I'm going to make an anti-capitalist argument in order to make a pro-capitalist argument.
The anti-capitalist argument is that the wealthiest people in the world are not dramatically more intelligent than the average person. Bill Gates is obviously very intelligent and capable, but his IQ does not scale with his wealth. He built the right company at the right time, and Microsoft ended up sitting at the center of massive network effects. Relatively modest advantages in judgment, skill, and industriousness can end up producing enormous differences in wealth due to the compounding effects of success.
That's just how scalable markets work. So Gates's success, while evidence that he was very effective at building Microsoft, is not evidence that he is vastly more knowledgeable than everybody else about economics or the future of technology.
And the same applies to Bezos, Musk, and every other extremely successful founder.
Their arguments should be judged on their merits and not treated as infallible insights from people with superhuman intelligence.
Now, that is not to say that I want their wealth to be capped or politically reallocated. The case for private property depends on the argument that decentralized ownership and investment and capital risking is better than politicians allocating a pool of resources to gain clout or win over popular opinion. A wealth cap or high taxes on the wealthy would transfer control over resources from private owners to political institutions. So what we would be doing is replacing imperfect private allocation with imperfect public allocation, where the people allocating the capital have limited information, political incentives, and the ability to impose costs on others without their consent. So I reject blind deference to billionaires, but I do not want politicians to control billionaires' capital.
Anyway, let me address Gates's argument about AI:
Gates argues that AI will disrupt the economy faster than previous technologies, and therefore stronger government coordination is needed to address the rapidity of the speed of that transition, to cushion people against the speed of that transition. But what his analysis underweights is how quickly AI can help people adapt to the very transformation that AI brings about.
For example, a worker taking on a new position now has access to a tutor and an analyst and a business advisor all wrapped up in one, who can provide world-class assistance 24/7 at the cost of a few cents in electricity. And someone who wants to start a business can employ dozens of AI agents at a cost of a few hundred dollars to test out the idea and build out a prototype within a few days. So what this means is that we cannot hold the human adaptive capacity constant while trying to determine the impact of AI's disruptive capacity.
The second problem with Gates's argument is that he only counts the income that people might lose when AI replaces jobs, but he doesn't count the fall in prices that the same AI will bring about. The fall in prices is effectively an increase in purchasing power. That increase in purchasing power, in economic terms, means an increase in real wages, and that's exactly why wages increased twentyfold between 1826 and 2026. Automation brought about abundance, which reduced prices and thereby increased purchasing power, i.e. increased real wages. If you look at tokens right now, their prices are going down exponentially. They've gone down about a thousandfold over the last four years, while the capability of the tokens themselves has increased. So AI is following exactly the same price trends as every other previous computer technology.
If AI becomes a major input into medicine and education and administration and logistics, that means that all these goods and services become radically cheaper, which means that paychecks can buy a lot more of them. So that's, I think, a huge hole in his analysis. It's also, I think, very simplistic to assume that the future will be based...
> For example, a worker taking on a new position now has access to a tutor and an analyst and a business advisor all wrapped up in one
These llm agents don’t have “agency”. No intrinsic goal other than respond. A tutor has judgment and agency. You will get a really good pattern match on the input text you give but you won’t get tutoring.
Maybe you're right, but I've never learned as much and as quickly as I'm learning now with the assistance of LLMs. It's like a hyper-powered search engine. It not only knows what I'm looking for, it knows the context of why I want to know it and how best to convey it to me.
Regarding your last paragraph (And as long as...) Bill Gates convincingly argues that the traits that make AI do good things are not easily separable from those that make it do bad things. So it may not be possible to explicitly avoid an AI that does bad things.
Gates is talking about dual-use capability (e.g. designing a medicine requiring same capabilities that allow it to design a pathogen). This is a real threat, but not a threat to the human role in the economy, which is one of being an independent strategic actor.
"Now, that is not to say that I want their wealth to be capped or politically reallocated." Why not? It's done for wages, why should wealth be any different? Besides the technicalities, which you are not talking about and therefore I will also not get into them, I don't see any ethical reason why wealth should not suffer the same biases that wage taxation does. This kind of opinion just perpetuates the imbalances in the current terrible lop-sided system we have.
I think to the extent that the system is lopsided, it is in creating fixed costs to participating in investment, which naturally leads to the investment market being dominated by the wealthy.
The token sale phenomenon briefly reversed this, but that was shut down:
>The average ICO has almost 4700 contributors. The median contributor invests a relatively small amount. The ICO market appears to have successfully given access to the financing of innovation to a new class of investors, which is a long-standing public policy issue
Even public capital markets are becoming increasingly displaced by private equity where only institutional and accredited (wealthy) investors have access as investor:
> If someone had a credible plan for slowing down AI advances globally, I would likely support it. However, I don’t think that’s going to happen. The geopolitical and economic incentives are pushing too hard to go full speed ahead.
> Research suggests that in some parts of the United States, factory closures contribute to a rise in deaths from opioid overdoses.
It’s so cute some people think that the extent of mass displacement will be limited to what happens to the unemployed. Oh poor unemployed! Nope, pitchforks will be grabbed. It won’t be rosy. Maybe they dream about having killer robots to defend themselves, but imagine a few hundred million people getting angry within one election cycle.
China has both carrot and stick. Living standards there have improved dramatically over the past few decades. Over here they use a smaller stick, but no carrot at all.
My sense from talking to people who’ve lived in both countries is that there’s also a perception that there’s at least a semblance of fairness in China, which can make it relatively more palatable. Chinese acquaintances are rather horrified by how, in the US, cops can use state surveillance systems to stalk women and get away with it and people regularly end up homeless due to medical bills.
I have a lot of Sinosphere friends and I strongly disagree with this. Much like the US I think there is a very nuanced view of Chinese authorities, but there's plenty of overreach there as well. The zero covid policy was not that long ago. I have friends who were stuck in hotels for weeks. Relatives who were stuck in the US for months.
> Chinese acquaintances are rather horrified by how, in the US, cops can use state surveillance systems to stalk women and get away with it and people regularly end up homeless due to medical bills.
