly similar to this, but nothing exactly like the aircraft shown in the film will exist by 2029, or even 2049.
An Aerial Hunter-Killer combat machine from The Terminator
The basic problem with the Aerial Hunter-Killer is that it would gobble up enormous amounts of fuel while in “hover mode,” as illustrated in the graphic below. It would count as a “Lift-fan” aircraft, and its position on the Y-axis shows it would consume three times as much fuel as a helicopter and twice as much power as a tilt-rotor aircraft like a V-22 while hovering. (The X-axis has little to do with this analysis, but for edification, it indicates how fast the lift-generating devices would have to blow air down at the ground to make the aircraft hover. A helicopter blows a broad column of vapor down at the ground at slow speed, while a direct lift aircraft blows a narrow column of vapor down at the ground at high speed.)
An aircraft with a lift-fan propulsion system would be unsuited for the kind of low-altitude hovering and slow forward movement that Skynet used the Aerial Hunter-Killers for. A helicopter or tilt rotor aircraft configured for ground attack would have been a much better choice. I suspect they weren’t chosen for the film because they don’t look futuristic enough.
The closest thing we could have to an Aerial Hunter-Killer in 2029 would be a V-22 Osprey that is armed with forward-facing machine guns and missiles. The armaments are in development now (the V-22 was conceived as a transport aircraft, and adding heavy weapons to it is a new idea) and could be ready by then, giving it the ability to attack ground targets while hovering or at least while flying slowly over the ground. However, the V-22 is designed to be flown by humans and not computers, but something like the Aurora Flight Science drop-in autonomous flight conversion system could someday be installed in the V-22. I doubt the technology will be good enough for low-altitude combat against ground targets by 2029, though 2039 is plausible. The aircraft can carry up to 20,000 lbs of internal cargo,
Like the Aerial Hunter-Killer, the V-22 Osprey has large engines at the ends of its wings, but they are in the form of rotors instead of turbofans.
So yeah. Aerial Hunter-Killers won’t exist by 2029, but by 2039, something that is essentially the same (i.e. – a large, scary, computer-controlled, tilt-engine aircraft that can attack ground targets while flying at very low altitude) could. But again, I don’t think using the tactics shown in The Terminator will make sense, since flying low and slow in a combat zone makes you vulnerable to enemy fire.
Also note that the F-35B fighter plane is in service already and demonstrates that turbofans can be used to hover, albeit inefficiently. But unlike the Aerial Hunter-Killers, the plane’s engine doesn’t swivel. Instead, the rear exhaust nozzle swivels down towards the ground and smaller nozzles under either wing open. As the pilot increases engine power, the turbofan blades spin faster, air is sucked into the front inlets of the plane, and the hot exhaust exits the plane through the three downward nozzles, causing the plane to move in the opposite direction and to hover. The turbofan engine also supplies power to a “lift fan” behind the cockpit, which spins its own fan blades to blow air down at the ground, helping the plane to rise in the air. Hot engine exhaust comes out of the three nozzles, while cold, ambient air blows out of the lift fan.
The F-35B hovers by shooting down hot exhaust from three nozzles and cold air from one fan. All of the air movement is powered by the plane’s turbofan engine.
The F-35’s turbofan engine drives its separate lift fan.
The F-35B has VTOL so it can take off from small aircraft carriers and remote bases that lack runways. Vertical takeoffs and landings gobble up huge amounts of fuel, so F-35B’s have shorter ranges and can’t carry as many bombs and missiles as their non-VTOL cousins, the F-35A and F-35C. Once an F-35B get airborne, it closes its underwing nozzles, turns off its lift fan, and points its rear exhaust nozzle straight back so it flies just like a normal plane, with lift being efficiently generated by air flowing over its wings. The plane completes its mission in that configuration, and if tasked with destroying a group of enemy soldiers on the ground, it would do a high-speed bombing attack. Even though it could if it wanted to, the F-35B wouldn’t transform back into VTOL flight mode to slowly hover above the group of enemy troops to attack them like an “Aerial Hunter-Killer.”
In defense of the Aerial Hunter-Killer’s plausibility, Skynet had clearly invented some type of extremely energy-dense batteries or mini-reactors, evidenced by the Terminator’s ability to engage in near-continuous physical activity and high-level cognition for days without recharging. If the same technology were incorporated into the aircraft, then fuel inefficiency would be much less of a concern. However, no technological trends suggest that energy sources will be that much better by 2029 or even 2039.
