Ian Welsh

The horizon is not so far as we can see, but as far as we can imagine

Category: Science and Technology Page 1 of 3

Maybe Early AI Adoption Is Stupid?

By

On August 11, 2026

14 Comments

The thing about AI is that it takes previous costs (your engineers and their tools) and makes them more expensive. Often much more expensive:

if you built your saas business before ai, your entire business was designed around one assumption which was that software has ~zero marginal cost.

ai blows that up. every ai action costs money. incumbents now have to create more expensive tiers or introduce usage based pricing. both force customers to make a new purchasing decisions while revenue doesn’t automatically increase just cuz the product became more expensive to operate. so your cost per seat rises, your gross margins compress, & adoption remains uncertain.

meanwhile, the model labs are subsidizing ai usage like crazy, so they can undercut you. ai native startups are burning venture capital to acquire users, so they can undercut you too.

(lack of capitalization in original)

I really don’t know what to think of AI for coding. I’m no longer a coder, haven’t been for almost thirty years now, so I don’t have enough firsthand knowledge, though I did have an AI generate basic sorting and random number generation code just to get my hands a little wet. Among devs I trust, opinion is split. Some love it, some hate it, some are on the margins. Here’s one in the hate category:

Are companies actually seeing massive productivity gains from their AI adoption?…

… All of the AI projects we have observed as a team are failing. Every single one – we have seen 0% success in a year and a half, not only amongst projects we have been asked to participate in2, but even within projects that we have observed in passing while doing totally unrelated work. Even if you grant that AI tooling accelerates specific workloads, the method and scale of the current investments is senseless. Frequently the failure is not related to AI itself, but rather that companies are terminally bad at running software projects effectively, and as I have remarked previously, AI projects are subject to all the failure modes of normal projects plus you can get everything right and then still fail because of the method’s novelty. Very few companies are so good at shipping software that they can afford the extra risk profile…

…

our team has quickly learned while on an engagement not to ask anything about ongoing AI projects in any context – by the time that project has started, it is too late for the management team, and intervention is not possible until a crisis point is inevitably reached. There is no conceivable positive outcome. The failure rate is so high that even basic inquiry leaves us in an untenable position. Any coherent question about how it’s going, what the goal is, who is using it, constitutes an inadvertent attack on the chain of command responsible for the work because there are no good answers to anything. Even in rare cases where my interlocutor has stated that things are going well (usually while the project is still mid-flight and failure has not had a chance to manifest), it is generally obvious that they are doomed, but at least in these cases I can simply agree and then go home to scream into a pillow for six hours straight3.

All of this is to say that I am very confident that almost every report at a company about “massive AI productivity gains” is untrue as a matter of brute fact. Even if some companies are seeing clear gains, this is the exception, not the norm.

It’s a long post and the entire thing is worth reading. This is a high end engineering consultant who gets a view on a lot of companies.

My feeling/guess is that AI is not mature, that the use cases exist, but that it introduces significant risks of new failure modes because AIs aren’t reliable and aren’t actually intelligent: they don’t know what they’re doing. If you don’t do the work yourself or audit it to the level where you might as well have done it yourself, it seems easy for errors to creep in, for hard to maintain or understand code to be created, and for new failure modes to exist.

If I were an executive in most businesses I wouldn’t be using AI for much, if anything yet, though I’d let a few people have an account and test it, and would do so myself. It’s moving fast, it clearly adds risks and mistakes that we don’t know how to mitigate yet, it’s expensive and that’s with massive subsidies, so becoming dependent on it when prices are likely to increase significantly is unwise.

There are going to be exceptions: if you’re Google, not sure you have much choice. But if you manufacture widgets or sell hamburgers or build homes or are a lawyer (who can’t afford to have the AI hallucinate cites) I’d give it a pass for now, or use it very warily in exploration mode. I use it for research myself, especially as search engines get worse and worse, but I also check the sources it uses.

And, as we’ve discussed before, non open models have huge risks, since prices can easily be raised, the government can intervene to cripple the model or deny to other nations, and the company itself can make changes to the model which make it less useful to you. So even if using AI I’d be primarily focusing on Open models (which means mostly Chinese.)

But my best guess is that this is a real tech, with real uses, which is not yet near to mature and which is being deployed before people understand what it’s good at or how to use it safely and cost effectively, or, more importantly, when NOT to use it. I’d also guess that at least so far, it’s not the second coming, the next great thing, in the way that its evangelists preach.