This sounds more a product of censorship rather than China being particularly better in those aspects. Police in China are less likely to kill you because guns are uncommon, but they are just as shady if not more, especially in more remote areas. Healthcare access and affordability are also worse than what you'd expect from a communist country. Homelessness in China is definitely a lot worse than what state numbers suggest. They just drive out homeless people who they're less visible.
False equivalence. Maybe that’s what they fool themselves with. Salivating, looking at China, not understanding the massive upward social mobility in recent time there. Why stop at China, look at Saudi Arabia, Qatar, etc.
The social credit system is regional, disjoint, and composed of a handful of various trackers. It can hardly be compared with public opinion.
Xinjiang is an exception; it is heavily publicly surveilled.
Internet censorship is strong. The public still has sway over government policy. When search terms are blocked, the public continues with euphemisms. The government is responsive to public opinion. The government does promote influencers with opinions favorable with CCP goals.
Chinese foreign policy is more dovish than public opinion, especially regarding Taiwan.
We have also to consider what if their voices not only don't matter but are actively being suppressed. It's hard to start a revolution if who's in control can control communications and can preemptively identify people who are more likely to start dissent. Maybe this time we won't even have the slightest chance to revolt.
Said democracy does have to be real, though. If we're talking about that country between two oceans that waged armed conflicts almost constantly since its very inception, has supported or helped put in place straight-up dictatorships not that long ago, and sports one of the highest wealth inequality in the West, well... I'd think long and hard before I call that a real democracy.
Also: The US government is structured to put checks on democracy thereby ensuring wealth and power inequality. See: The Electoral College, The Senate, and even the Constitution itself among other examples.
Not sarcasm. For example: The Senate was originally not directly elected. Furthermore Wyoming and California both get 2 senators each. And there’s the filibuster on top of that.
Everyone seems to be forgetting that the US did have a pitchfork event, it was called Jan 6th and might best be characterized as a coup d'etat that fizzled with remarkably few casualties.
>>>> imagine a few hundred million people getting angry within one election cycle
If that many people get angry, you can bet they're all going to be different people with different motivations. This is partly why revolutions tend to suck no matter how evil the regime was they overthrew; the US got extremely lucky with being able to postpone the post-revolution civil war for a century.
During the economic transformation in the former Soviet Bloc, entire industries disappeared, and pitchforks weren't grabbed. I saw it myself, growing up in a heavy industry city that quickly turned into a rust belt city.
IDK why so many people romanticize pauper riots and try to wish them into existence. In practice, as long as people don't outright starve, they don't become violent just over losing their jobs.
All the post-1989 wars east of the former Iron Curtain were triggered by nationalism. Not a single one by a class conflict.
It's cathartic to feel that your views are correct but simply suppressed by the establishment. You don't simply have a political opinion, you are ultimately correct, and the true absolution of revolution will make your opinion manifest. My family is from a developing country so the particulars are different, but yes people don't just descend into violent revolution because they lose their jobs. Unemployment is indeed a factor in revolution but is one of many.
It's one of those unfortunate aspects of internet information silos where people repeat this phrase to each other but don't really contextualize their opinions in the greater body of history and political thought.
Additionally, I think a lot of people with this opinion don't realize the vast difference in material standards we have today and what was true in the age of Communist revolution and decolonialization. It's tempting to think of material standards as relative and I think inequality is indeed a destabilizing factor in socities, but minimum material standards are true in so much of the world today that I think a class-conscious revolution is much harder in most of the developed and developing world absent other external political forces (e.g. nationalism, decolonoialism, etc)
Judging from your username you were from a country that ended up integrated into the EU. Good for you. But in Russia you can draw a direct line between the economical collapse and drop in living standards of the 90s and the rise of Putin as the big man who will put an end to the disruption.
But this still wasn't case of pitchforks. It was "just" an election, and the Putin regime, for the next 10 years since his election, wasn't even particularly grim.
> During the economic transformation in the former Soviet Bloc, entire industries disappeared, and pitchforks weren't grabbed
Wouldn't it be easy to argue that this is because there was more a social safety net to prevent said starvation and by that point people were extremely cynical about revolutionary ideas in a way that is not at all comparable?
The social safety net of a relatively poor country is not really better than the current social safety net in the US, because of the overall lower economic level that limits the amount of taxes collected.
Looking at the "hollers" in the Appalachia, the situation is quite comparable to that of the rust belts of the 1990s in Central and Eastern Europe. You can buy groceries from welfare money, so you don't starve. Young people escape, old people sit and drink (or worse), but don't form a violent mob.
This is how economic decline usually looks like, including in the US. Reminds you of the poem about how the world does not end with a bang, but with a whimper.
The social safety net was precisely the thing which had collapsed! There were basically two options: scrabble around trying to find work in the newly evolving private sector to stay alive, or give in to despair. There was a huge drop in the life expectancy of Russian men after the end of the USSR, which took a while to climb back up again.
Pitchforking really requires a clearly identifiable villain. You can't fight a "system". In fact it usually requires the "system" to have turned on its leader. See the death of Ceaucescu for an example.
I wouldn't rule out the possibility of Amerimaidan in the coming years if the country doesn't reverse course, but it won't be over primarily economic issues. it will be against paramilitary policing.
We did grab pitchforks further East, in Romania, they were called Mineriads (the last one took place in early 1999 [1]), it's only that the Westernized elites painted that as barbarism (i.e. the miners were seen as barbarians) versus them, the Westernized elites, and with Western money on their side said elites prevailed in the end.
I think he's literally conceding your point by writing this essay. That there will be massive upheaval and yes maybe pitchforks if we don't get ahead of the problems.
I can imagine "a few hundred million people getting angry within one election cycle" but the question is, what's the probability of that happening? I would say given current evidence, close to zero. Modest productivity gains from AI are expected, but the current projection in hiring reduction is less than 1%. And that's as reported by a minority of executives, so it's likely way too optimistic.
>More than 90 percent of executives report no effect of AI use on employment over the past three years, and 89 percent report no impact on labor productivity. Among the minority that do report effects, the results tilt slightly positive for productivity, with an estimated average boost of 0.29 percent. However, executives anticipate much larger effects going forward. They forecast that over the next three years, AI will increase labor productivity by 1.4 percent on average while reducing employment by 0.7 percent, implying a net output gain of roughly 0.8 percent. About two-thirds of the anticipated workforce reduction is expected to come from reduced hiring.
