Ex-Google Exec Says "The Idea That AI Will Create New Jobs Is 100% Crap"
windowscentral.com
windowscentral.com
In his last interview from this series in 2023, Gawdat said that LLMs are "alive in every possible way," have a "very deep consciousness," are "definitely aware," and "feel emotions."
His current endeavors involve an unlaunched combination LLM relationship coach and dating site (matchmaking humans with humans, not with LLMs, as far as I can tell).
...has he looked at a calendar lately? I just did, and it turns out that 2026 is less than five months away.
I'm not even going to touch the "alive in every possible way" bullshit.
Given the Holocaust took ~12 years (including all the legal/political groundwork) and the Holodomor a year, the "between" doesn't fill me with confidence...
it's been only ~6 months or so since Trump took office and he's already annihilated the US Fed Gov and is now declaring martial law.
and he's concerned with appearances, sorta. AI won't be.
shoutout to Paul Virilio and how technology drives speed which drives war and destruction
We have gone from something like 90% of people employed growing food to something in the single digits (I'm too lazy at the moment to get exact figures, but this is close enough for my point).
Did all those advances in agriculture create jobs? I would say yes and no.
For the specific industry the advances happened in? No, definitely not. There are way fewer jobs in agriculture now.
But what did happen is that people were freed up to do different things. Those jobs were indirectly created by the fact that people did not need to spend so much time growing food. That allowed people to do other things to create value and improve our lives. Without those agriculture advances, we wouldn't have AI right now.
I suspect AI will be similar. We aren't going to have more jobs in the technology section because of AI. But that guy who hears "Code Monkey" for the first time and marvels that someone wrote a song about his life? He won't be a code monkey anymore he will be doing something else. If he's lucky something more personally fulfilling.
Anyways, if I put my cynnical hat on after looking at this: https://blog.waldrn.com/p/american-boys-have-become-less-sup...
The next generation of children is already being conditioned for women leaving the workplace and society reverting to a pre-WW2 social structure. That's a solution. Jobs are lost forever, women are out of the workforce and the number of workers is halved. We all live in a trad-conservative paradise.
Before they existed, few would have believed the disappearing jobs would be replaced by web designers, uber drivers, youtubers, drone pilots etc.
But the sensible thing is to expect strong trends to continue.
And yes, it is a real science!
what about rules/laws which allow predictions?
Maybe one day unemployed people will be truly useless to other people, but for the last 250 years, new professions have always emerged.
Free market capitalism lacks anything approaching a clear and exhaustive definition and the term is used for many arrangements - from market oligarchy to market anarchy but there is one thing that is absolutely certain about it:
"Past performance is no guarantee of future results"
In effect, the most enduring tenet of free market capitalism negates your argument. You need to provide material evidence, not glorious promises of some unknowable bright future, North Korean style.
The key point here is that "create value" -- new jobs are created if there are new human jobs that create value.
The bifurcation of incomes pre-AI should be a warning that we're reaching a point where the average human (in physical ability and knowledge) can no longer efficiently create value.
Imagine we have robots that can physically lift and manipulate as well as the average human. And AGI that can think as well as the average human (arguably LLMs might already pass that low bar).
Should an AGI future eventuate, the distribution of that abundance is not pre-ordained. Retooling government and society into a post-capitalist system, and deciding a fair distribution to people who have little power to demand it, is going to be a mess.
Or to put it another way, how many times in history have the rich en masse voluntarily given to the poor?
IMHO, the value will be in creating f2f human interactions — stuff that AI and robots will only be able to simulate but never replicate.
That may sound flippant, but real human interactions are already being sought after (relatively) widely, and I think that this trend will continue.
The masses will always be the masses, the elite will always be the elite, and there was a small anomalous blip in this trend in the mid- to late-20th century that has dribbled over into the early 21st century.
The vibe in the movie Titanic for how the upper decks and lower decks lived pre-collision (and to some extent post-collusion) is pretty much how I imagine life will be in an AI/robot-powered future.
[Note that the Titanic example can be nitpicked, but the upper deck people were shown to be having stuffy fun in a relatively posh setting, while the lower deck people were letting loose in a very relaxed and social way. That is my point in mentioning it.]
A superintelligent AI could meet all human needs. The only remaining jobs might be those where people prefer a human touch like therapy or sex work, but such roles would be rare.