So far it seems to increase failure rates on real projects and raise costs significantly at the same time. Unless I’m running a business where I must be in it, I think I’d take it slow.

There are very few businesses where “the way we did it in 2022” will expose you to significant risk and costs. If you’re in one of those businesses, chill a little and observe. Let other people pay the price and make the mistakes. If you need to, because it’s the current “everyone must do” just lie and say you’re into it, while doing the minimum.

This is certainly something I could be wrong about. I’m confident that Chinese AI will win, for example, I’m less certain about how useful AI will turn out to be. I do think that I’m almost certainly right that most companies adopting it in a big way, NOW, are making a mistake. Wait. See how it turns out. Let other people figure out what it’s good for, how to manage it, how to mitigate the risks and how to drive costs down.

And remember, even if it’s the true next big thing, early rushers often get badly burned, as with the dot-com bubble. Don’t buy hype, make sure what you’re investing in is real and you understand what it’s good for and what it sucks at.

What I write here is for the benefit of everyone, but alas, I live in capitalism and I, and the site, take money to keep running. If you value the writing here and can, please subscribe or donate.

Burning Down The Great Library Yet Again

By

On July 29, 2026

21 Comments

Sigh:

AI companies are bulk-buying rare books, scanning them through high-speed machines that cut the spines off, and shredding the originals. A service called ISBNdb facilitates orders of up to a million books and keeps buyers anonymous. Pre-2022 books are premium because they’re free of AI-generated text. A federal judge ruled the practice is fair use because eliminating the original means only one copy exists at a time. Anthropic hired the former head of Google Books partnerships to obtain “all the books in the world.”

My Take This got to me. A bookseller told 404 Media that rare books with almost no surviving copies are being fed into this pipeline. Books that survived wars, fires, and centuries of handling are being shredded so an AI can learn to write a better marketing email. ISBNdb’s website literally says “‘AI company destroys two million books’ is not a headline that generates sympathy,” and they still built an entire business around making it happen quietly. They offer NDAs as a feature. They coach clients to call it “digital preservation.”

I’ve covered AI companies scraping the internet, torrenting libraries, and stealing music. This is worse because it’s irreversible. You can re-upload a website. You can reprint a bestseller. You can’t replace the last three copies of an 18th-century botanical text once someone shreds them for training data. And the judge said it’s legal. So it’s going to accelerate. “We shred rare books and offer NDAs so nobody finds out” is a legitimate business model in 2026. What a timeline.

These people are barbarians. Philistines. As bad as destroying the library of Alexandria or Mongols burning libraries in Baghdad.

This knowledge is irreplaceable and it’s not worthless (if it was, they wouldn’t be scanning it.) If it were up to me this would be a capital crime. Not exaggerating for effect. Everyone involved would go for the long jump.

(To give credit where credit is due, apparently Musk does not shred books for his AI. He’s scum in many ways, but one should praise the correct actions of bad men.)

We all have a limited time here, to destroy that which could live much longer than us and speak of the past to the future is a monstrous crime.

Truly most of the people who run AI are scum.

 

What I write here is for the benefit of everyone, but alas, I live in capitalism and I, and the site, take money to keep running. If you value the writing here and can, please subscribe or donate.

We Live In Epochal Times

By

On July 28, 2026

16 Comments

History has cycles. When dealing with ideas about how to organize societies there are ideological and sub-ideological cycles. Various nations have been capitalist since the 16th century, but what the means kept changing, usually every 50 to 80 years. Britain started very protectionist. It had periods where imperialism and capitalism were very intertwined. It had laisszez faire periods. All of these were forms of capitalism, but they were quite different from each other. The Roman Empire is an empire from Augustus to the end, but the Rome of Augustus is very different from future Romes, with their serfs (something Augustinian Rome didn’t have), their German legionaires, and so on.

In our times there are people alive who remember two and a half transitions:

  • The transition from laissez-faire (in America) to New Deal.
  • The transition of New Deal to Neoliberalism.
  • The transition of neoliberalism to whatever the new system will be called. (This is not the old system, Trump’s crazed tariffs, export controls to major buyers, the severe cutting of research, etc… all show that.)

In China they remember:

  • The change from warlordism and Republicanism to Maoism.
  • The change from Maoism to Dengism.
  • The recent change from Dengism to Xiism.