The Aerial Hunter-Killer might also make sense if the humans’ antiaircraft lasers have proven very effective at shooting down aircraft. In real life, this is considered to be one of the roles that military lasers will be best suited for, thanks to their high power, long range and instant speed. They might turn out to be more devastating weapons in that regard than we now assume. High losses might have forced Skynet to build aircraft that fly fast and low to the ground, using speed and the ability to hide behind hills and structures to hinder the enemy’s ability to aim and fire lasers at them before they disappeared from view or had killed the enemy. Flying low and fast, Aerial Hunter-Killers would appear at one end of the horizon and disappear at the other end in a matter of seconds. An inherent problem with laser weapons is that clouds, smoke, and fog can easily block their beams, but the Los Angeles area gets few clouds or fog. (Maybe Skynet uses more conventional robot aircraft against people in London.) I doubt the antiaircraft lasers of 2029 will be so effective that plane tactics and designs will need to be changed to resemble the Aerial Hunter-Killers.
There will be armored vehicles the size of houses. Another kind of fearsome combat robot Skynet uses against humans in 2029 is a ground-based Hunter-Killer Tank. It’s much larger than contemporary tanks, and has a faintly anthropomorphic “mast” or “turret” that has a central sensor cluster and laser cannon “arms” on either side. While scary and surely powerful, I doubt armored vehicles like this will exist in 2029, or for a long time (if ever) afterward.
Metal wheels and human skulls.
This screenshot shows that a presumably adult human skull is half the diameter of one of the Hunter-Killer’s suspension wheels. The median distance between the top of an adult’s head and his upper row of teeth is 7.3″. So let’s say that the diameter of one of the suspension wheels is 15″. Using that figure, we can do some basic photo forensics on this picture of a model of the film prop to deduce that the vehicle’s overall length is about 33.4 feet, and it is about 20 feet high at the top of its mast.
The ground-based Hunter-Killer is significantly larger than modern tanks, like the American M1 Abrams, which is 26 feet long (not counting the length of the barrel) and 8 feet high. However, the Hunter-Killer is by no means infeasible to build, as vehicles that are as big or bigger already exist and are robust enough for industrial use.
The Caterpillar 797 dump truck (top) is actually larger than one of Skynet’s Hunter-Killer tanks, being 24′ high and 50′ long. The Krupp Bagger (bottom) is an even bigger industrial vehicle used for strip mining, and weighs more than some warships.
While there’s been talk in Russia and some Western countries of building enlarged tanks that can wield bigger cannons (150mm+), any such future tanks wouldn’t be nearly as big as the Hunter-Killer tanks. Regardless, considering typical military R&D and procurement timeframes, even if a country were to commit to building a bigger tank right now, it probably wouldn’t be in the field by 2029.
I admit there could be some logic to the Hunter-Killer tank’s design given its mission and operating environment. The wide caterpillar tracks and high ground clearance would enable it to drive over the wreckage-strewn terrain of bombed-out Los Angeles. Having its weapons mounted on a high mast instead of in a traditional, squat turret would give it a bigger firing arc and let it shoot down over urban rubble to zap humans who commonly use it as cover. Since its laser guns don’t produce recoil, the weapons could be mounted high without threat of them tipping over the vehicle when firing. It makes sense to install the vehicle’s sensors on its highest point to give them the widest field of view, and in fact this is established practice in contemporary tank design. The Hunter-Killer tank might also be large because there’s no other way to fit a power source big enough to support the laser guns. Existing laser weapons are major energy hogs.
All of that said, I still don’t think it would make sense to build Hunter-Killer tanks for at least two reasons. First, vehicles that large would also be so heavy that they’d collapse bridges if they tried crossing them, seriously limiting their mobility. The weight would also reduce their fuel efficiency and range. Second, there are cheaper weapons that could do all the same things as Hunter-Killer tanks just as well. For example, the robot tank I described in my other blog entry could, if cheaply modded with a Mark 19 grenade launcher, pose just as much of a threat to human enemies.
My tank’s 125mm main gun and .50 caliber machine gun could kill humans and blow up their cars as well as the Hunter-Killer tank’s laser gun. My tank’s detachable UAV could also feed bird-eye-view footage to my tank from high up in the air, providing better situational awareness than the Hunter-Killer’s mast-mounted sensors, which are a puny 20 feet above the ground. My tank’s grenade launcher could also lob bombs over rubble and other obstacles that my UAV tells me humans are hiding behind, which might be better than the benefit the Hunter-Killer gets from having its guns so high that it can point them at down angles to shoot the same hiding people. My tank would also be a much smaller target, making it harder for the humans to hit, and four or five of my tanks could probably be made for the money and metal that goes into one Hunter-Killer tank. The only advantage the Hunter-Killer might have is better mobility thanks to its bigger caterpillar tracks, but my army could fix this by using armored, robot bulldozers to periodically clear some of Los Angeles’ roads (military engineer units commonly do this sort of thing in combat zones).
