Amongst the sea of spilled ink about America’s purported “STEM shortage,” and policy proposals to address this disaster by training preschoolers to code, this article stands out as one of the best counterpoints I’ve seen:
‘[Although] I certainly believe that any member of our highly digital society should be familiar with how these [software] platforms work, universal code literacy won’t solve our employment crisis any more than the universal ability to read and write would result in a full-employment economy of book publishing.’ (SOURCE)
OUCH! The article goes on describe how lower-skilled computer programming jobs are being outsourced to India, leaving a pool of higher-skilled jobs here in the U.S., which will get more cutthroat as time passes.
I’ll add the following:
Computer coding is a dry, difficult job that few people are suited for. It requires tremendous patience, good math skills, a willingness to work brutal hours to get promoted, and it provides few (if any) opportunities for self-expression or emotionally interacting with clients. You sit in a cubicle looking at numbers and letters on a computer screen, tediously typing away and testing your software program over and over to work out the kinks. The notion that America can expand its white-collar workforce by incentivizing more people to become computer programmers rests on the flawed assumption that human beings are perfectly interchangeable widgets lacking innate strengths, weaknesses and preferences that together limit their job options. The vast majority of people just aren’t cut out to spend eight hours a day poring over computer code.
Wages will decrease if labor supply increases. As with any other profession, computer programmer salaries are determined by supply and demand. If the STEM Shortage Chicken Littles get their way and the number of American computer programmers sharply increases, then median wages will decrease unless there’s an equivalent rise in demand for their services. Lower pay will make an already dull and difficult job not worth it for many coders, and people will start fleeing for other jobs, counterbalancing the inflow of new coders.
If we do think that there’s a shortage of computer programmers in America, then there’s a fair case to be made that the best way to fix it is to focus on retaining existing talent rather than trying to attract new entrants to the field. Complaints about age discrimination against older workers, low pay, and overly demanding work schedules seem pervasive if the news articles out of Silicon Valley are to be believed, and are supported by high rates of turnover in computer programming companies.
Bloomberg New Energy Finance just released an analysis, “Electric Vehicle Outlook 2017,” that estimates all-electric cars will get as cheap as traditional gas-powered cars sometime between 2025 and 2030. Importantly, the estimate assumes that government subsidies for electric cars are discontinued by that time, so the future price figures are market rates.
Bloomberg thinks electric car prices will drop thanks to price-lowering economies of scale and to competition among carmakers. It doesn’t assume any technological breakthroughs like new batteries that can store twice as much energy. This is good: making predictions about the future that hinge on a technological breakthrough that may or may not actually happen is always a bad idea, and will get you thinking something like the Singularity is right around the corner.
The Bloomberg Executive Summary is here: https://data.bloomberglp.com/bnef/sites/14/2017/07/BNEF_EVO_2017_ExecutiveSummary.pdf
Interestingly, the analysis also concludes that better electric cars will make plug-in hybrids obsolete since the latter are more mechanically complex and hence more expensive. A shortage of at-home car charging stations will also limit the potential customer base for electric cars, and cause electric cars as a share of the total passenger vehicle fleet to stabilize at about 50% by 2040. I wish I had access to the full report, so I can only guess that the at-home car charging problem will be most acute for poorer people who can’t afford to install them or who live in rental properties that lack them.
Futurist, transhumanist, and singularitarian Ray Kurzweil just did a short interview regarding his views on the future of jobs and some other topics. You can see it here:
My thoughts:
0:12 – Kurzweil nods to the Boring Truth of our age: The clash of ideologies ended in the 20th century (except for the ongoing sideshow that is non-viable Islamism vs. everyone else), and there’s consensus among academics and leaders in the industrialized world that having a mixed economy and some social welfare programs is close to the optimal setup for a country. In the West, conservatives and liberals push and pull, but within narrow boundaries. Similarly, the new political faultlines pit “nationalists” against “globalists,” but no one in the former camp wants to completely forsake trade. There really is a lot less drama today than the news media makes you think.