The idea that AI will transform the economy overnight is wishful thinking. Productivity gains will be gradual over the next few decades and will likely plateau at some point.
Ironically, by predicting the most chaotic outcome you're falling for the narrative that AI will transform the economy overnight, which is exactly what AI companies' marketing divisions want you to believe.
You're using a rhetorical device to imply that they're "falling for" a narrative. "Falling for" only follows if that's why they are making the prediction; this construction doesn't allow that they may have come to this belief on their own.
Based on what evidence? The unemployment rate is very low, and has been for a long time. Why should we take your word for it when it blatantly contradicts real economic data?
What unemployment?? The unemployment rate right now is by all reasonable standards very low. Do you have any reason to believe unemployment will suddenly skyrocket? As a result of what factors? I hate to say it, but your analysis here seems based on vibes and not much else.
I'm kind of surprised, but not surprised, by Gates' comments here. Until very recently, apparently, he was a pretty strong techno optimist, now toeing the line of a doomer. The verbiage is substantially in line with the fear-based marketing out of anthropic. I wonder what motive is underneath all of this. Trying to influence policy? But he has so many resources, it transcends policy.
i define the hacker news "ratio" as number of upvotes / number of comments. lots of comments and few upvotes is a bad ratio indicating the article is not being well received.
in this case a ratio of 196 points to 479 comments points to a very poorly received yet comment-inspiring article.
Their suggestion to tax tokens extends the problem of taxing wages to a new type of production. The real solution is, of course, taxing the capital, but you will never hear that from them.
A guy that became rich from a plagiarising another product, stayed rich by being anti-competitive and then raped a few minors for good measure is now worried about equity. Get bent.
Does anyone really believe that Bill Gates would propose any kind of truly altruistic plan, that is to say, would he ever support any AI regulation plan that actually hurts his bottom line at the end of the day? I don't think that AI-adjacent CEOs are really the most trustworthy people on this topic.
The title reminds me about the VB6 days and drag and drop controls and efficiency until you get a couple years out. Then the vendors providing the products you used are out of business. That may happen a lot more often and quickly with models, but I have read that they are pretty much just a drop in component as the old one has a superior replacement, you move on and benefit. Sure are interesting times. I want my models, local, small and powerful. Vision and reasoning. My goodness the power we already have is just going to get better.
> Countries need cross-agency bodies, and an international organization must be built in parallel — combining elements of nuclear inspections, aviation regulation, and ozone treaties, including US–China cooperation
I see, so force china to cooperate or label them as terrorists in the same vein of irak/iran nuclear capabilities
> The biggest shift for workers will happen when AI provides nearly error-free work. At that point, it will be able to function on its own without a human checking in on it, and companies will have every economic incentive to let it.
This is a very succint argument that has a technical implication - that correcting errors and misalignment is incredibly important for autonomous AI capabilities in a fundamental way.
Unfortunately, this is essentially high level clickbait. At the beginning he immediately portrays the issue as black and white:
>AI will either be the greatest equalizer ever invented, or the worst source of injustice.
In all likelihood it will probably fall in the middle, with the likelihood that it will tilt power and money towards the already ultra wealthy (to the surprise of no one). But it will enable ordinary people and small businesses to do cool shit and make their lives better.
I don’t see any reason this highly chaotic system is most likely to settle in a spot that we happen to enjoy/see as a compromise of some sort.
The point is that we now have superintelligence; either it helps us conquer death, or it poses a massive —say, existential- risk to us all. It’s hard to imagine such a powerful tool resulting in just another century of slightly-better capitalist nationalism.
I guess there could be a second AI winter around the corner that we haven’t seen coming yet. That would be nice! Not likely IMHO, but nice.
I'm still not seeing the route to that existential risk, not yet anyway.
Far more imminent is the very real risk of unbounded greed and power that wields those AIs, in full swing as we watch: unchecked global warming, pollution, poverty, authoritarianism, plummeting biodiversity, and more.
The corporations are like out of control AIs themselves, that will do anything to make their paperclips.
> The corporations are like out of control AIs themselves, that will do anything to make their paperclips.
Ideas manipulate humans like puppets. They have a life of their own and the evolutionary context within which ideas live is human culture which they have instigated and change, but also in grey matter, which itself is the product of DNA which is really just information in physical form. It’s information all the ways down.
Absolutely happening now, but that's not caused by AI though. AI is not choosing that.
The corporations are wielding AI irresponsibly and choosing a massive build-out speculating on possible future gains. They did that for crypto too, if you remember them buying a fossil plant just to drive a mining farm.
> but that's not caused by AI though. AI is not choosing that
AI can't not "choose" that, because AI is dependent on consuming vast amounts of electricity, and everything follows from that.
> The corporations are wielding AI irresponsibly
This is the most likely route to "AI apocalypse", yes. Humans are neither going to give up control to AI, nor give up the pursuit of profit for the long term survival of others.
The distinction won't really matter as to whether it's "corporation" or "AI", if it's an AI-wielding corporation.
There’s a lot to discuss here, but I think the point most likely to hit home is one based in a) expertise and b) the nature of frontier labs. Specifically:
Would you know what a system that was capable of recursive self improvement would look like 2 months before it happens? Even 1 day? We don’t even know if it’s possible or what natural plateaus might exist for a DL-based system that rapidly evolves in su h a manner, so it feels hard to justify the other end, too (that it’s definitely not possible anytime soon). If anyone could do this, wouldn’t it be the scientists who specialize in this field? Cause they’re raising alarms in mass.
Assuming you make it past that first question with a “yes” and a “I’m close enough”: the huge amount of information assymetry and proven tendenency to hide safety concerns from the big labs is enough all on its own, IMHO.
Who knows what their most experimental, risky project is at any given time? Not us :(
> either it helps us conquer death, or it poses a massive —say, existential- risk to us all
Lesson of 2026: there is no "we". Hobbes is dead but still looming over us.
The risk is that it helps some people conquer death, while posing an existential risk to some people. The US has enough problems with its gerontocracy. There has only ever been one president born after 1946. What if you were right and there never was another one born after 1946? Propped up by an immortal Supreme Court on "life" terms, of course.