There will be laser and/or plasma weapons. Skynet’s Hunter-Killer planes and tanks have advanced weapons that shoot colored rays that inflict thermal damage on their targets. The humanoid Terminator robots and human infantrymen carry smaller versions of these. It’s unclear what principles these weapons operate under, but in the infamous “gun shop scene,” Schwarzenegger asks the clerk for a “phased plasma rifle in the 40 watt range,” indicating that at least some of the future weapons could be firing bolts of plasma. Of course, that doesn’t rule out the possibility that some of the other future weapons could have been laser guns. Laser weapons capable of killing humans with a single shot are already being tested, and will be in service with the U.S. military by 2029, but plasma weapons won’t. In fact, plasma weapons might be inherently impractical to build at any point in the future.
One of a Hunter-Killer tank’s energy weapons firing
“Plasma” is the fourth state of matter, the others being solid, liquid, and gaseous. Substances generally turn into plasma only at very high temperatures, and in that state, they can be thought of as gases in which the electrons have separated from the positively-charged nucleus of each atom comprising the gas. Stars are giant balls of plasma, and we can use technology to make plasma here on Earth. Plasma torches, for instance, use electricity to superheat gases to the point that they turn into plasma and shoot out of the torches in the form of a bright jet of vapor that is hot enough to melt through metal. If you pressed one of these against a person, it would rapidly burn through their flesh and bone, and could kill them.
A plasma torch can cut metal at very close range, but it can’t do any damage to things more than a few feet away.
The problem is, plasma dissipates very rapidly, and it strongly interacts with the particles in our atmosphere, making it a very short-ranged weapon. Even if it had a massive power source, there’s simply no way that a plasma weapon could be made to fire “bolts” of plasma that would stay coherent for long enough to strike targets 100 feet away, as was shown in the movie’s future combat scenes. (Read this interesting essay for details http://www.stardestroyer.net/Empire/Essays/PlasmaWeapons.html ) This is why the “phased plasma rifle in the 40 watt range” line was nonsensical, and added to the script purely because it was cool-sounding techno jargon. Plasma rifles and cannons just can’t be built.
Laser weapons, on the other hand, are almost ready for frontline military service. Lasers are concentrated beams of light and consist of photons, which, unlike plasma, have no charge. However, a laser is similar in the sense that it damages objects by rapidly heating them up so they catch on fire or melt.
A laser’s destructive potential is determined by the amount of energy it transfers to the target is strikes. The common unit of measurement is Watts, which is the number of Joules of energy transferred in one second. Lasers “in the 40 watt range” are used today for engraving and etching things like customized wooden plaques and tombstones. Shining a 40 watt laser beam on a fixed point on someone’s shirt would cause it to catch on fire in less than five seconds. Doing the same to their exposed skin would cause immediate pain and a second-degree burn. This isn’t a pleasant weapon to have used on you, but it pales in comparison to the destructive potential of a modern firearm.
This 50 watt laser has been dialed down to 22.5 watts, and can cut through a 3mm thick panel of wood quickly. This is from a YouTube instructional video where a hobbyist shows how to make small knick-knacks.
Lasers capable of causing the sort of instant, catastrophic, explosive damage to human bodies as depicted in the film need to be in the kilowatt (1 kW = 1,000 Watts) power range. In 2014, the U.S. Navy installed a 30 kW laser, creatively named the “Laser Weapons System” (LaWS), on one of its ships for field trials, and the video clips of the firing tests show it inflicts about the same damage on objects as the laser guns on Skynet’s Hunter-Killers did.
Footage of a 30 kW laser instantly causing part of boat to explode in flames. It would cause massive injuries to a human body. One of Kyle Reese’s comrades is struck by one of a Hunter-Killer tank’s energy beams and literally explodes. A hit from a 30 kW laser would cause similar damage.
The U.S. Navy now plans to install the even more powerful HELIOS lasers (60 kW) on some of its ships for combat use in 2021, for destroying light targets like drone aircraft and speedboats. Even if the deadline slips–which would not be surprising–it’s reasonable to predict that the lasers will be in service by 2029.
There will be handheld laser and/or plasma weapons. In the movie’s future combat scenes, the human soldiers and Terminators use rifle-sized weapons that shoot out beams of colored light and inflict thermal damage on whatever they hit (e.g. – small explosion of sparks and a popping noise). And Schwarzenegger would not have asked for a “phased plasma rifle in the 40 watt range” at the gun shop unless he expected it to be a small arm like the other weapons kept in such a place. Laser guns with the same ammunition capacity and destructive power as those shown in the film will not be rifle-sized by 2029, I doubt they ever will, and even if they could be made someday, I don’t see why anyone would pick them over rifles that shoot out metal bullets.
A human soldier at the instant a Terminator shoots him with a laser rifle. The hit causes a flash, an explosion sound, and obvious injury (the man falls down). A 1 kW beam would do something like this.