1:50 – The reasons for the rise of welfare states in the early 20th century are more complex than that, but Kurzweil makes a good point that they wouldn’t have been sustainable had there not been the economic surpluses made possible by Industrialization. If you take the long view like Kurzweil does, and you assume that technology keeps improving, the concomitant economic surpluses keep growing, and social welfare programs grow in an intelligent manner, then a future where all humans are on the dole and few if any people work is indeed the logical endpoint.
3:06 – Uh-oh. Kurzweil makes predictions that will be true in “a decade.” So by 2027, 3D printers will be able to make “at low cost, all of the physical things we need,” including large Lego-like pieces of building materials that you will be able to “snap together” to make your own house. Vertical farms will also be making “very high-quality” food at “very low prices” by 2027. Yikes. I’m skeptical of the 3D printed house prediction because the construction industry and consumers have failed to even embrace modular buildings (there’s a great report on this here: http://www.mckinsey.com/industries/capital-projects-and-infrastructure/our-insights/reinventing-construction-through-a-productivity-revolution). The notion that something even more radical like 3D printed Lego houses will become common in just ten years bucks the trend way too much. Also, I don’t see how an average person in 2027 will be able to assemble his or her house from giant Legos considering: 1) the need to pour solid concrete foundations will still exist, 2) local governments are highly unlikely to relax building codes to allow unlicensed, inexperienced people to build houses, and 3) few people have the skills to even put such Lego pieces together, particularly with enough accuracy to ensure the surfaces are truly level and plumb. Maybe what Kurzweil is trying to say is that, in 2027, there will be some construction companies that will specialize in building cheap, prefabricated houses comprised partly of 3D printed components. Plausible, but only a tiny bit different from how things are today. As for vertical farms, they’ve proven to be much more expensive to run than normal “flat” farms and haven’t caught on thanks to basic economics. If Kurzweil knows of some way that they can make food at “very low prices” in just ten years, then he should quit his job at Google and pursue it full-time since it will be worth billions of dollars. And he should also ask himself whether it would be more efficient and profitable to use that secret method to improve “flat” farms. For example, if Kurzweil thinks vertical farm costs will drop thanks to cheap, 3D printed building techniques, then won’t the same techniques also make it possible to cheaply build greenhouses over standard cropfields? If farm robots will eliminate labor costs at vertical farms, won’t they do the same at flat farms? Why would the vertical farms benefit more?
4:09 – Kurzweil observes (as he has in the past) that most of the Earth’s surface is sparsely populated, meaning there is ample room for humans to spread out. While true, it’s important to remember the reasons why: Beachfront property in Florida is more aesthetically appealing and provides more opportunities for recreation than a plot of land in the middle of Nebraska. The climate in San Francisco is more conducive to human life than that of Minot, North Dakota. Humans are also social animals (particularly when young), meaning they like to live in places where there are other people. The high (and still rising) rates of suicide and substance abuse in rural America attest to the ill effects of isolation and lack of varied things to do. He doesn’t say it in this interview, but I know from his books that his response would be something like “future technologies will substitute for all that,” meaning virtual reality will be as real as The Matrix someday, so hanging out on virtual reality Miami Beach while you’re actually lying in a VR pod in your living room will feel as real as hanging out on the real Miami Beach with your actual body. Whether or not sufficiently advanced brain-computer interfaces can be made to do that is an open question, but for sure, I doubt the technology will exist by 2027, or even 2057.
5:00 – Kurzweil predicts that, by 2027, virtual reality and “virtual avatars” will be so good that many people won’t need to live in cities anymore, and he seems to suggest there will be a detectable change to the global urbanization trend. Thanks to virtual reality, people will be able to work and play from anywhere, so they’ll choose to live outside of cities to save money. I think this is a prime example of a prediction that Kurzweil can’t possibly get wrong, and that is also almost useless. As he admits around this part of the interview, many of his colleagues at Google already work remotely, and most of us know someone who works from home. It doesn’t take a futurist or economist to see that the practice is getting more popular, so it’s a simple assumption that it will be more common by, say, 2027. Technologies related to computing, videoconferencing, and virtual reality are all obviously improving, and it’s just common sense that they will make it easier for people to work remotely. And while the number of people living in cities is growing, so is the number of people living outside of them in the suburbs and exurbs. By 2027, the suburban/exurban population could be growing faster than the truly urban population, which Kurzweil could cite as proof his prediction was right. So on close analysis, Kurzweil’s prediction is nothing more than a simple synthesis of three long-running trends in America that most adults are already aware of through direct experience. It will be almost impossible for him to be wrong, but the prediction about the future is so general and so incrementally different from today that it has no real value.
