I just returned from Manifest, a bay area rationalist conference hosted by the prediction market platform, Manifold. The conference was great and I met lots of cool people!
A common topic of conversation was AI and its implications for the future. The standard pattern for these conversations is dueling estimates of P(doom), but I want to avoid this pattern. It seems difficult to change anyone’s mind this way and the P(doom) framing splits all futures into heaven or hell without much detail in-between.
Instead of P(doom), I want to focus on the probability that my skills become obsolete. Even in worlds where we are safe from apocalyptic consequences of AI, there will be tumultuous economic changes that might leave writers, researchers, programmers, or academics with much less (or more) income and influence than they had before.
Here are four relevant analogies which I use to model how cognitive labor might respond to AI progress.
First: the printing press. If you were an author in 1400, the largest value-add you brought was your handwriting. The labor and skill that went into copying made up the majority of the value of each book. Authors confronted with a future where their most valued skill is automated at thousands of times their speed and accuracy were probably terrified. Each book would be worth a tiny fraction of what they were worth before, surely not enough to support a career.
With hindsight we see that even as all the terrifying extrapolations of printing automation materialized, the income, influence, and number of authors soared. Even though each book cost thousands of times less to produce, the quantity of books demanded increased so much in response to their falling cost that it outstripped the productivity gains and required more authors, in addition to a 1000x increase in books per author, to fulfill it.
Second: the mechanization of farming. This is another example of massive productivity increases but with different outcomes.
Unlike authors and the printing press, the ratio of change in production per person and change total consumption was not low enough to grow the farming labor force or even sustain it. The per-capita incomes of farmers have doubled several times over but there are many fewer farmers, even in absolute numbers.
Third: computers. Specifically, the shift from the job title of computer to the name of the machine that replaced it. The skill of solving complicated numeric equations by hand was outright replaced. There was some intermediate period of productivity enhancement where the human computers or cashiers used calculators, but eventually these tools became so cheap and easy for anyone to use that a separate, specialized position for someone who just inputs and outputs numbers disappeared. This industry was replaced by a new industry of people who programmed the automation of the previous one. Software engineering now produces and employs many more resources, in shares and absolute numbers, than the industry of human computers did. The skills required and rewarded in this new industry are different, though there is some overlap in skills which helped some computers transition into programmers.
Finally: the ice trade. In the 19th and early 20th centuries, before small ice machines were common, harvesting and shipping ice around the world was a large industry employing hundreds of thousands of workers. In the early 19th century this meant sawing ice off of glaciers in Alaska or Switzerland and shipping it in insulated boats to large American cities and the Caribbean. Soon after the invention of electricity, large artificial ice farms popped up closer to their customer base. By WW2 the industry had collapsed and been replaced by home refrigeration. This is similar to the computer story but the replacement job of manufacturing refrigerators never grew larger than the globe spanning, labor intensive ice trade.
This framework of analogies is useful for projecting possible futures for different cognitive labor industries. It helps concretize two important variables: whether the tech will be a productivity enhancing tool for human labor or an automating substitute and whether the elasticity of demand for the outputs of the augmented production process is high enough to raise the quantity of labor input.
All of these analogies are relevant to the future of cognitive labor. In my particular research and writing corner of cognitive labor, the most optimistic story is the printing press. The current biggest value adds for this work: literature synthesis, data science, and prose seem certain to be automated by AI. But as with the printing press, automating the highest value parts of a task is not sufficient to decrease the income, influence, or number of people doing it. Google automated much of the research process but it has supported the effusion of thousands of incredible online writers and researchers.
Perhaps this can continue and online writing will become even more rewarding, higher quality, and popular. This requires the demand for writing and research to expand enough to more than offset the increased productivity of each writer already producing it. With the printing press, this meant more people reading more copies of the same work. With the internet, there is already an essentially infinite supply of online writing and research that can accessed and copied for free. Demand is already satiated there, but demand for the highest quality pieces of content is not. In my life, at least, there is room for more Scott Alexander quality writing despite being inundated with lots of content below that threshold. AI may enable a larger industry of people producing writing and research of the highest quality.
If quantity demanded, even for these highest quality pieces of content, does not increase enough then we will see writing and research become more like mechanized farming. An ever smaller number of people using capital investment to satiate the world’s demand. Highly skewed winner-take-all markets on youtube, larger teams in science, and the publishing market may be early signs of this dynamic.
The third and fourth possibilities, where so many tasks in a particular job are automated that the entire job is replaced, seems most salient for programmers and paralegals. It would be a poetic future where programmer, the job title which rose to replace the now archaic “computer,” also becomes an archaic term. This could happen if LLMs fully automate writing programs to fulfill a given goal and humans graduate to long term planning and coordination of these programmers.
The demand for the software products that these planners could make in cooperation with LLMs seems highly elastic, so most forms of cognitive labor seem safe from the ice trade scenario for this reason. Perhaps automating something like a secretary would not increase not increase the quantity demanded for the products they help produce enough to offset that decline in employment
When predicting the impact of AI on a given set of skills most people only focus on the left-hand axis of the 2x2 table above. But differences in the top axis can flip the sign of expected impact. If writing and research will get enhanced productivity from AI tools, they can still be bad skills to invest in if all of the returns to this productivity will go to a tiny group of winners in the market for ideas. If your current job will be fully automated and replaced by an AI, it can still turn out well if you can get into a rapidly expanding replacement industry.