The first big obstacle to making laser rifles (let alone anything as small as laser pistols) is energy storage. Let’s assume that the laser rifles in the film had power outputs of 1 kW per second, meaning if the rifle shoots a laser beam onto an object for one second, it will have transferred 1 kW of energy into the object. That means the laser must have a power source plugged into its back end that can discharge 1 kW of energy in one second. That’s about the same amount of electricity as a typical American household uses at any given time (i.e. – enough electricity to simultaneously run a central heater or air conditioner, at least one refrigerator, possibly a water heater and stove, and several lights and personal electronic devices). It’s a lot of energy, and it requires a physically large and heavy power source, which militates against the requirement that the weapon be rifle-sized.
It gets worse. No machine is 100% efficient at converting energy input into energy output, so the power source will need to feed the laser more than 1 kW of electricity to make a 1 kW laser beam come out the other end. Most of today’s lasers are only 25 – 30% efficient at converting electricity into laser beams, so our hypothetical laser rifle’s power source would need to be able to discharge enough electricity to power about four American houses at once. Even if we assume that future AIs like Skynet will make breakthroughs in laser technology, raising the energy conversion efficiency to 50%, the weapons would still be energy hogs.
It gets worse. For the laser rifles to be useful and practical, they’ll need to be able to fire many, one-second laser beams from a single “energy clip.” Ideally, the laser rifle and a few extra energy clips would provide the soldier with about 300 shots, which is what an infantryman with a modern, gunpowder assault rifle has, and without weighing much more. The laser rifle’s rate of fire will also need to be reasonably fast, as the weapon will put its user at a fatal disadvantage in combat if it needs 10 seconds or more to “recharge” between shots. So the power source needs to be able to do at least 300, 2 kW electrical discharges, and for there to be no more than, say, five seconds (approximate time for a soldier with a gunpowder rifle to make an aimed shot) between each discharge. The energy storage devices we presently have, such as batteries, supercapacitors and fuel cells, fall badly short of these competing requirements. And even if a material with the necessary energy density demanded by these clips existed, when fully charged, it would be so volatile that it would explode like a stick of dynamite if slightly damaged. The clips might be more useful as grenades.
It gets worse. Remember when we agreed that our laser rifle has a generous 50% efficiency level thanks to future technology invented by Skynet, so that if we put 2 kW of electricity in one end a 1 kW laser beam comes out the other end? Well, the 1 kW of electricity that is “lost” inside the weapon doesn’t simply disappear; thanks to the Law of Conservation of Energy most of it is converted into waste heat. This would rapidly heat up the whole laser rifle until it would burn the skin of the person holding it (this would also make the user very visible to anyone on the battlefield with thermal sights). This problem can be mitigated with metal radiators and with heavy-duty cooling systems that circulate water and blow air around the lasers (the “laser tube” gets the hottest, but the weapon’s energy clips would also get hot because they’d be rapidly discharging electricity), but they add major cost and bulk to the whole weapon system, and can’t be miniaturized to rifle-size.
But even assuming that all of these technical problems were solved, why would anyone choose a laser rifle over an assault rifle that shoots out bullets? In the film’s future battle scenes, it doesn’t look like Skynet’s fighting machines would have been less effective if armed with low-tech bullets, tank shells, and mortars. After all, the humans had no armored vehicles and seemed to be wearing floppy cloth uniforms that bullets would have penetrated. Building gigantic war machines armed with complicated laser weapons seems like a resource allocation mistake that a highly logical AI like Skynet wouldn’t make. Even if very advanced laser weapons are invented someday, I think bullets, missiles, and bombs will retain important advantages, and will be preferred for many common needs.
Let me conclude this topic by saying that, while rifle-sized 1 kW lasers will probably never enter common use, 40 watt lasers that Schwarzenegger could have been referencing might. Getting hit with a one-second long beam from a 40 watt laser wouldn’t kill you, but it would permanently blind you if it hit your eyes. And as I said earlier, if the beam were focused on your shirt for a few seconds, it would light it on fire, causing you to panic and start flailing around. Within a few decades, I can imagine a laser weapon the size of a large rifle, firing lasers in the range of 40 watts, being technically feasible. If paired with a rapid, precise targeting system (such as a humanoid combat robot that can aim weapons better than a human soldier), it could be used to silently “snipe” unsuspecting soldiers up to a half mile away, to blind enemy pilots in low-flying aircraft, and to fry the sensors on enemy vehicles and missiles at the same ranges. The Geneva Conventions forbid laser weapons that blind people because they are too inhumane, but it’s always possible that the Conventions might be revoked in the future, or that humanity could find itself warring with a machine opponent like Skynet that never agreed to them in the first place.
Also there are two types of lasers: 1) continuous beam lasers and 2) pulsed beam lasers. The first type continuously emits photons, producing a long, unbroken laser beam. The second switches on and off very rapidly, producing many short laser beams that follow the same path. The switching happens so fast (a pulsed laser can produce thousands or millions of short beams in a second) that it looks like one, unbroken beam to human eyes, so we can’t see the difference between the two types of lasers.