Past revolutions either made work more efficient or created new ways for humans to serve each other. Superintelligence would replace every form of work that meets human needs, so historical analogies don’t apply.
Even with current technology, there are already almost fully automated POCs of farms and so called “dark" factories and we are talking about superintelligence, not LLMs.
Even in small rural areas the demand for locally produced sustainable and consciously produced agricultural products is largely unmet.
Funny thing is, it wasn't agricultural advances, at least initially. It was chemistry and mechanics. Haber-Bosch process of ammonia synthesis created cheap fertilizer, mechanization dramatically increased productivity. New plant varieties with higher yields came later.
It's very counterintuitive, but for 250 years tech has constantly eliminated jobs, while unemployment rates have stayed the same.
People have always predicted this would lead to mass unemployment, they've always been wrong, and yet they've kept predicting it.
One way to explain it is that unemployed people are an unused resource, and free market economies are very good at finding uses for those.
Two examples out of plenty others, where people had to move out as means of survival.
unions kept wages high and led to offshoring, but inefficient garbage tier cars from Detroit lost ground to better engineering and no planned obsolesce coming from European and Japanese models. A lot of that came from a decline in steel, coal, and acute pain points related to the oil crises in the 70s.
I'm joking, but not. I think a lot of fat is going to get trimmed of the bone of most industries. That may include myself which is a worry.
With this much shoddy code out there, I'm expecting a LOT more testers are going to be needed. This is job creation but not the good kind.
Think Tesla Optimus robot grabbing popcorn. Instead of one cashier filling bags of popcorn, you now have a dedicated teleoperator + sweeper to pickup all the popcorn that's spilled. One job turned into two jobs.
Perhaps AI will finally raise the bar so high lower-IQ individuals won't be able to find any meaningful employment, but this has never happened in before and I doubt it'll happen again. I don't think I've seen a respected economist go on a news broadcast and say AI will lead to mass unemployment.
Baby steps, I guess.
Is it ?
> when something makes the economy more efficient
How does "ai" make "the economy" more efficient exactly ?
> people get new, better jobs
Which law dictate that these jobs are better ?
I don't see people getting better jobs personally, I see a shit load of people pushed into more and more precarious jobs, with less and less workers rights and job security, with stagnating wages despite rise in productivity, &c.
Yes, if you look up any video of an economist talking about AI, they'll bring up the industrial revolution and point out how people thought it would bring the end of employment and instead created more and better jobs. The generally accepted viewpoint is AI will make things more efficient, and thus people will be freed to do other, more important work.
In a post-labor-scarcity world, there is no longer any "other, more important work."
Which means there's no basis for paying labor currency, which means there's no way for labor to buy things, which means we need a new method to ensure people can get things.
99% of economists couldn't see any of the big downturns up to 1hr before shit hit the fan. Yet they all retrospectively can upload 12 hours long video on youtube to explain you how the dot com or 2008 crisis were so obvious you'd have to be blind to not have seen them coming...
> and thus people will be freed to do other, more important work.
"freed" is a weird word to employ, given the century+ long race to automation shouldn't we all have been freed by now ? Why do so many people work min wage or other type of precarious jobs ? Why will I retire later than my parents and grand parents ?
And what do you mean by "more important" ? Most jobs are bullshit and workers know it, they don't want important jobs, they want stables jobs that pay enough to live life, and "ai" is going to do anything but improve that
Still doesn’t explain how new jobs are created. Efficient economy doesn’t mean more jobs. You could displace a thousand workers and create a single new better job, but that means nothing.
For a simple model, let's say you hire programmers for three reasons:
1. Because you have (X) work to get done to run your business. Once that work is done, there is no more work to do.
2. Because they get work done that makes more money than you pay them, but with diminishing marginal returns. So the first programmer is worth 20x their salary, the 20th is worth 1.01x their salary.
3. Because you have some new idea to build, and have enough capital to gamble on it. If it succeeds before you run out of money, you'll revert to (1) or (2).
Let's assume AI comes along that means a programmer can do 4x the work. If most programmers are in the first bucket, then you only need 1/4 as many programmers and most will be fired.
If most programmers are in the second bucket, then suddenly there's _much_ more stuff that can be built (and money made) per-programmer. So businesses will be incentivized to hire many more programmers.