Globally we remember multiple transitions as well

  • The transition from the multi-great power world centered on America/Europe to the Cold War world dominated by the USSR and US. (Note this transition involved two world wars.)
  • The transition from the cold war world to the unipolar superpower world with the US the world hegemonic power.
  • The current transition from the American hegeomonic world to whatever will replace it.

In energy technology we have one and a half transitions:

  • The transition from the coal world to the petrochem world;
  • The current transition from the petrochem world to the electrical world based on other power sources.

Note that the technological cycle is a LOT slower. (And yes, I’ve left out the digital/telecom, but you could easily make the same sort of schema.

There is also the environmental question, and that’s part of various cycles. We aren’t different from other cultures in this regard: one thesis for the fall of Mesopotamia from the world’s leading region for six to eight thousand years is environmental. They aren’t the only ones.

We have multiple changes income all once:

  • Sub ideological changes in how capitalism and democracy run in Europe and the US.
  • A more or less complete change in China from Dengism to Xi-ism.
  • A change to the world political system, with the rise of China and the weakness of America.
  • Environmental collapse
  • Climate Change

All of these at the same times. Not only is “nothing ever happens” bullshit (as it always was) but we live in a period of radical change, and it’s only going to become more obvious over time.

But there are people alive, albeit not many, who remember the pre-WWII world.

Change is constant. It seems slow to humans, but those who live a long life have seen a great, great deal of change. And we didn’t even discuss the way daily life has changed due to the change from the analog world to the digital one. (A change I personally think was probably more bad than good, though I certainly love large parts of it.) And, of course, there’s the revolution in war making technology, strategy and tactics, as big as the change to trench warfare, then the change to blitzkrieg and air war.

Everything changes. And if you have a perspective longer than a couple decades, everything changes pretty fast, actually. Certainly the world I grew up in, that I remember from the 70s and early 80s no longer exists.

In human affairs nothing is constant save change.

What I write here is for the benefit of everyone, but alas, I live in capitalism and I, and the site, take money to keep running. If you value the writing here and can, please subscribe or donate.

America’s Losing the AI Race Hard

By

On July 21, 2026

32 Comments

Amazing stuff, and sooner than even I expected. China’s Kimi is now about equal to GPT and Claude:

For the first time, a Chinese model Kimi K3 has taken on the Frontend Code Arena and is scoring at or near the frontier on other benchmarks.

And the majors are noticing:

Meanwhile Trump is talking about banning Chinese AI. 

Alternatively, the smart lads at Open AI have a regulatory plan:

I would guess that the Trump Administration will at some point realize that their best strategy here would be to create large amounts of regulatory risk around the use of open-weight Chinese models. You don’t need to “ban open source” (one of the dumber motifs of AI policy discussion). You just need to direct every agency to issue soft law that creates FUD. “A Federal Reserve Advisory Bulletin found that there may be backdoors in Chinese AI models.” It needn’t be that well justified. You just create enough regulatory risk that every regulated enterprise backs off. You probably don’t want to create so much regulatory risk that you scare off the hyperscalers from serving Chinese models; this will just drive startups to sketchier providers. There’s a happy middle ground here. I’d assume they will do some version of this.

Even more fun, the Feds want to approve all AI releases and not allow everyone the full models.

The Trump administration has taken new steps to assert more control over the rollout of future artificial intelligence model releases by dictating which companies and entities are allowed access to the latest frontier models, two people familiar with the matter told CNBC…

….

This week, the administration launched its own program, dubbed “Gold Eagle,” aimed at collaborating with the private sector to find and fix cyber vulnerabilities.

The so-called clearinghouse would put the White House in charge of greenlighting which companies can access new AI models, according to a person familiar with the matter, who spoke on condition of anonymity in order to discuss information that is not public.

China is going to eat America’s lunch on this. If they ban Chinese AI (harder than it seems, given it’s open source) all that means is writing off the rest of the world. And since American models are handicapped, smaller American companies will be stuck with worse AI.

Since Chinese AI is far cheaper, as well, I’d expect American companies to set up subsidiaries overseas to use it, rather than be stuck with American AI.

The entire situation is a complete clusterfuck. Major companies have taken on serious debt in order to build data centers which have a lifecycle of five to seven years, and often less (since new generations of GPUs are much better.)  But the Chinese product is cheaper, open source, lacks nearly as much sovereign risk and I’ll bet multiple models will soon be about as good as Anthropic and OpenAI’s.

Where’s the business case that spending all these trillions of dollars is going to produce enough revenue from US AI to pay for all of it?