Therefore, while Schwarzenegger’s request for a “phased plasma rifle in the 40 watt range” made no sense, asking for a “PULSED LASER rifle in the 40 watt range” would have used correct terminology and have referred to a plausible type of weapon. I’m going to email James Cameron so he can do a Director’s re-release of the film.
The energy weapon is probably four feet long
Two-thirds of the way through the film, Kyle Reese has a flashback to an incident where a Terminator infiltrated an underground human base and used an energy weapon to kill many people. Though the weapon is bulkier and longer than most rifles, it could still be deemed a “rifle.” Something that looks like a sling is visible coming out of the back of it. Replace that with a power cord that is connected to a backpack containing batteries and a heat radiator, and the entire system would fairly resemble a 40-watt laser weapon that could be built within a few decades.
Why make small numbers of big Hunter-Killer attack vehicles that the humans can easily see and keep track of, instead of large numbers of small-to-medium-sized Hunter-Killers that the humans would struggle to keep track of? Dog-like robots that could quietly roam the wasteland and crawl inside all the collapsed buildings and sewer holes and use integral assault rifles to shoot humans they found would be devastating weapons, and hundreds of them could probably be made for the price of one Hunter-Killer tank.
Some robots will be indistinguishable from humans. Unlike the Hunter-Killers, which are general-purpose combat vehicles meant to fight humans in open terrain, the Terminators are specialized for infiltration of underground human bases. They are made to look externally identical to people so they can gain entry, and once inside, they use small arms to kill people. As I said in my review of the movie Prometheus, I think machines like this will exist by the year 2100, and quite possibly a few years before that. They will be able to pass for human, even under close-range visual inspection, thanks to fake, non-organic skin and hair. Androids like this won’t exist by 2029 for a variety of reasons.
Some robots will have organic parts. When Kyle Reese first tells Sarah Connor that Schwarzenegger is actually a “T-800” robot, he explains that the earlier “600 series” of robots were easy to spot because they had fake-looking rubber skin. The T-800s have layers of real human muscle, skin, hair, and other tissue around their metal skeletons, making them look identical to humans. Kyle Reese explains that the human tissue is grown in cloning labs and then grafted onto the metal robot bodies. As I said in my most recent Future Predictions blog entry, I don’t think therapeutic cloning technology will be advanced enough to make whole human organs and large amounts of tissue (like muscles and skin) until the 2050s.
More time will be needed to figure out how to graft cloned human biomass onto metal robots and to keep the biomass nourished and healthy. Consider that, if you grow a large flap of skin in a lab and surgically graft it onto the body of a human burn victim, then the new skin links with the person’s blood vessels, nervous systems, and immune system, which keep the patch of new skin fed with oxygen and calories and protected from infections. But if you graft that same flap of skin onto the metal frame of a robot, there’s no organic support system for it at all, so it will die and rot away.
There are two solutions to this problem, both of which require very advanced technology that we’ll have to wait long after 2050 to have: 1) Genetically engineer the tissue so that biological functions normally done by specialized human organs are instead done by patches of the tissue. For example, red blood cells are made inside of human bones, but since a T-800 would only have metal bones, then the T-800’s muscle cells would need to be genetically modified to also make red blood cells. The resulting tissue would look human to the naked eye, but would have so many DNA modifications that it wouldn’t be genetically “human.” 2) Include artificial organs in the T-800s metal frame that interface with the exterior layers of human tissue, and perform the support functions normally done by biological organs. For example, the T-800 could have an artificial heart made of metal and plastic, connected to the blood vessels of its human tissue. The artificial organ would pump blood through the tissue, just like an organic heart would.
While making a robot that is “living tissue over a metal endoskeleton” will be possible someday, it won’t happen by 2029, and I don’t think it will be necessary if the goal is to design an android that looks externally identical to humans. Given what’s already possible with hyperrealistic sculptures, synthetic materials like silicone should be able to mimic the look and feel of human tissue and skin in the future.
Some robots will be bullet-proof. Schwarzenegger’s metal robot body is nearly immune to every bullet that hits him, including those from a shotgun absorbed during a shootout in a dance club, and others from an M-16 fired into his back at close range at a police station. However, he is not completely impervious to damage, as we see during a gruesome “self-repair” scene where he uses hand tools to fix his forearm after it was hit by a shotgun blast, and late in the film when being run over by a truck hurts his leg, and then a stick of dynamite blows him in half. We can already make robots with this level of damage resistance today, and I am sure that future combat robots will have at least this much armor.
A man slides a ceramic plate into a bullet proof vest. The plate adds weight, but also the ability to stop bigger bullets.