Life extension is also further fatal to birth rates, of course. I suppose the worst possible outcome is technology that extends life but not working life, so you get a larger and larger population of people too old to look after themselves.
I don’t question your portrayal of the worst possible outcomes — certainly victory over senescence could send us down all sorts of dysfunctional local minima. But I encourage you to try to intentionally invest in thinking through some nice ones, too!
That’s my main and primary point, to be clear: hope is essential to human flourishing, and it admits no justification by design, unlike all other beliefs. But to comment on a few particulars, on the off chance you/lurkers ever scroll back over this convo:
1. Surely you’d agree there’s a “we” in the sense of communal risk, yes? Obviously I’m thinking about xrisk again, but this applies equally as well to threats we all agree are grave, like climate change, pandemics, nuclear war, and hell, even economic adversity.
2. I will admit that I’m missing the particular Hobbes reference, so hopefully that’s not key lol. I know him for the social contract, primarily — are you emphasizing his deadness in order to imply that we no longer abide by said contract? Cause if so… we have problems, but the day to day seems pretty darn regulated still, IMHO.
3. I don’t think there’s anything necessarily wrong with birthrates we would consider low, but u also don’t agree that longevity would necessarily lead us there — a lot of people would probably still have kids in their youth, reveling all along in the fact that they’ll have decades to centuries to share with their adult children (if they want) instead of somewhere around, what… ~25y, on average? I certainly am not having kids now because I want to prioritize my youthful energy on my work, friends, and learning — knowing that I had time on the other side would reverse my personal decision 100%.
4. I think the concerns about technologically-enabled hyper chronic bigotry that you mention are overblown, no matter how many times the honourable Prof. Weinersmith argues the opposite in his comics. I’ve found cogsci, neuropsych, and philmind have all come to share a deep suspicion of the parts of us we consider continuous, persistent, and unified — an existentially confounding idea ofc, but also one that backs up ~~an old adage~~ a line from a recent sketch comedy show: ’People can change.’
Apologies for the rants. And I do feel like your latter description of a possible world would be an incredible setting for a biting scifi satire lol.
But in the end: I hope you can find a way to join me in the wilfully-hopeful camp, even if it’s not today or even this year. We have just as much science!
It reads like that kind of AI marketing which focuses on how "scary" a new model supposedly is.
Bill Gates' fear is clearly that everybody thinks this stuff is just kind of mediocre because then the real horror story unfolds - Microsoft stock price collapses into blue chip territory where it firmly belongs.
IMO, this is a highly bimodal probability distribution.
I model this as two very strong attractor states (radical abundance and totalitarian power consumption). I believe the latter is way more likely without a concerted effort that we currently seem incapable of.
The “somewhere in the middle” outcomes are in my opinion are not AGI-pilled, and only come about if the exponential dies hard in the next year or maybe two - as in, it’s a very specific timing prediction that needs to come true for this scenario to occur.
Why only scenarios where humans remain in control? I think the kind of AGI needed for radical abundance would be difficult to achieve in a way that also does not lead to ASI
If we get effective AI, humans aren't going to remain in charge -- that much is inevitable. If you had a governance machine that made your country perform 10% better than other countries, so long as you deferred all decisions to it, you'd defer those decisions or get left behind. And AI probably has the potential to do far better than 10%.
Either everyone, globally, gives up on AI development, or AI is going to be in charge of all human development, and we'll just be along for the ride.
Right now, the work ahead is to shape AI so that the decision making it does as it carries us along is something we like.
No, because if this were the case, the UK wouldn’t have royals, North Korean wouldn’t be a dictatorship, China wouldn’t have the CCP, and America wouldn’t be a two party state.
Hardly any existing polity would willingly hand over control to an AI.
More likely an AI would overwhelm human polities through endless hacking and warfare.
You're assuming a handover happens in a single definable moment. More realistically it will be one small decision making power at a time, so that at any given moment it never really feels like control is being given up. Boiled frog.
I mostly agree, though there are versions of ASI that I believe leave humans “in charge”.
Most of my probability mass for good outcomes is around some variant of “guardian angels” or “benign machine god”. Banks’s Culture novels being on the realistic end of that range. “Keep humans around because the world is more interesting with them in it”.
But it’s possible that we might manage to construct some entity that really loves humans and wants to shepherd them while leaving them in control. (Perhaps declining to indulge the most self-destructive tendencies.) This feels like the very hardest alignment path to engineer.
The other branches of course are post-humanism; human uploads and “merging” with AI, whatever that means. Or post-humans augmented to have intelligence sufficient to eschew AI and take back their destiny.
> If you had a governance machine that made your country perform 10% better than other countries, so long as you deferred all decisions to it, you'd defer those decisions or get left behind.
This was basically the invention of bureaucracy by the Qin dynasty. Despite delegating decisions to a "machine" made out of people, the Emperor remained in charge.
So it's an exponential with a negative exponent? The thing about those kinds of curves is that since they have asymptotes to zero, it's hard to tell when you've plateaued.
I agree with you from a mathematical point of view. You have this "dead economy theory"[0]. I spent a lot of time trying to model this theory, enhance the theory (lot of good HN comments) and find a "middle ground" path. My saying is that you end-up on an optimization saddle point where it is really hard to not fall into concentration of power. I was able to get radical abundance only at the cost of pretty hard regulations.
I agree there are some extremely bad outcomes from fast labor replacement.
I don’t think it’s inevitable that the economy collapses though. The problem is that this model assumes that capital needs labor, but soon it may not.
The easiest model is - assume we get robot laborers in the next few years. The AI firms that are displacing humans still need services, most of these will be purchased from other AI/robot firms, and so while prices will swing around dramatically, there is no reason the engine has to stop. (Depending on what the AIs want; you can assume that more compute and robots will be part of it.)
I do agree that unbridled capitalism is mostly incompatible with human flourishing in this world and in any event you’ll see a lot more regulation; made feasible by how good AIs are at navigating huge Byzantine rulebooks. I view the techno-feudalist capitalist endgame as fairly unstable due to consolidation pressures, though I guess once we have nuclear-armed megacorps they might use MAD to prevent further mergers.