There isn’t one. It doesn’t exist.

And that means that, at least in America, this is an AI bubble. All bubbles burst and this will not be an exception. If the government bails them out it will be the last major US bailout.

This is also very likely one of the last major tech revolutions which will start in the US (which it did.) Going forward they China will produce the vast majority of them.

This is the endgame. The turning point where China obtains not just the industrial base but the absolutely undisputed tech lead. From now on China will like America in the 1950s — it’s where almost everything new is created, the dynamic center of the world, and soon people will be competing to move there, because everyone knows it is the future.

 

What I write here is for the benefit of everyone, but alas, I live in capitalism and I, and the site, take money to keep running. If you value the writing here and can, please subscribe or donate.

 

Xi Lays Out His Principles for AI Development

By

On July 17, 2026

16 Comments

 

There were rumors recently that China was going to put export restrictions on AI. That now seems… unlikely. Here are the principles Xi laid out:

First, adhering to the principle of openness and win-win cooperation while boosting innovation-driven development. Xi highlighted the importance of encouraging open-source, openness, collaboration and sharing to facilitate technological innovation, industrial development and scenario-based application of AI. (My emphasis.)

Second, strengthening risk awareness and ensuring that AI is secure and controllable. Stressing the need to ensure that AI is always under human control, Xi urged all sides to jointly oppose overstretching the national security concept in the field of AI or placing one country’s security over that of others.

Third, encouraging inclusiveness and promoting mutual learning among civilizations. AI development and its application should not erode or undermine the diversity of world civilizations or the uniqueness of cultures of different countries, according to Xi.

Fourth, advocating solidarity and improving global governance. The important role of the United Nations should be recognized, Xi said, calling for further alignment and coordination on AI development strategies, governance rules and technical standards.

I’ve predicted, for a couple years now, that Chinese AI models will be the main models used in most of the world, including in much of the West, assuming they aren’t banned outright, because they’re open and cheap. Costs of running them are about twenty times lower than the US frontier models made by OpenAI and Anthropic. They’re almost as good, and they aren’t that far behind.

The problem with US models is not just that they’re expensive (though that’s huge, there are tons of reports of AI use being cut back) but that they are CLOSED: meaning you can easily be cut off, or have prices raised, or have the model changed on you with no recourse. Open models you can adapt the model, you can run it on your own servers, or various server companies can, will and do run them for you on their servers which you rent.

It’s clear that Xi gets this, and thus that the CPC understands it as well. Open Source isn’t a liability, there’s a reason why Linux runs most of the world’s servers: closed tech is the liability. Open Source is the advantage.

Notice the second bit: on AI always being under control. I wonder if Xi is thinking of Israel and the US for military targeting and how that has possibly contributed to hitting civilian targets like schools. (Possibly because Israel and the US are run by psychopaths and I bet they’d do it anyway. But no human in the loop may make it even worse.)

The third principle is about avoiding US (and Chinese) cultural hegemony. A nice thought, and open source certainly could be adapted to different nations and cultures, so that AI models aren’t all giving the same generic results.

Finally, the fourth principle. I’ve always found it interesting just how much China plays up the United Nations. I don’t know if the respect is genuine, but the words are consistent. Honestly, I think the UN should move its main HQ out of America. America’s been pulling stunts like denying diplomats visas. Not sure if it should go to China (though if it did Shanghai or Hong Kong seem like good fits) but they’d be better stewards than the US, and in any case, if the UN is going to be in the most important great power, that’s now China. (That said, I’d favor something more neutral. Perhaps Singapore.)

China just keeps coming across as smarter, more strategic and more human than the West. It’s sad, in a way, but it is what it is.

And I remain convinced that Chinese AI will be the winner over American.

What I write here is for the benefit of everyone, but alas, I live in capitalism and I, and the site, take money to keep running. If you value the writing here and can, please subscribe or donate.

 

The Old Gray Lady Runs RussiaGate 2: They’re Coming for OpenAI

By

On July 10, 2026

12 Comments

Guest post by Nat Wilson Turner.

New York Times: "China, Russia and Others Seek to Inflame Debate Over A.I. Data Centers"

Thursday’s New York Times brings back their old RussiaGate spirit with a front page banner headline about “foreign interference” and data center opposition.

Here’s a key quote:

…a push by foreign adversaries to seize on what polls have shown is deep ambivalence — verging at times on hostility — about the spread of the data centers needed to power A.I. in the United States and elsewhere.