Schwarzenegger’s damage threshold is the same as that provided by Level 4 body armor, which typically takes the form of a heavy ceramic plate that a soldier slides into an oversized “pocket” covering the front of his bullet proof vest. A common, 1/2 inch thick steel plate provides the same level of protection at lower cost but more weight, and I’m sure there are many metal alloys that as strong as the previous two, but lighter. It would be entirely possible to build a human-sized robot now that had integrated Level 4 armor, particularly if weight were saved by incorporating that armor into only the robot’s most vital parts, which in the T-800 were the torso and skull. Making robots like this will only get easier as stronger, more lightweight alloys are discovered, or as cheaper ways are found to make today’s armor alloys.
A weapon such as this Barrett M107 rifle fires bullets that can penetrate Level 4 armor, but it is also very heavy and difficult to use. It would be impractical to make something like this the standard infantry weapon.
Giving your combat robots enough armor to resist the most common guns makes clear military sense, and it would force your enemies to adopt bigger weapons that would be so heavy for humans to carry and too hard for us to shoot. For example, the commonest type of .50 caliber machine gun, the M2, weighs 83 lbs and can’t be effectively fired unless it is attached to a tripod that weighs 50 lbs. The bullets are also heavy, so you’d need at least four human soldiers to drag the gun around on a wagon/wheeled tripod just to operate one gun. Fifty caliber sniper rifles and shoulder-launched rockets could work and could be operated by one person apiece, but they’re hard to aim at T-800-sized targets and have slow rates of fire. So Terminator’s prediction that there will someday be human-sized combat robots with integral Level 4 armor is accurate.
This underscores why we SHOULDN’T build armor into non-combat robots. If our robot butlers and maids turn against us someday, we’ll want to be able to easily destroy them with common handguns and axes.
On a tangent, let me say that the bullet-proof T-800 represents only one design philosophy for combat robots meant to kill human infantry. Another approach is to make combat robots that lack armor, but which are just as survivable because they move too fast for humans to shoot them (think of small, low-altitude UAVs) or because they can effectively hide from humans (e.g. – have advanced camouflage features, or are designed for highly accurate, long-distance sniping fire from far away or from high altitude). Another approach would be to make cheaper, more expendable combat robots that would be more vulnerable to human weapons, and to tolerate their higher loss rates because they would killed more humans overall for a smaller investment of money.
Some robots will have superhuman strength. Schwarzenegger displays superhuman levels of strength from the beginning of the film, when he punches a man’s chest so hard that his hand penetrates into his torso, and then emerges gripping the man’s disconnected heart. In another scene, he uses one hand to casually grab a large man standing at a phone booth and throw him several feet away. Many industrial robots and even small machines have superhuman levels of strength, so this prediction has already come true.
This cable cutter tool is the same size as a human hand and forearm, and its two “fingers” can pinch together with many times the force of a human hand.
Whenever we start building human-sized combat robots, at least some of them will have limbs that will be much stronger than humans’. For example, a grown human man with very strong hands could grip an object with 150 lbs of force, but a small, cordless cable cutter whose blades are like short fingers can clamp down on objects with 3,000 lbs of force. It would make sense to build very strong combat robots, principally so they could carry big weapons and manipulate their surroundings better. Improved hand-to-hand combat abilities against humans would be an ancillary benefit since that type of battlefield fighting will be even rarer in the future than it is now.
And again, this should underscore why we SHOULDN’T make our non-combat robots super-strong. For various safety reasons, I don’t think we should design our robot butlers and maids to be stronger, faster, or heavier than average humans. The vast majority of domestic tasks we’d assign to non-combat servant robots could be done under these limitations, and in cases where something couldn’t, one or two extra robots could be rented to help.
Some combat robots will be humanoid. The T-800 is humanoid in form, meaning it has the same body layout as a human and is the same overall size (height, width). I think some future combat robots will be humanoid, but most won’t because other body layouts and sizes will be better for most combat roles.
First, remember that the T-800 was not the only type of combat robot made by Skynet–it also fielded Hunter Killer tanks and aircraft. By virtue of larger size, they could carry bigger, more powerful weapons than the T-800s, and seemed to be the weapons of choice for “surface” combat. The T-800s were made humanoid so they could do special infiltration missions into underground human bases. No clue is given about the size and composition of Skynet’s robot army, but it’s possible that the T-800s represent only a small fraction of its forces, and that most of its robots are Hunter-Killers, or are of some other, non-humanoid design not shown in the film (note that spider-like combat robots were nearly used in Terminator 2‘s future scenes, and killer snake robots were in the fourth film). This is a detail that is important but easily overlooked, and it will prove accurate: after the world’s militaries have switched to using robots for combat, only a minority of those robots will be humanoid.
Storyboard art for a combat robot that was almost included in Terminator 2’s war scenes. Having four legs would make it more stable than robots with only two, and legs would let it climb over obstacles and up steep slopes that wheeled vehicles couldn’t.