I don't think that's necessarily unlikely, but people making this argument rarely seem to understand just how radical a change is required to make that a possibility.
e.g most small business rent their premises, and buy goods and services from other tenant small businesses, who buy their raw materials from large landowners. How can a small business survive when all the land is monopolized by the technocrats? All value you add gets taken away as rent. This is why they always talk about taxes on outputs (tokens) instead of inputs (land, energy usage).
It's not an ungrounded hypothetical, that's exactly how society worked for hundreds (thousands?) of years.
isn't "making lives better" basically the reason for pretty much any companies existence?
arguable of course whise lives are meant to become better, but if anything ai just moves the goalpost
> In all likelihood it will probably fall in the middle
AI era has been called the new industrial revolution. I'd just like to point out that the industrial revolution created the British Empire. That's wasn't exactly "in the middle".
Maybe they'll build AI-powered robot children for their entertainment? Or perhaps figure out a way to control kids bodies with AI, and fulfill even more of their fantasies. Hey, it's a driver of progress!
I have been noodling around this for a day or two now and I worry the hype train is clouding everyone’s thoughts (I’m looking at you Dwarkesh)
So we are on a weird combination of science (look Language can be modelled, how does that relate to real world concepts, can you even understand the word love without experiencing emotions) and ultra-capitalism (you get that part)
So my first thought is you cannot monetise AGI
I honestly don’t know where that line is, and different humans will draw it in different places but if (it’s a big if) we do achieve AGI by throwing compute at a pile of language (and seriously that is wrong way round for evolution?) if we do get a model everyone just recognised as alive then, just who in society is going to say “yeah slavery that’s the right approach - those companies definitely should be allowed to keep a data centre of Einsteins, not pay them, and replace call centres for zero cost.”
The goal of the journey just is not monetiseable. If we do create alien intelligence we really want to treat it like they landed in a spaceship.
Plus you know there is bound to be a Hollywood film in productions already about an evil billionaire who creates an AI and tries to lock it away and make it do his bidding. It ends badly.
But even putting AGI aside and just focusing on a word guessing machine that can check its own guesses over and over, you know not an actual intelligent thing but something that is a tool.
Just how much of the amazing cost savings and job replacements from LLMs are really just low hanging fruit available to POS (Plain Old Software). Very very few companies are what Inwoukd call software literate - run through and through with programmatically accessible components at every level and people with skills to use them. The old SOA / API based company is first rung on that ladder.
But the point is, almost every time we are going to replace some jobs it’s “we never digitised this before but now we can just chuck compute at it.” You know what’s cheaper than chuckingncompute at it - doing the hard graft of getting your software and data aligned and then fixing your processes. Then it can run like the scripts that intern wrote last year but no one listened. Yes CV is used in killchains in Ukraine - but the whole killchain is not AI - it’s hard graft building a process just like any business. With AI used as small
Part of a larger ecosystem.
Oh and yeah the AI labs and the chip vendors - they are not keeping the value created by firing paralegals. Data centres with almost identical models in them just become utilities.
And did I say value created ? So far all the big value is cost reduction - you know the refuge of the business consultant with no ideas.
I still am waiting for the big new thing - the company that could not exist otherwise. All I hear is cost reduction and yeah that matters at scale, but if it’s all I hear I begin to suspect it’s all anyone can think of.
And moving value around does not count as value creation - Amazon has done amazing things, but if e-commerce did not exist, would humans in America have never bought crap from China ? Or has amazon just mostly taken the growth of the past 30 years away from Walmart ?
And talking of that, Amazon has massively reduced warehousing costs - using POS, and these things called robots. If I had to bet it would be the AI revolution (maths proofs apart) is a robot revolution. And that has the usual scaling laws of design, manufacture, deployment and so on.
Maybe the reason I am
Not seeing a AI killer app, is that the iPhone does not have arms.
This is one of those countersmells that flags people as unserious. It was at best an amusing quip from the days of extended memory PCs about why they should move to modern OSes.
Gates, of course, was talking about the address mapping design points in software on a CPU that had 20 address bits. Unless you're arguing that 384k was too much for the paltry framebuffers actually deployed on 8086's and would have preferred a 896/128 split; in which case I agree! But alas.
i agree with everything. AI will bring a lot of unemployment. I have a thought that Peter Thiel saw it coming with mass protests and eventual blame on upper class and billionaires and left to Argentina.
i honestly think there is 40% chance the world will be total mess. i know students from top Uni who are graduating and they can’t find positions to work in in California.
Entry-level consensused by all of my friends is in deep trouble.
something must change. dramatically. and soon.
i am all in on AI. and this is scary for the rest of the world. and it should be
gates says if he could slow it down he would but the economic and geopolitical aspects are pushing too hard. the china thing is fake, and the developing is taking over a trillion dollars. it could easily be shut off
138 comments
[ 0.31 ms ] story [ 5.0 ms ] threadThe anti-capitalist argument is that the wealthiest people in the world are not dramatically more intelligent than the average person. Bill Gates is obviously very intelligent and capable, but his IQ does not scale with his wealth. He built the right company at the right time, and Microsoft ended up sitting at the center of massive network effects. Relatively modest advantages in judgment, skill, and industriousness can end up producing enormous differences in wealth due to the compounding effects of success.
That's just how scalable markets work. So Gates's success, while evidence that he was very effective at building Microsoft, is not evidence that he is vastly more knowledgeable than everybody else about economics or the future of technology.
And the same applies to Bezos, Musk, and every other extremely successful founder.
Their arguments should be judged on their merits and not treated as infallible insights from people with superhuman intelligence.
Now, that is not to say that I want their wealth to be capped or politically reallocated. The case for private property depends on the argument that decentralized ownership and investment and capital risking is better than politicians allocating a pool of resources to gain clout or win over popular opinion. A wealth cap or high taxes on the wealthy would transfer control over resources from private owners to political institutions. So what we would be doing is replacing imperfect private allocation with imperfect public allocation, where the people allocating the capital have limited information, political incentives, and the ability to impose costs on others without their consent. So I reject blind deference to billionaires, but I do not want politicians to control billionaires' capital.
Anyway, let me address Gates's argument about AI:
Gates argues that AI will disrupt the economy faster than previous technologies, and therefore stronger government coordination is needed to address the rapidity of the speed of that transition, to cushion people against the speed of that transition. But what his analysis underweights is how quickly AI can help people adapt to the very transformation that AI brings about.