China, Russia and, to a lesser extent, Iran have sought to use state media outlets to turn the controversy over data centers in the United States into “a domestic fracture point,” according to a new analysis by Alethea, a threat intelligence company, which identified scores of articles and posts on social media this year.

These campaigns, whose impact on public opinion remains to be seen, have raised alarms in Washington, where A.I. is seen as a top issue heading into this year’s midterm elections.

The foreign efforts appear intended to stoke the debate over data centers that has united political figures across the political spectrum — from Senator Bernie Sanders of Vermont, a progressive, to Stephen K. Bannon, the erstwhile adviser to President Trump.

“Foreign actors aren’t manufacturing American debates over the future of A.I., they are exploiting them,” said Jessica Brandt, a former official with the Office of the Director of National Intelligence who tracked foreign influence efforts during the Biden administration.

The goal, she added, is to “deepen our divisions in order to dent our appeal and weaken us from within.”

Interesting sources, some company called Alethea and a former Biden admin DNI spook as our sources.

We’ll come back to them later, but first I’m curious as to why the NYT is only now covering this story when OpenAI put out a press release saying basically the same thing on June 10 except focused solely on China.

After all, OpenAI’s report was convincing enough to sway such luminaries as Senator Tom Cotton (R-AR), Republican Leaders on the House Energy and Commerce Committee, Rep. Brett Guthrie (R-KY), Interior Secretary Doug Burgum, The Bitcoin Policy Institute and prominent tech investor Kevin O’Leary, per WIRED.

Read More

Just How Heritable Is IQ

By

On January 15, 2026

29 Comments

The IQ debates are, to me, tiresome. I’m pretty high IQ, not what I consider genius level (I’ve encountered true geniuses) but just under, in the one-in-ten thousand range. Which is to say, if I’m around 10K people I expect that no one is smarter than me, unless it’s a place that selects for IQ. At MIT I’d be nothing special.

But what I’ve also noticed is that high IQ, and I’ve spent a lot of time around high IQ people and reading them, has very little correlation to being right about the sort of problems which interest me. Virtually all the high IQ economists were wrong about, well, everything, for generations. Larry Summer is extremely high IQ and he’s reliably wrong. If you want to be right about something, find out what Larry Summers thinks, and you at least know one wrong view.

IQ is very good at following rules, even very complicated ones, at seeing correlations and at pattern matching. Without judgment all IQ does is get you to where everyone who shares your priors, as the youngs say (I call them axioms or assumptions), faster.

I also believe that IQ can change over time. The more you do of something, the better you get at whatever that is. Being good at economics makes you better at economics and the types of reasoning and math it uses. (It does not make you better at understanding economies, that’s something entirely different.)

And I think that IQ is only somewhat heritable.

Right now we’re in an period where the consensus among smart non-specialist is that IQ is highly heritable and most of this comes from the result of twin studies.

David Bessis, a mathematician, has a long post based based on a lot of work where he actually looked at the twin studies. You should read it.

The conclusion is that these studies are extremely flawed and can’t be used to make the claims made. The twins were often placed separately not immediately after birth, in fact in some cases as late as eight years old. The effects of mother’s on babies in the womb is huge (smoking, drinking, lead exposure) and that’s environment, many of them were placed with extended family and almost all were placed with middle class families similar to the ones they came from.

This debate matters. High heritability means that certain families are just superior. Bessis has a good summary of this. (The current strong case is 80% heritability.) I’m going to quote him here:

Let’s say, for example, that you are a genetically average person. How much does that affect your prospects?

  • Surprisingly, at 30%, it’s as if your genes didn’t matter at all. With an average potential, you still have a decent chance of landing at the top or bottom of the IQ distribution. Actually, in this specific random sample, one of three smartest people around (the top 0.3%) happens to have an almost exactly average genetic make-up, and the fourth dumbest person has a slightly above-average potential.
  • At 50%, being genetically average starts to limit your optionality, but the spread remains massive. Had you been marginally luckier—say, in the top third for genetic potential—you’d still have a shot at becoming one of the smartest people around.
  • At 80%, though, your optionality has mostly vanished. It’s still possible to move a notch upward or downward, but the game is mostly over. In this world, geniuses are born, not made.