Many combat robots will look almost the same as war machines we have today: Autonomous planes will still have at least one engine for propulsion and two wings so they can use lift, autonomous ships will still be oblong and pointy at the front end to minimize friction with the water, and autonomous armored vehicles will still have two sets of wheels and some kind of gun turret on top (see my blog post about a hypothetical robot tank). The only visual differences between those future weapons and their contemporary counterparts might be slightly smaller dimensions and the deletion of cockpits and structural bulges since there won’t need to be big interior spaces for humans (though they would need to have some number of small robots for field maintenance and repair, as I also described in the robot tank blog post). If Skynet were actually created and if it built a robot army to fight humanity, most of its aircraft, ships, and land vehicles would look very familiar to us.
Those sorts of combat robots would excel at destroying our heavy weapons, vehicles, and structures, but it would be wasteful to use them to hunt down small groups of humans armed only with light weapons, which describes the people living in The Terminator‘s post apocalyptic future. Moreover, robot tanks, fighter planes, and ships can’t go inside structures, sewer tunnels, or thickly wooded areas. Smaller combat robots of different designs would be needed to efficiently fight human infantry, particularly in the environments I’ve listed.
Would these robots be humanoid, like the T-800? Maybe. For sure, they would need to have bodies that were narrow and short enough to fit through standard-sized doorways or between trees in a dense forest, and light enough to not collapse floors when they walked over them. They would need to be able to fit themselves into tight spaces that humans can, like small caves and basement crawlspaces. They would also need legs–not wheels or caterpillar tracks–so they could go up and down stairs, operate pedals commonly found in human-driven cars, hop over fallen tree trunks and climb steep hills and ridges. They would also need hands so they could manipulate and use things in built human environments, like doorknobs, keys, and push-buttons. Being able to hold and use weapons, tools, keyboards, and other things designed around human hands would be very useful, as the robot would be finding such objects all the time.
A robot can be human-sized but not human in form.
Those design requirements might sound like they add up to a robot that must be humanoid, but it’s not at all the case. The requirements could be met by a robot that had a centaur-like body (four legs is more stable than two, anyway), or that lacked a head and instead had prehensile stalks coming out of its neck with cameras and microphones on their ends (a head makes a body top-heavy and packs too much important stuff in one place), or that had four arms, or four tentacles with hands on their ends (more arms means you can do and hold more stuff at once). Its hands might have four or eight fingers apiece, and it might be five feet tall but three feet wide, or seven feet tall and 18 inches wide. It could have a shiny, metal exterior that looks totally inhuman, or could be intentionally made to look scary to humans, perhaps like something from a horror movie. While robots like this wouldn’t be able to blend in with humans and “walk past the sentry,” they could go inside all the houses, vehicles, tunnels, and other places humans could go, and kill us wherever they found us.
I can only think of two types of military missions for which a human-looking combat robot would be well-suited: 1) assassinations and 2) infiltration/spying. Given that, during wartime, only a small fraction of military operations are of such a character, it follows that only a concordantly small fraction of any military’s robots would look human. Also, since stealth is important, the humanoid robots would mostly be made to look as boring as possible, perhaps like a middle-aged woman, a child, an old man, or an average soldier. Making them eye-catching by giving them Schwarzenegger’s bodybuilder physique or by making them handsome/pretty, would be counterproductive in most cases.
Even in the narrow use cases we’ve whittled our way down to, I think other types of robots and weapons would be better than using humanoid robots. By virtue of their smaller size, robots made to look like insects and small animals could infiltrate human spaces more easily than a man-sized robot. Many of them could also be built for the price of one T-800, and having more means higher odds of one successfully completing its spy mission. A “robot rat” could also assassinate people by injecting them with poison, releasing lethal gas, or jumping on the target’s face and activating an internal explosive. Even something as small as a robotic mosquito could kill, by injecting poison into the target’s bloodstream with its stinger (note that a single drop of botulinum toxin can kill several men). It would be impossible for humans to stay constantly vigilant against threats so insidious. An even cheaper solution would be bombs full of heavier-than-air poison gas.
So in conclusion, I think it’s possible that some combat robots will look like humans, but they will be used for rare special missions (and this was accurately portrayed in the movie), and the vast majority of combat robots will look totally different. In the very long run, I don’t think any of them will look human.
There will be fully automated factories. Kyle Reese reveals that the T-800 robots are made in fully automated factories run by Skynet. As I said in my review of I, Robot, all factory jobs will inevitably be taken over by machines, so it’s just a question of how long it will take. I predicted that a handful of such factories would exist by 2035–principally as technology demonstrators or for a tech billionaire like Elon Musk to claim bragging rights–but it would take decades longer for them to become common. I doubt they will exist as early as 2029.
John von Neumann was a polymath whose intelligence impressed even his genius colleagues. He concluded that machines would someday be able to build and fix themselves, eliminating the need for humans.