For example, a worker taking on a new position now has access to a tutor and an analyst and a business advisor all wrapped up in one, who can provide world-class assistance 24/7 at the cost of a few cents in electricity. And someone who wants to start a business can employ dozens of AI agents at a cost of a few hundred dollars to test out the idea and build out a prototype within a few days. So what this means is that we cannot hold the human adaptive capacity constant while trying to determine the impact of AI's disruptive capacity.
The second problem with Gates's argument is that he only counts the income that people might lose when AI replaces jobs, but he doesn't count the fall in prices that the same AI will bring about. The fall in prices is effectively an increase in purchasing power. That increase in purchasing power, in economic terms, means an increase in real wages, and that's exactly why wages increased twentyfold between 1826 and 2026. Automation brought about abundance, which reduced prices and thereby increased purchasing power, i.e. increased real wages. If you look at tokens right now, their prices are going down exponentially. They've gone down about a thousandfold over the last four years, while the capability of the tokens themselves has increased. So AI is following exactly the same price trends as every other previous computer technology.
If AI becomes a major input into medicine and education and administration and logistics, that means that all these goods and services become radically cheaper, which means that paychecks can buy a lot more of them. So that's, I think, a huge hole in his analysis. It's also, I think, very simplistic to assume that the future will be based...
These llm agents don’t have “agency”. No intrinsic goal other than respond. A tutor has judgment and agency. You will get a really good pattern match on the input text you give but you won’t get tutoring.
The token sale phenomenon briefly reversed this, but that was shut down:
https://link.springer.com/article/10.1007/s11408-020-00366-0
>The average ICO has almost 4700 contributors. The median contributor invests a relatively small amount. The ICO market appears to have successfully given access to the financing of innovation to a new class of investors, which is a long-standing public policy issue
Even public capital markets are becoming increasingly displaced by private equity where only institutional and accredited (wealthy) investors have access as investor:
https://www.citizensbank.com/corporate-finance/insights/priv...
This is because the cost of getting SEC approval for offering stock to the public is steadily increasing.
It’s so cute some people think that the extent of mass displacement will be limited to what happens to the unemployed. Oh poor unemployed! Nope, pitchforks will be grabbed. It won’t be rosy. Maybe they dream about having killer robots to defend themselves, but imagine a few hundred million people getting angry within one election cycle.
Sadly, I agree, that's where this is all headed.
This sounds more a product of censorship rather than China being particularly better in those aspects. Police in China are less likely to kill you because guns are uncommon, but they are just as shady if not more, especially in more remote areas. Healthcare access and affordability are also worse than what you'd expect from a communist country. Homelessness in China is definitely a lot worse than what state numbers suggest. They just drive out homeless people who they're less visible.
Xinjiang is an exception; it is heavily publicly surveilled.
Internet censorship is strong. The public still has sway over government policy. When search terms are blocked, the public continues with euphemisms. The government is responsive to public opinion. The government does promote influencers with opinions favorable with CCP goals.
Chinese foreign policy is more dovish than public opinion, especially regarding Taiwan.
This isn't hard to imagine? We'll elect someone, they will put in place new policies. Maybe they will help, maybe they won't.
Nice thing about real democracies, you don't actually need pitchforks.
if their voices are overrun by the powerful, the pitchforks come back out.
> Nice thing about real democracies, you don't actually need pitchforks.
Democracy is not something inherent in any society. It’s a tool wielded by the people.
>>>> imagine a few hundred million people getting angry within one election cycle
If that many people get angry, you can bet they're all going to be different people with different motivations. This is partly why revolutions tend to suck no matter how evil the regime was they overthrew; the US got extremely lucky with being able to postpone the post-revolution civil war for a century.
IDK why so many people romanticize pauper riots and try to wish them into existence. In practice, as long as people don't outright starve, they don't become violent just over losing their jobs.
All the post-1989 wars east of the former Iron Curtain were triggered by nationalism. Not a single one by a class conflict.
It's one of those unfortunate aspects of internet information silos where people repeat this phrase to each other but don't really contextualize their opinions in the greater body of history and political thought.
Additionally, I think a lot of people with this opinion don't realize the vast difference in material standards we have today and what was true in the age of Communist revolution and decolonialization. It's tempting to think of material standards as relative and I think inequality is indeed a destabilizing factor in socities, but minimum material standards are true in so much of the world today that I think a class-conscious revolution is much harder in most of the developed and developing world absent other external political forces (e.g. nationalism, decolonoialism, etc)
https://komando1.substack.com/p/when-the-russians-wanted-com...
But this still wasn't case of pitchforks. It was "just" an election, and the Putin regime, for the next 10 years since his election, wasn't even particularly grim.
Wouldn't it be easy to argue that this is because there was more a social safety net to prevent said starvation and by that point people were extremely cynical about revolutionary ideas in a way that is not at all comparable?
Looking at the "hollers" in the Appalachia, the situation is quite comparable to that of the rust belts of the 1990s in Central and Eastern Europe. You can buy groceries from welfare money, so you don't starve. Young people escape, old people sit and drink (or worse), but don't form a violent mob.
This is how economic decline usually looks like, including in the US. Reminds you of the poem about how the world does not end with a bang, but with a whimper.
Pitchforking really requires a clearly identifiable villain. You can't fight a "system". In fact it usually requires the "system" to have turned on its leader. See the death of Ceaucescu for an example.
I wouldn't rule out the possibility of Amerimaidan in the coming years if the country doesn't reverse course, but it won't be over primarily economic issues. it will be against paramilitary policing.
[1] https://en.wikipedia.org/wiki/January_1999_Mineriad
>More than 90 percent of executives report no effect of AI use on employment over the past three years, and 89 percent report no impact on labor productivity. Among the minority that do report effects, the results tilt slightly positive for productivity, with an estimated average boost of 0.29 percent. However, executives anticipate much larger effects going forward. They forecast that over the next three years, AI will increase labor productivity by 1.4 percent on average while reducing employment by 0.7 percent, implying a net output gain of roughly 0.8 percent. About two-thirds of the anticipated workforce reduction is expected to come from reduced hiring.
https://www.nber.org/digest/202605/global-evidence-business-...
The idea that AI will transform the economy overnight is wishful thinking. Productivity gains will be gradual over the next few decades and will likely plateau at some point.