This discussion is generally omitted by hereditarians, which is unfortunate, because it is the only way to clarify the stakes. There is a fundamental asymmetry in the debate. Heritability matters a lot when it is extremely high, because it then supports genetic determinism, but for the rest of the range the exact figure has little practical significance.1

Now while Bessis doesn’t go into it, what I find even more disturbing are the racial/ethnic version of genetic IQ determinism. I think they’re largely bunk (that’s another post) but many very smart people believe them. Koreans and Chinese and Ashkenazi Jews are smarter than whites who are smarter than blacks and so on, and this is taken to explain differences in how well various countries do, not their history or their environment. Blacks are, in this view, innately stupid. It’s not that they were colonized and brutalized and that the environments they grow up in are harmful to IQ development, nope, it’s innate.

If heritability is 80%, well, they just “deserve” their fates, and there’s really nothing that can be done about it. (If IQ determines national success, which is also BS if you ask me. If it was that simple, China would never have had its century of humiliation and whites shouldn’t have ruled the world for hundreds of years when Chinese and Koreans and Ashkenazi Jews are so superior to us.)

It’s not, in this view, that Talmudic study and cultures that place an obsessive value on learning like Korea and China do, develop higher higher IQs, it’s that they start smarter.

Now, as with Bessis, I think there IS a genetic component to IQ. It’s not like it doesn’t matter at all. I just think other things matter too, and that IQ matters less than people think it does.

We may revisit this issue, though I’m unsure. For a lot of my writing career I spent a great deal of effort debunking bullshit. The problem is that it never works, most people aren’t convinced, it takes longer to debunk than produce, and there’s always more of it because the pernicious types of bullshit are highly funded. It’s hard to compete with entire think tanks spewing out garbage, and that’s the job of 90% of think tanks: what they believe is pre-determined, donors want “intellectual” arguments to back up what they already believe or what they want others to believe because it is beneficial to them.

If excellence, however determined, is 80% hereditary, then aristocracy, however defined, is justifiable. The best people come from certain genetic lineages and deserve their place in the world. Whites deserve to be above blacks, Chines and Koreans above whites, and Ashkenazi Jews are the super race. (As an aside, though not genetic, trans women blow Ashkenazi out of the water in terms of average IQ, which I find hilarious, since it means that the people who love IQ and think it’s determinitive, should love trans women.)

It also means that there’s one less reason to improve circumstances of the majority of people. The few sports will rise to their level of genetic fitness and everyone else deserves to be where they are and doesn’t need support to improve their excellence, since that’s determined by genetics not environment.

This stuff is fought over because it matters, just like the divine right of Kings mattered. It’s about justification of how society runs, or an argument to change how society treats different people. Material circumstances matter, but so do ideas. We are slaves to what we believe the world is like and what we believe people are like. We often act on those beliefs. As the sociological maxim says “things believed true have real consequences even if not true.”

Twin studies don’t show 80% hereditability because those studies were extremely flawed. That matters.

This site is only viable due to reader donations. If you value it and can, please subscribe or donate.

Is China Going To Win The Humanoid Robot Race & End Capitalism As We Know It?

By

On November 18, 2025

14 Comments

Elevated from Comments. Piece by KT Chong

China is now entering the next phase of its economic-growth engine — humanoid robots.

And just like with EVs, the shift is happening fast, quietly, and with the same pattern: Chinese companies industrialize before Western analysts even realize it’s begun.

UBTech, Unitree, XPeng — they’ve all started mass-producing and delivering humanoid robots. This is not “prototype hype” or “lab demo” stuff anymore. It’s real machines getting shipped to real factories, hospitals, and even homes. China’s humanoid sector is going to be the next multi-hundred-billion-dollar growth curve, and the West is, once again, completely oblivious.

Frankly, IMO it’s already too late for the West to catch up.

Anyway, my point here today is… the Unitree G1 Ecosystem.

While reading deeper, I found something much more important: a lot of these new humanoid startups aren’t building from scratch. Instead, they’re standing on the Unitree G1 frame and layering their own proprietary AI on top. That means Unitree has quietly become the default hardware platform for China’s humanoid boom — like the Android of robot bodies.