The top axis also has the more relevant question for the future of cognitive labor, at least the type I do: writing and research. Is there enough latent demand for good writing to accommodate both a 10x more productive Scott Alexander and a troupe of great new writers? Or will Mecha-Scott satiate the world and leave the rest of us practicing our handwriting in 1439?
I think AIs will be able to do all cognitive labor for less resources than a human could survive on. In particular, “Scott Alexander quality writing” and “long term planning and coordination of programmers”—tasks that you assume will stay with humans—seem to me like tasks where AIs will surpass the best humans before the decade is out. Any “replacement industry” tasks can be taken up by AIs as well, because AI learning will keep getting better and more general. And it doesn’t seem to matter whether demand satiates or grows: even fast-growing demand would be cheaper met by building more AIs than by using the same resources to feed humans.
(This is also why Ricardian comparative advantage won’t apply. If the AI side has a choice of trading with humans for something, vs. spending the same resources on building AIs to produce the same thing cheaper, then the latter option is more profitable. So after a certain point in capability development, the only thing AIs and AI companies will want from us is our resources, like land; not our labor. The best analogy is enclosures in England.)
Maybe it’s equivalent, but I have been thinking of this as “The price at which humans have comparative advantage will become lower than subsistence, so humans refuse the job and/or die out anyway.” AKA this is what happened to most horses after cars got cheap.
Yeah, I think your formulation is more correct than mine.
That seems unlikely to me. The occupations of professional scribe and author did not much overlap. Can you name 3 examples of authors from 1400-1500 whose name anyone would know without checking Wikipedia, who primarily made a living as scribes and so might have been ‘terrified’ to learn of Gutenberg’s press, rather than elated? This sounds like a strawman. (WP, for example, provides a list of notable scribes; the only name I recognize is fictional.) How does this scenario where authors are valued primarily for their ‘handwriting’ work economically, in terms of specialization and comparative advantage? Geoffrey Chaucer sits down to write the next installment of Canterbury Tales and on completion, instead of handing over it to one of the legions of professional scribes to copy, he… starts copying out by hand 100 copies himself? Surely whatever compensation Chaucer (or other authors) received which motivated him to write, it was not being paid a few pounds for a stack of 100 copies he made and then sells in the streets.
On a side note, you still haven’t responded to the comments on your most recent post about misrepresenting critics to provide a strawman.
Hmm fair enough, I didn’t consider that there would already be a lot of specialization between authors and copyists pre-press. Still, I think I can rewrite the paragraph to remove this error and preserve the parts relevant to the overall post:
>First: the printing press. In 1400, the labor and skill that went into copying a book made up the majority of its value. Authors confronted with a future where the most valuable part of each of their books is automated for a tiny fraction of the cost might understandably be terrified. Each book would be worth a tiny fraction of what they were worth before, surely not enough to support a career.
That still makes no sense. Why would authors be terrified by scribes being disemployed, when authors received no percentage or payment whatsoever from scribes per copy? At the worst, they would be indifferent. It would matter as much to them as, say, someone discovering a replacement for parchment or vellum which threatened the livelihoods of sheepherders (like Chinese ‘paper’ made from plants rather than animals).
A couple of reasons why authors might be worried about the press:
It’s a massive change to the technology of what they produce. This comes with lots of uncertainty and fear.
It commodifies books and massively decreases the unit price. Depending on how much you think quantity demanded will change, it could easily decrease your income. E.g, if the press came around and no one read any more books, it would be scary for authors and many would be out of work since now a single author can produce 100x more books.
This is nicely thought out and composed. It seems like you’re asking the question: what will skills be worth in the new equilibrium? I think this is an intuitive question, but in this case it’s probably not a useful question to ask. The new equilibrium is that machines outcompete us dramatically for almost all jobs.
The only alternative is not believing that we’ll get actual general AI any time soon. If we do, it will handily outcompete humans at everything it’s allowed to do, and very quickly. So less than subsistence, is the very likely answer to how much we’ll make, for almost all trades.
The relevant question is how quickly jobs become almost worthless. We could ask how long MY job will keep me fed, but it’s probably more productive to think beyond that.
The most relevant question is: what are we all going to do about it? If we solve technical alignment, we’re still faced with a world with vastly increased productivity but vastly decreased job opportunities. Capitalism as we know it isn’t going to work. A gentle reshuffling, a la minimum basic income, probably won’t even cover it for the richest countries, let alone the world.
To keep everyone alive let alone happy, we need radical new plans for the transition from capitalism to post-scarcity.
I’ve tried to listen to everyone thinking about the job loss issue. I haven’t found anyone smart and credible who’ll even claim to have a good guess about transition rates and whether we’ll get economic collapse. To me, it seems quite likely we will. Maybe it’s time to start thinking about how we’ll survive the transition even if we get the heaven scenario from AI in the longer run.
What fraction of job loss can the economy sustain over a short time? If we lose 5% of jobs globally in two years, we might be fine. The actual numbers are probably much higher. I’m no economist, but it seems like we’re heading for a massive recession during the transition to a almost-no-work-for-humans “economy”.
Yeah. The way the world works now, if technical alignment work is successful, it will just lead to AIs that are aligned to making money or winning wars. There need to be AIs aligned to human flourishing, but nobody wants to spend money training those. OpenAI was started for this purpose, but got taken over by money interests.
An AI aligned to making money is not that much better than one aligned to making paperclips.