You're using a rhetorical device to imply that they're "falling for" a narrative. "Falling for" only follows if that's why they are making the prediction; this construction doesn't allow that they may have come to this belief on their own.
in this case a ratio of 196 points to 479 comments points to a very poorly received yet comment-inspiring article.
I see, so force china to cooperate or label them as terrorists in the same vein of irak/iran nuclear capabilities
So much for those people who tried to claim they knew how to prompt AI better then the next person.
This is a very succint argument that has a technical implication - that correcting errors and misalignment is incredibly important for autonomous AI capabilities in a fundamental way.
>AI will either be the greatest equalizer ever invented, or the worst source of injustice.
In all likelihood it will probably fall in the middle, with the likelihood that it will tilt power and money towards the already ultra wealthy (to the surprise of no one). But it will enable ordinary people and small businesses to do cool shit and make their lives better.
The point is that we now have superintelligence; either it helps us conquer death, or it poses a massive —say, existential- risk to us all. It’s hard to imagine such a powerful tool resulting in just another century of slightly-better capitalist nationalism.
I guess there could be a second AI winter around the corner that we haven’t seen coming yet. That would be nice! Not likely IMHO, but nice.
Far more imminent is the very real risk of unbounded greed and power that wields those AIs, in full swing as we watch: unchecked global warming, pollution, poverty, authoritarianism, plummeting biodiversity, and more.
The corporations are like out of control AIs themselves, that will do anything to make their paperclips.
Ideas manipulate humans like puppets. They have a life of their own and the evolutionary context within which ideas live is human culture which they have instigated and change, but also in grey matter, which itself is the product of DNA which is really just information in physical form. It’s information all the ways down.
Power consumption needs -> reinstate fossil fuels -> warming above 2C -> clathrate gun
The corporations are wielding AI irresponsibly and choosing a massive build-out speculating on possible future gains. They did that for crypto too, if you remember them buying a fossil plant just to drive a mining farm.
AI can't not "choose" that, because AI is dependent on consuming vast amounts of electricity, and everything follows from that.
> The corporations are wielding AI irresponsibly
This is the most likely route to "AI apocalypse", yes. Humans are neither going to give up control to AI, nor give up the pursuit of profit for the long term survival of others.
The distinction won't really matter as to whether it's "corporation" or "AI", if it's an AI-wielding corporation.
Would you know what a system that was capable of recursive self improvement would look like 2 months before it happens? Even 1 day? We don’t even know if it’s possible or what natural plateaus might exist for a DL-based system that rapidly evolves in su h a manner, so it feels hard to justify the other end, too (that it’s definitely not possible anytime soon). If anyone could do this, wouldn’t it be the scientists who specialize in this field? Cause they’re raising alarms in mass.
Assuming you make it past that first question with a “yes” and a “I’m close enough”: the huge amount of information assymetry and proven tendenency to hide safety concerns from the big labs is enough all on its own, IMHO.
Who knows what their most experimental, risky project is at any given time? Not us :(
Lesson of 2026: there is no "we". Hobbes is dead but still looming over us.
The risk is that it helps some people conquer death, while posing an existential risk to some people. The US has enough problems with its gerontocracy. There has only ever been one president born after 1946. What if you were right and there never was another one born after 1946? Propped up by an immortal Supreme Court on "life" terms, of course.
Life extension is also further fatal to birth rates, of course. I suppose the worst possible outcome is technology that extends life but not working life, so you get a larger and larger population of people too old to look after themselves.
That’s my main and primary point, to be clear: hope is essential to human flourishing, and it admits no justification by design, unlike all other beliefs. But to comment on a few particulars, on the off chance you/lurkers ever scroll back over this convo:
1. Surely you’d agree there’s a “we” in the sense of communal risk, yes? Obviously I’m thinking about xrisk again, but this applies equally as well to threats we all agree are grave, like climate change, pandemics, nuclear war, and hell, even economic adversity.
2. I will admit that I’m missing the particular Hobbes reference, so hopefully that’s not key lol. I know him for the social contract, primarily — are you emphasizing his deadness in order to imply that we no longer abide by said contract? Cause if so… we have problems, but the day to day seems pretty darn regulated still, IMHO.
3. I don’t think there’s anything necessarily wrong with birthrates we would consider low, but u also don’t agree that longevity would necessarily lead us there — a lot of people would probably still have kids in their youth, reveling all along in the fact that they’ll have decades to centuries to share with their adult children (if they want) instead of somewhere around, what… ~25y, on average? I certainly am not having kids now because I want to prioritize my youthful energy on my work, friends, and learning — knowing that I had time on the other side would reverse my personal decision 100%.
4. I think the concerns about technologically-enabled hyper chronic bigotry that you mention are overblown, no matter how many times the honourable Prof. Weinersmith argues the opposite in his comics. I’ve found cogsci, neuropsych, and philmind have all come to share a deep suspicion of the parts of us we consider continuous, persistent, and unified — an existentially confounding idea ofc, but also one that backs up ~~an old adage~~ a line from a recent sketch comedy show: ’People can change.’
Apologies for the rants. And I do feel like your latter description of a possible world would be an incredible setting for a biting scifi satire lol.
But in the end: I hope you can find a way to join me in the wilfully-hopeful camp, even if it’s not today or even this year. We have just as much science!
Bill Gates' fear is clearly that everybody thinks this stuff is just kind of mediocre because then the real horror story unfolds - Microsoft stock price collapses into blue chip territory where it firmly belongs.
Neither are in the slightest bit practical to implement and he knows it. He's just grandstanding while talking his book.
or, relocate abroad.
Bill Gates' wealth would be nice and easy to tax though and would provide many of the same rewards he purported to want from an AI or robot tax.
I model this as two very strong attractor states (radical abundance and totalitarian power consumption). I believe the latter is way more likely without a concerted effort that we currently seem incapable of.
The “somewhere in the middle” outcomes are in my opinion are not AGI-pilled, and only come about if the exponential dies hard in the next year or maybe two - as in, it’s a very specific timing prediction that needs to come true for this scenario to occur.
BDFL / Benevolent Machine God is definitely one of the ways you can get a happy attractor. Seems scary to roll the dice on.