A few examples:

1. A-Bots Robotics (Shenzhen, 2024)

• Focus: precision assembly, modular SDK

• AI layer: Baidu Ernie-ViLM for object manipulation

• Notes: 150+ units in Foxconn trials; ~$22k package; tuned for fragile electronics

2. HPDrones Tech (Guangzhou, 2023)

• Focus: warehouse logistics + drone hand-off automation

• AI layer: proprietary SLAM + multi-floor routing

• Notes: partnered with Unitree; 500-unit rollout for e-commerce warehouses in Q1 2026

3. LeRobot Labs (Beijing, 2024)

• Focus: open-source robotics + reinforcement learning

• AI layer: embodied datasets, tool-use improvisation

• Notes: hacked 20+ G1s for universities; GitHub repo exploded; expanding to eldercare

4. Weston Intelligence (Hangzhou, 2023)

• Focus: healthcare — vitals scanning, bedside conversations

• AI layer: Tencent Hunyuan conversational model

• Notes: deployed in Shanghai hospitals; sub-$20k price; measurable patient-compliance benefits

5. DexAI Dynamics (Shenzhen, 2024)

• Focus: dexterity — folding fabric, micro-adjustments, teleop self-supervision

• Notes: $80M raised; 100 units deployed in garment factories; arguably the best hands in China now

And then there’s MindOn — the one that caught my eye earlier — using the G1 frame to build a full butler/housekeeping robot (“MindOne”). One of their engineers even said they eventually want their own frame, but that’s the point: everyone is starting on Unitree first.

Unitree has locked down the humanoid robot ecosystem

All these startups — even if they eventually design their own skeletons — are still tying their early models to:

• Unitree’s frames

• Unitree’s actuator supply chain

• Unitree’s low-cost motor ecosystem

• Unitree’s software layer and APIs

Once you build your first few generations on someone else’s chassis + firmware, you’re effectively locked into their ecosystem. Switching costs explode. You’d have to rewrite half your AI stack.

So Unitree has already achieved what Western robotics companies wish they could do:

Become the default hardware substrate for an entire national robotics industry.

This is exactly how China overtook the West in EVs — standardized hardware, cheap mass manufacturing, and dozens of startups building on top of the same base.

Unitree is still a private company.

Given everything above, the most obvious question becomes: When does Unitree IPO?

On 15–16 November 2025 (literally this weekend), Unitree completed its pre-IPO regulatory tutoring with CITIC Securities — an unusually fast four-month process that normally takes 6–12 months.

The company publicly stated in September that it expects to submit the formal prospectus and listing application to the Shanghai STAR Market between October and December 2025.

Market sources still quote a targeted valuation of up to US$7 billion (≈50 billion RMB).

Once the prospectus is accepted (usually 2–4 rounds of CSRC questions), the actual listing can happen remarkably quickly in a hot sector — sometimes inside 3–6 months. A Q1/Q2 2026 listing is the base case, but a very late-2025 listing is still possible if the regulator fast-tracks it the way they have the tutoring.

What About America?

Meanwhile… America’s Great White Hope Elon Musk is already behind.

Elon Musk promised that the U.S. would lead the humanoid robot race with Tesla Optimus — but the timelines have slipped, and the window has basically closed. By the time Musk’s robot is actually ready for real-world deployment — 2 years from now? 3? — China’s robotics companies will already be deep into mass production, with tens of thousands of units deployed across factories, warehouses, homes, hospitals, and service industries.

And let’s be real — we all already know this:

Tesla will NOT be cost-competitive. Not even close.

China has already hit the sub–$20k price point for serious humanoids. Several G1-derived platforms will likely break below $15k. Meanwhile, Tesla Optimus — if it gets out of prototype limbo — will land somewhere between $20k–$40k+, before customization, localization, or integration costs. It’s the exact same pattern we saw with EVs, solar panels, drones, lithium batteries, telecom gear — the U.S. builds one expensive proof-of-concept; China builds ten factories and ships globally.

So yes, Tesla’s robot may survive inside the U.S., but only through:

• tariffs,

• import bans,

• national-security excuses,

and whatever industrial-policy tool Washington can wield.

It won’t survive on merit. It will survive on protectionism.

But step outside the U.S.?

Why would any ASEAN, Middle Eastern, African, or Latin American country buy a Tesla robot when Unitree, UBTech, XPeng, and others are offering machines that are:

• cheaper,

• and available now — not in 2027,

• generations ahead and more advanced by 2027.

You think Indonesia, Malaysia, Brazil, Mexico, Turkey, or Saudi Arabia is going to pay double the price for a worse robot just to keep Washington happy? You think they’re going to turn down a $12k Unitree or $16k UBTech because Trump tries to bully them into paying for a $35k American robot instead?