For programmers in the third bucket, our AI makes more likely to get built in time, thus ups the odds of success a little.
How you think the market is structured decides how you think AI will effect job creation and destruction.
> If most programmers are in the second bucket, then suddenly there's _much_ more stuff that can be built (and money made) per-programmer. So businesses will be incentivized to hire many more programmers.
How do these two reconcile? If hiring more programmers is diminishing returns, why would a business hire more?
That said, I'm sort of assuming simple tasks will get done with AI, which isn't a given and is something I could easily be wrong about. AI could easily just benefit everyone, as has been the case for every prior technological advancement in human history
"Meaningful" is a bit weasely here. If a skilled factory worker had their job offshored in the past and wound up employed at Walmart after, they did not find "meaningful" work
It has happened plenty of times
Geoffrey Hinton seems to be in the headlines constantly going "godfather of AI" and giving some random obtuse warning or call for action.
Eric Schmidt also spends a lot of time yapping.
And of course people like Gary Marcus.
Simon Willison also does a lot of writing but at least his blog is actually useful information of actual tests and news.
I would not include simonw with the other 3 at all. His opinions are informed by use instead of just other thoughts. Simon also releases a lot of utility tools.
It's pretty wild.
One team to re-do the work to double check the right answer. The second team to reconcile the right answer with the AI result.
I'm not even joking ... I've professionally been extremely embarrassed once by an AI result. Now I check it so often that I might as well just do the work I asked it to do.
No doubt, quick questions and rough ideas, LLM is the bomb.
Professionally? Nah. Not yet.
It's easier to start the next widget company because building widgets with the technology is cheaper.
It's easier to consume other things because goods are cheaper to make with the tech.
A third option is that the tech enables something all together new, eg television, that starts a new industry.
As far as direct job creation, the third way is the most obvious but probably not the case at the moment. So I guess we're stuck waiting for goods and services made with AI to get cheaper.
As amazing as LLMs are, the statistical unreliability of the SOTA models means human intervention is always necessary. That's based on today's SOTA models which are already so absurdly gigantic that the DCs are disrupting civic infrastructure due to their water and power demands, and these companies still have to rate limit aggressively just to keep the service up.
Everyone knows quadratic growth doesn't scale, so I don't see how these models continue to grow in capability, while also growing in capacity to meet the demand of an "agentic economy", while still ultimately needing to pay human SMEs to verify correctness and fix mistakes. It doesn't add up.
My prediction is that we'll see increased efficiency in knowledge workers and reshaping/consolidation of certain job functions, but the structure of the economy over the next two decades won't be upended due to LLMs.
If you're a business that offers 3 services to the public and "AI" could automate one person's salary out of your payroll, why would do that as opposed to keeping your headcount and instead offer 9 services to the public generating more streams of revenue?
> If you're a business that offers 3 services to the public and "AI" could automate one person's salary out of your payroll, why would do that as opposed to keeping your headcount and instead offer 9 services to the public generating more streams of revenue?
If AI could automate one person's job, why would you keep him? You can offer 9 services with one less person.
It created dump truck driver / maintenance / engineer / safety / fuel supply / parts factoring / etc jobs. I wouldn't say that's an example of creating zero jobs.
Those are primarily intellectual/service jobs.
> I wouldn't say that's an example of creating zero jobs.
I never said it didn't create zero jobs. Reread my comment. I said it created more intellectual/service jobs. It's the first line of my comment.
> Automation has already created more job opportunities, why should this be any different? Jobs don't go away, they just transform
Which seemed to be what you're arguing against at the same time.
Similarly, software is a crucial input to pretty much everything in modern society. More / better software would create tons of new jobs.
I'm not so sure that more worse software will create those jobs though, and that seems to be the direction we're moving with AI.
Sure, with far less human labor. That's my point.
> That road system directly and indirectly created millions of jobs.
Yes. That's my point. It created millions of service/intellectual jobs. Or are you saying the highway system created more jobs for those who own shovels?
If people moved from physical to intellectual labor in the past century. What will people move into if intellectual work is mostly done by AI. Assuming that the AI the evangelicals promise comes to fruition.
This is the "the climate has changed before" argument for economics.
> why would do that as opposed to keeping your headcount and instead offer 9 services to the public generating more streams of revenue?
Because there might not be enough demand for those 9 services.