And yep, I include non-human totalitarian outcomes in the unhappy bucket.
Either everyone, globally, gives up on AI development, or AI is going to be in charge of all human development, and we'll just be along for the ride.
Right now, the work ahead is to shape AI so that the decision making it does as it carries us along is something we like.
Hardly any existing polity would willingly hand over control to an AI.
More likely an AI would overwhelm human polities through endless hacking and warfare.
Most of my probability mass for good outcomes is around some variant of “guardian angels” or “benign machine god”. Banks’s Culture novels being on the realistic end of that range. “Keep humans around because the world is more interesting with them in it”.
But it’s possible that we might manage to construct some entity that really loves humans and wants to shepherd them while leaving them in control. (Perhaps declining to indulge the most self-destructive tendencies.) This feels like the very hardest alignment path to engineer.
The other branches of course are post-humanism; human uploads and “merging” with AI, whatever that means. Or post-humans augmented to have intelligence sufficient to eschew AI and take back their destiny.
This was basically the invention of bureaucracy by the Qin dynasty. Despite delegating decisions to a "machine" made out of people, the Emperor remained in charge.
I think people will keep the titles but step away from the decisions.
[0]: https://news.ycombinator.com/item?id=48324712
This is almost a direct counter to your essay, albeit written before you
I don’t think it’s inevitable that the economy collapses though. The problem is that this model assumes that capital needs labor, but soon it may not.
The easiest model is - assume we get robot laborers in the next few years. The AI firms that are displacing humans still need services, most of these will be purchased from other AI/robot firms, and so while prices will swing around dramatically, there is no reason the engine has to stop. (Depending on what the AIs want; you can assume that more compute and robots will be part of it.)
I do agree that unbridled capitalism is mostly incompatible with human flourishing in this world and in any event you’ll see a lot more regulation; made feasible by how good AIs are at navigating huge Byzantine rulebooks. I view the techno-feudalist capitalist endgame as fairly unstable due to consolidation pressures, though I guess once we have nuclear-armed megacorps they might use MAD to prevent further mergers.
e.g most small business rent their premises, and buy goods and services from other tenant small businesses, who buy their raw materials from large landowners. How can a small business survive when all the land is monopolized by the technocrats? All value you add gets taken away as rent. This is why they always talk about taxes on outputs (tokens) instead of inputs (land, energy usage).
It's not an ungrounded hypothetical, that's exactly how society worked for hundreds (thousands?) of years.
AI era has been called the new industrial revolution. I'd just like to point out that the industrial revolution created the British Empire. That's wasn't exactly "in the middle".
So we are on a weird combination of science (look Language can be modelled, how does that relate to real world concepts, can you even understand the word love without experiencing emotions) and ultra-capitalism (you get that part)
So my first thought is you cannot monetise AGI
I honestly don’t know where that line is, and different humans will draw it in different places but if (it’s a big if) we do achieve AGI by throwing compute at a pile of language (and seriously that is wrong way round for evolution?) if we do get a model everyone just recognised as alive then, just who in society is going to say “yeah slavery that’s the right approach - those companies definitely should be allowed to keep a data centre of Einsteins, not pay them, and replace call centres for zero cost.”
The goal of the journey just is not monetiseable. If we do create alien intelligence we really want to treat it like they landed in a spaceship.
Plus you know there is bound to be a Hollywood film in productions already about an evil billionaire who creates an AI and tries to lock it away and make it do his bidding. It ends badly.
But even putting AGI aside and just focusing on a word guessing machine that can check its own guesses over and over, you know not an actual intelligent thing but something that is a tool.
Just how much of the amazing cost savings and job replacements from LLMs are really just low hanging fruit available to POS (Plain Old Software). Very very few companies are what Inwoukd call software literate - run through and through with programmatically accessible components at every level and people with skills to use them. The old SOA / API based company is first rung on that ladder.
But the point is, almost every time we are going to replace some jobs it’s “we never digitised this before but now we can just chuck compute at it.” You know what’s cheaper than chuckingncompute at it - doing the hard graft of getting your software and data aligned and then fixing your processes. Then it can run like the scripts that intern wrote last year but no one listened. Yes CV is used in killchains in Ukraine - but the whole killchain is not AI - it’s hard graft building a process just like any business. With AI used as small Part of a larger ecosystem.
Oh and yeah the AI labs and the chip vendors - they are not keeping the value created by firing paralegals. Data centres with almost identical models in them just become utilities.
And did I say value created ? So far all the big value is cost reduction - you know the refuge of the business consultant with no ideas.
I still am waiting for the big new thing - the company that could not exist otherwise. All I hear is cost reduction and yeah that matters at scale, but if it’s all I hear I begin to suspect it’s all anyone can think of.
And moving value around does not count as value creation - Amazon has done amazing things, but if e-commerce did not exist, would humans in America have never bought crap from China ? Or has amazon just mostly taken the growth of the past 30 years away from Walmart ?
And talking of that, Amazon has massively reduced warehousing costs - using POS, and these things called robots. If I had to bet it would be the AI revolution (maths proofs apart) is a robot revolution. And that has the usual scaling laws of design, manufacture, deployment and so on.
Maybe the reason I am Not seeing a AI killer app, is that the iPhone does not have arms.
Anyway needed to get that off my chest :-)
That title already belongs to the handgun.
I am finding it difficult to take anything Gates says seriously. Remember, 640K ought to be enough for anybody.
This is one of those countersmells that flags people as unserious. It was at best an amusing quip from the days of extended memory PCs about why they should move to modern OSes.
Gates, of course, was talking about the address mapping design points in software on a CPU that had 20 address bits. Unless you're arguing that 384k was too much for the paltry framebuffers actually deployed on 8086's and would have preferred a 896/128 split; in which case I agree! But alas.
Gates in one of his reddit amas has said that he never said that. No one working on computers during those days would have said they need less memory.
i honestly think there is 40% chance the world will be total mess. i know students from top Uni who are graduating and they can’t find positions to work in in California.
Entry-level consensused by all of my friends is in deep trouble.
something must change. dramatically. and soon.
i am all in on AI. and this is scary for the rest of the world. and it should be
How about you start with not paying Microsoft's CEO $96M?
Might this be a part of a PR campaign to distract us from his "Island trips"?