The U.S. will absolutely try to pressure, coerce, or outright threaten developing countries into “buying American” — the same way it pressures them on telecom, semiconductors, energy infrastructure, ports, and industrial policy. But this time I don’t think most countries will obey.

They have options now.

By the time the U.S. finally ships its first commercially deployable humanoids in 2–3 years, the rest of the world will already be locked into the Chinese robotic ecosystem — Unitree frames, Chinese actuators, Chinese SDKs, Chinese AI integration, Chinese supply chains.

The EU, Australia, Japan, South Korea, and Taiwan — effectively U.S. satellites — may follow Washington’s orders and switch to American robots. Maybe. If their economies in two years can still afford it.

Everyone else?

Forget it.

Forcing U.S. factories and businesses to buy “American-only” humanoid robots — which will be more expensive and less advanced — will cripple U.S. competitiveness across the board.

If American companies are stuck paying $30k–$40k per unit for less capable Tesla or U.S.-made robots, while factories in China, Malaysia, Indonesia, Brazil, Vietnam, Mexico, Turkey, and everywhere across the Global South are deploying $12k–$18k Chinese robots at scale, the cost gap between U.S. and foreign manufacturing will explode. And it won’t stop at robotics — it will cascade downstream into every single sector that depends on automation:

• logistics
• warehousing
• construction
• agriculture
• textiles
• electronics assembly
• packaging
• even retail, service, and hospitality

If U.S. firms are locked into a high-cost, low-capability robotic ecosystem while the rest of the world uses cheaper, better, faster machines, then every American industry that relies on automation gets structurally handicapped. That’s not just a disadvantage — that’s YUGE and permanent.

So Trump’s protectionism will actually accelerate the decline of U.S. manufacturing competitiveness. Because the battlefield is no longer labor cost — the battlefield is automation cost.

And China will win that fight by orders of magnitude.

This is also why I doubt even America’s closest aligned countries will follow U.S. orders when Washington eventually demands they drop Chinese robots and buy American ones. Unless they’ve developed a death wish for their own industries, they simply can’t afford to sabotage themselves like that — especially when their economies will likely be in even worse shape two years from now.

Except Europe. Europe will probably obey, because their heads are shoved so far up America’s arse they can’t even think straight — and then there’s that incessant, obnoxious demand of theirs: “You must stop be friend with Russia first or we won’t play with you!”

In my opinion China will eventually move toward some form of universal income or redistribution. Once robots replace most human labor, the state will simply “tax” robotic productivity — in whatever form it chooses — and channel that output back to the population. China can do that because the government actually has the authority, the ideology, and the political structure to redistribute.

After all, that’s the logical endgame of communism, isn’t it? A fully automated productive base supporting human welfare.

America? No such luck.

In the U.S., the elites — the top 5%, or really the top 1% — will own the robots. They’ll own the factories, the logistics chains, the land, the means of production, and the automated labor force. Everyone else below them will get… nothing. No jobs, no prospects, no future, nada. Just a growing underclass structurally locked out of the new automated economy, where human labor is obsolete and redundant.

And unlike China, the U.S. government can’t — and won’t — redistribute. It won’t tax robots because it won’t tax the ultra-rich. It won’t implement a universal income. It won’t structurally rebalance anything. The millions displaced by automation will simply be left to rot — not because the technology is bad, but because the political system is incapable of adapting to it.

And if there’s one thing I’ve learned comparing Americans and Chinese: Americans are astonishingly ideologically rigid, stubbornly wedded to outdated principles even when reality punishes them. The Chinese, by contrast, are pragmatic — willing to bend, adapt, and change. That adaptability will matter a lot when robots replace human labor and make capitalism, as we know it, obsolete.

That’s why America is panicking. They know they can’t adapt.

Ian Comments: again, China is ahead in most technologies and they have an unparalleled ability to scale. Once they scale, no one else can compete. You either find a place where you’re ahead and concentrate on staying ahead, or you find a niche. It used to be that China didn’t feel the need to be ahead in everything, but Trump, in his first time, with his sanctions, changed that. The Chinese realized they had to own full stack of everything.

One side effect of this is that Musk isn’t going to get his one trillion dollar payday. It’s based on him hitting targets, including in humanoid robots which he won’t be able to make, because Tesla’s too far behind and lacks the ability to scale.

More on the transition away from labor-distribution capitalism soon.

And great piece by KT. Thanks for letting me post it.

Page 1 of 3

Next

Powered by WordPress & Theme by Anders Norén