6:35 – He says we will use 3D printers to make clothes, without giving a date for when the prediction will come to pass (by 2027?). Regardless of when or if it happens, this has always struck me as a useless application of 3D printers. Today, I can buy a pack of six new cotton undershirts from Wal-Mart for $15, and they will last for years before falling apart. I can go to a local thrift store and buy durable, surprisingly good-looking used clothes that are 75% discounted from their original prices, and which will also last me many years. I can go on Craigslist right now and find people in my area who are giving away clothes for free. There is no evidence at all that our existing textile technology is deficient making clothes, or that our “standards of living” will meaningfully improve if we started making clothes with futuristic 3D printers. Even if we assume 3D printers are so superior at making clothes that they’re (almost) “free,” how much better is that than the present condition? Clothes are already free or trivially cheap. Lowering the price farther might free up enough money for you to buy a slightly bigger morning coffee at Starbucks, but that’s it. The only real beneficiaries would be fashion-obsessed people who shudder at the thought of wearing the same outfit twice and want their 3D printer to spit out some zany new creation each morning. Yay for the coming empowerment of vain people.
8:10 – Kurzweil cites changes to the nature of jobs over the last 100 years (workforce transformed from hard labor on the farm and factory to doing computer stuff in office buildings) as proof that there will always be jobs for humans in the future. While humans have always managed to move up the skills ladder and create new, gainful work for themselves as machines took over the less skilled jobs, there’s no reason to think the trend will continue forever. His argument also gets muddled when he equates people in college with people who have jobs. Studying poetry or art in college isn’t the same thing as being gainfully employed. Moreover, its a common fate for such students to have problems finding employment after college, and for them to settle for jobs that are unsatisfactory because they pay little, or because they have nothing to do with what they studied (think of the waitress with the Literature B.A.). I think it’s much safer to predict that “Humans in the future will be able to find things to do with their days but they won’t necessarily get paid much money or any money at all for what they do, and automation will be good overall for humans since it will eliminate unpleasant drudge work.”
Aeon has a good article about the religious undertones to Singularitarianism. (FYI, “Singularitarianism” is the belief that the Technological Singularity will happen in the future. While Singularitarians can’t agree if it will be good or bad for humans, they do agree that we should do whatever we can until then to nudge it towards a positive outcome.) This passage sums up the article’s key points:
‘A god-like being of infinite knowing (the singularity); an escape of the flesh and this limited world (uploading our minds); a moment of transfiguration or ‘end of days’ (the singularity as a moment of rapture); prophets (even if they work for Google); demons and hell (even if it’s an eternal computer simulation of suffering), and evangelists who wear smart suits (just like the religious ones do). Consciously and unconsciously, religious ideas are at work in the narratives of those discussing, planning, and hoping for a future shaped by AI.’
Having spent years reading futurist books, interacting with futurists on social media, and even going to futurist conferences, I’ve come to view Singularitarians as a subcategory of futurists, who are defined by their belief in the coming Singularity and by the religious qualities of their beliefs. Not only do they indulge in fantastical ruminations about what the future will be like thanks to the Singularity, but they use rhetorical hand-waving–usually by invoking “exponential acceleration of technology” or something like that–to explain how we’ll get there from our present state. This sharply contrasts with other futurists who are rigidly scientific and make predictions by carefully identifying and extrapolating existing trends, which in turn almost always results in slower growth future scenarios.
A sizable minority of Singularitarians I’ve encountered also seem to be mentally ill and/or poor, and the thought of an upending of daily life and of the existing socioeconomic order, and the thought of an end to human suffering thanks to advanced technologies appeal to them for obvious reasons. Their belief in the Singularity truly is like the psychological salve of religion, so challenge them at your own risk.