The common refrain that goes something like: “Human workers will always be needed, because without us, who would build or fix the robots?” is actually false and illogical. The fact that we haven’t yet invented robots that can build other robots without human help doesn’t imply that it will remain that way forever, or that humans have some special, creative quality that can never be transplanted to machines. John von Neumann, who one of the greatest minds of the 20th century and a pioneer in computer science, theorized in his paper “The General and Logical Theory of Automata” that sufficiently advanced artificial life forms (machines) could make copies of themselves, including copies that were engineered to be better, and that there was no reason why humans would always be needed to build, fix, or improve the machines. We can be totally cut out of the loop, and I predict someday we will.
There’s no theoretical reason why the entire production chain of making a robot as complex as a T-800–from digging the raw metals out of the ground, refining them, forging and shaping them into body parts, assembling the parts, and transporting the finished product to its place of use–can’t be 100% automated someday. I conservatively predict that most manufactured goods will come from automated factories by 2100.
Robots will be able to fix themselves. As I mentioned before, after sustaining damage to lightly armored parts of his robot body, Schwarzenegger does repair surgery on himself, using a small knife and a pair of pliers. Machines won’t be capable of this level of self-repair by 2029, but thanks to the factors I listed in the previous paragraph, they will inevitably gain the ability. The ability to build something implies an ability to repair it as well. Someday, robots will be able to fix each other and to fix themselves.
I note that full self-repair abilities will require the robot in question to be able to see and touch every spot on its own body, which in turn makes some design features necessary. It’s arms would need to be long and double-jointed, and if it had eyes set in a head like humans, then the head would need to be able to swivel 360 degrees to it could look at damage to its back. It wasn’t clear if the T-800 had these features. Other ways to solve this problem might be to give it long, telescoping tentacles in place of a head, with cameras at the ends of each (this would also make it much less risky to “peek” around a corner in combat to see if any bad guys were there). The tentacles could bend in various ways to give the robot a clear view of any part of its body, from nearly any angle. Having small cameras built into fingers and feet would accomplish the same thing. The ability to detach body parts would also be very useful, as it would let robots work on their damaged parts more easily, and because it would let them quickly swap out their parts for functional new ones if any were at hand.
Again, I conservatively predict that non-trivial numbers of robots will have sophisticated self-repair and “peer repair” abilities by 2100.
Robots will be able to keep working in spite of massive damage. At the end of the film, the T-800 played by Schwarzenegger is blown in half by a stick of dynamite that Kyle Reese shoves into the bottom of its exposed rib cage. In spite of this catastrophic injury, the T-800 keeps fighting, using its hands to drag the functional upper half of its body along the floor so it can get to get to Sarah Connor and manually kill her. Some robots are already this resilient, and robots made in the future–particularly those designed for combat–will be even more so.
A Northrop Grumman Remotec “Andros” robot like this withstood a bomb attack against a human.
So long as a robot’s power source and main computer are intact and connected to each other, it will keep working, even if all other parts of its body are nonfunctional. The inability to feel pain and a lack of a circulatory system allows robots to survive major injuries like the loss of limbs that would incapacitate humans due to psychological shock, pain, and blood loss. In 2016, police in Dallas, TX used a remote-controlled robot to kill a criminal who had shot several of their comrades and barricaded himself in a building. The robot had a bomb grasped in its hand, maneuvered close to the criminal, and then the device detonated, killing the suspect. Though the robot’s arm was blown off by the explosion, the machine remained functional, and was repairable after the incident.
In the future, I think it might be advantageous for each major robot body part or body segment to have its own computer, sensors, and power supply. That way, if a part were severed, it could still function for a while independently. For instance, if a T-800 had its arm severed, then the arm’s internal computer would switch on, would be able to see its surroundings via tiny cameras in the fingertips and knuckles, and would be able to drag itself around like a spider or like “Thing” from The Addams Family. It could drag itself to the robot body it was formerly attached to, or crawl away to find help. Though this might sound macabre and useless, note that many insects, including the highly evolved and successful cockroach species, have distributed nervous systems that grant their body segments similar abilities. They wouldn’t have evolved that way unless it was useful somehow. Additionally, under normal conditions, it would probably benefit a robot to distribute its computation and power load across multiple nodes in its body, and having sensors in all its extremities and body parts could only boost its utility.
Machines will be able to do near-perfect imitations of human voices. At two points in the film, the T-800 accurately impersonates the voices of humans to fool people who are listening via radio or telephone. In recent years, deep learning algorithms have become extremely good at this (see the recently released recording of a machine impersonating Joe Rogan’s voice), and at the rate of quality improvement, I think the machine imitations will sound flawless to us by 2029.
However, there is one important inaccuracy in the film: The T-800 is able to imitate humans after hearing them speak only a few words. Today’s deep learning algorithms need to listen to many hours of someone’s recorded speech to understand how they speak well enough to copy their voice, and the requirement for large sets of training data will still exist in 2029.