Funny how all these boosters bragging about automating the economy seem to conveniently forget how much of the economy, in some way, is based around headcount and consumption - both of which are going down due to AI with no replacement being pitched, almost as if that were the plan all along.
It's really not, you cannot cause a recession by increasing productivity, money doesn't disappear.
Money cannot disappear, but circulation can be confined to ever-smaller segments of the economy. Let's assume a world in which AI is good enough to do 90% if all white-collar jobs and companies have replaced positions with AI accordingly. Then the amount of money that entered the wider economy through salaries will now stay locked up in B2B transactions between AI companies, their corporate customers and their contractors. How will all the laid-off people get money? Some would probably get into blue-collar work, others will try their hand starting a business (that will also compete with AI), but all of them? What will happen to the businesses that were relying on those employees as customers?
Yes, that's exactly what will happen and you can even now see examples of unemployed software engineer building successful one-man gigs with the help of AI(not competing).
I think the concept of stable long term jobs as we knew is gone. Which might not be a bad thing.
I think it's a very different question to ask "how would this create jobs" and "will this create jobs...in today's market."
Because cost cutting increases share-holder value. See Jack Welch at GE. Over time he destroyed the company but also raised the share price tremendously.
The simpler way to view this is that AI creates value. We can argue how much and what kind, but our economy is reliant on ever-increasing value and is pretty good at utilizing it.
Microsoft didn't get smaller when computers got better, their market got much bigger.
Maybe, just maybe, we could slow things down, profit from the time saved instead of trying to squeeze more out of every fucking thing we put our hands on.
Seems like a current trend.
The idea that a massive tax increase (aka "tariffs") will create jobs is also 100% crap.
1. Say it will protect children
2. Say it will create jobs
But now I put it like that, it's a little strange I've not heard of any politicians even implying that their policies will get people more sex…
(Especially as I did say "even implying")
There's always the chance that "this time it's different". What's left after machines can think better than people? Dexterous manual work that mechanical arms can't perform?
The ideal situation would be that we reap the benefits of technological abundance and work less.
What happened in the past wasn't magic. It was logical progression from a very basic civilization to an optimized one. But AI advocates push the past as 'everything changed but magic made it work out so that will happen again'. The past’s transformation was a one-off fueled by obvious, unexploited gains. There is no “next level” waiting in clothes, shipping, energy, or most other mature systems.
You get the idea. This is bad, it’s always going to be bad, and the only fix is a fundamental reorientation of global civilization away from “profit at all costs” and towards better metrics that reflect the health and wealth of a populace than GDP. Doing that would require existing power barons cede their control and cooperate on building such a new system, which they emphatically do not want to do.
We’re fucked, basically, has been our position all along. Either AI via LLMs is the real deal and the economy collapses because corporate and political leaders are too stupid to understand macroeconomics and labor issues, or AI via LLMs is bullshit and the economy collapses because we let the rich and powerful light money on fire for years in pursuit of a boondoggle.
There is no good ending here under current incentives and motives, and not one AI company has actually bothered to challenge or change those so that AI becomes a positive force.
They can stay irrational longer than we can stay above water
Make friends. Start a game night or hackathon with friends. Pick up new skills you’ve been putting off. Finding personal meaning outside of work is the key to surviving times like these.
Looking at the product, the idea that this would take 350 developers is likely complete nonsense. It also indicates that the recent big tech layoffs and hiring reductions are simply removing bloat, not AI driven.
Seems like something a senior could build a prototype in a week and a product in a month.
Perhaps the "AI matchmaking" feature would take a lot of resources not done with AI, but that goes toward the point of AI creating things not possible before, not replacing existing development
The risk is more that the quantity is less than what is removed. And that the people losing jobs aren’t suitable to retrain. Retraining say truck drivers as data scientists is an uphill battle.
ie substantial risk of specific people getting left behind
That's why I'm thinking about my new SAAS, revolution as a service! (Because this is HN I have to shill a product)
There's almost no question that tech giants will have negative headcount over the next 15 years but the changes coming from AI haven't even really started to percolate into the broader economy.
Eventually, at some point. Most industry would have been disrupted. There should be a working group of the major players/stake holders in the space to figure out how this is implemented.
I work in tech and I don't see that happening for another decade and half or so.
Where is this discussion thread?
└── Dey well