Singularitarians could also be thought of as a subcategory of Transhumanists, the latter being people who believe in using technology to upgrade human beings past their natural limitations (such as intelligence, lifespan, physical strength, etc.). If you believe that the Singularity will bring with it the ability for humans to upload their minds into computers and live forever, then you are by default a Transhumanist. And you’re a doubleplus Transhumanist if you go a step farther and make a value judgement that such an “upgrade” will be good for humans.
With those distinctions made clear, let me say that I am a futurist and a Transhumanist, but I am not a Singularitarian. I plan to explain my reasons in depth in a future blog post, but for now let me summarize by saying I don’t see evidence of exponential improvement in artificial intelligence or nanomachines, which are the two pillars upon which the Singularity hypothesis rests. And even if an artificial intelligence became smarter than humans and gained the ability to rapidly improve itself, something called the “complexity brake” would slow its progress enough for humans to have some control over it or to at least comprehend what it was doing. Many Singularitarians believe in scenarios where the Singularity unfolds over the course of literally a few days, with a machine exceeding human intelligence at the beginning, and all of planet Earth being transformed into a wonderland of carbon nanotube structures, robots, humans sleeping in Matrix pods, and perhaps some kind of weird spiritual transcendence by the end. The transformation is predicted to be so abrupt that humans will have no time to react or to even fully understand what’s happening around them.
The Economist has a rather disturbing article about how advances in “generative adversarial networks” will soon make it possible to create computer-generated audio and video footage that is indistinguishable from the real thing. The potential for spreading misinformation is obvious. The article offers some ways that such fakes could be spotted:
‘Yet even as technology drives new forms of artifice, it also offers new ways to combat it. One form of verification is to demand that recordings come with their metadata, which show when, where and how they were captured. Knowing such things makes it possible to eliminate a photograph as a fake on the basis, for example, of a mismatch with known local conditions at the time.
…Amnesty International is already grappling with some of these issues. Its Citizen Evidence Lab verifies videos and images of alleged human-rights abuses. It uses Google Earth to examine background landscapes and to test whether a video or image was captured when and where it claims. It uses Wolfram Alpha, a search engine, to cross-reference historical weather conditions against those claimed in the video. Amnesty’s work mostly catches old videos that are being labelled as a new atrocity, but it will have to watch out for generated video, too. Cryptography could also help to verify that content has come from a trusted organisation. Media could be signed with a unique key that only the signing organisation—or the originating device—possesses.’
However, it would be naive to think that these methods couldn’t be defeated with better CGI algorithms and through hacking file metadata and cryptographic keys.
And even if the “good guys” manage to forever stay one step ahead, we’re still rapidly approaching an era where the forgeries will be so good that unaided human eyesight and hearing won’t be sensitive enough to detect them, and humans will have to rely on machines to tell them what is real and what is fake (which is itself an interesting state of affairs from a philosophical standpoint, but that’s a talk for a different time). Something like a few fragments of aberrant computer code embedded in an otherwise perfect-looking fake video might be the only thing that reveals the lie. Considering the short attention span and low level of scientific and technological literacy in most countries, how could the computer forensic findings in such a case ever be explained to average people?
They couldn’t, which means belief or disbelief in accusations of forgery will twist in the winds of whatever preexisting biases each person has, which is how it is now. Americans will believe it when their government tells them a video originating in Russia is fake, and Russians who mistrust America will reflexively disagree and believe their own government’s claims it is genuine. The truth will of course be out in the open, but so abstruse that only a small minority will be able to see it clearly on their own.
Moreover, the ability to make perfect computer generated audio and video imitations of people could lead to disaster in crisis situations where the intended target lacks either the ability or the time to verify their authenticity using their own technology: Imagine a military battle where one side transmits false orders to the other, in the voice of the latter’s commander, or a situation where a hacker posing as a rich investor calls his stock broker and insistently tells him to trade some massive number of shares.
*Update (7/13/2017): Computer scientists at the University of Washington have developed a way to merge audio recordings of someone speaking with video footage of them, so their mouth appears to be moving in sync with the words, even though the audio and video are from two different sources. Here’s a sample of them manipulating a speech by Barack Obama:
