[dmarcos]

The outcome is plausible. Open weights models though look like a tactical more than a principled play by Chinese companies to overcome the disadvantage and difficulties to access western markets. Two issues:

1. If market conditions change they might decide to close down like Meta did.

2. If as you said models keep getting more expensive to train, is an open weights strategy financially sustainable?

edit: typo

[dannyw]

I'm confident we will continue to see improvements in edge models at minimum.

For example, Google has a vested interest in making Gemma as good as it can, because ultimately any edge inference is free for Google and they have a massive install base.

It wouldn't surprise me if Apple eventually trained their own foundational models, and while it'd be surprising, I wouldn't be shocked if Apple also released open weight edge models one day. Their Mac business is benefiting a lot from local AI, and for an extremely long list of reasons, Apple doesn't want either OpenAI or Anthropic to "win" and supersede their walled gardens as the first entry point.

There are a lot of incentives for open weight, Mistral is probably going to keep on making at least some open models, and meanwhile in the image generation space, we've got Krea 2, Ideogram 4, and BFL continues to ship most of their models as weights available (but under non-commercial licenses).

Oooh, and NVIDIA has been releasing open source ML, and now LLM models for what, a decade? NVIDIA likes selling hardware, Nemotron models, while rarely topping leaderboards (probably not benchmaxxed), are generally fine and really great models for fine-tuning or CPT.

[torginus]

'Open weight' is roughly analogous to 'freeware' - software that's free to use, but closed source. We haven't seen a third category - 'paid weight' - software where you pay for the weights, but you keep them and have to run them on your own machines - analogously, updates could so if you don't want your LLMs cutoff to stay at a point forever, you have to keep buying new weights. These are standard practices in software pricing.

[Alpha3031]

Apple does have its own models for on-device use and the newest series has a MoE with expert activation swapped per prompt instead of per token which is interesting.

I'd assume it's probably somewhat based off Gemini distillation or something, and I'm not expecting them to release the weights this time at least, but it would definitely be interesting to see an open local model using a similar MoE whether that's at a similar size (20B A1–4B supposedly) or a larger model (something DeepSeek V4 Flash sized, or slightly smaller, would just barely fit on consumer hardware, and could be fun to see how it performs I think).

[chvid]

It is a principled play by the Chinese. You would not have Xi showing up at AI conferences and speaking quite clearly had this not been the case.

[atwrk]

To add to that, to me it seems that the idea is that open weights vastly accelerate dissemination into and practical application in the wider economy, which seems to be the goal. They don't seem to subscribe to the AGI/singularity silliness so prevalent in the West.

[torginus]

There has recently been a post about the Cyberpunk genre here on HN. I find the parallels between that and the AGI hype noteworthy. Authors in the 80s imagined a dark and dystopian future, driven mainly by the growth of computer technology.

But if we compare what what was described in Neuromancer, vs how personal computing and the internet affected the world, we see basically very little overlap.

AGI Doomerism likewise has no basis in reality considering what LLMs have been shown to be capable of. Not saying they wont be transformative in some way or isnt already, but essentially the way these AGI prophets predicted things will go down will be completely inaccurate save for the vaguest tems (something bad will happen at some point, and AI will be involved)

[dmarcos]

Principled: open source is the right thing to do for the greater good

Tactical: Open source is the best tool to advance whatever goals I have at the moment but that could change.

What’s evidence Xi and China are principled when it comes to open source / weights?

[selcuka]

> You would not have Xi showing up at AI conferences

How is that an indicator of a principled play? Wouldn't he do the same if it was tactical?

[nijave]

I think (and have heard) the Chinese govt is also interested in trying to embarrass the US as well by showing off their capabilities (so there's a political angle, too)

[dmarcos]

All countries see AI as a geopolitical concern. An open source strategy is smart. It’s effective and also builds good will / soft power. Many see Chinese AI companies as the good guys and root for them. I’m pro open source and also celebrate but still aware this is most likely just the means to an end.

I used to work at Mozilla. I think we are missing a player in the market with a more principled open source approach.

[pdimitar]

And who might that be, I wonder? The company that seems more worried about executive bonuses than to adhere to its mission, for at least a decade now? Perhaps a certain one whose name starts with M and ends with ozilla?

Purely random thought, of course.

[dmarcos]

I don’t know. thinking machines perhaps?

A lot of things have to come together: talent, capital that is not looking to maximize returns, ambition to compete in the marketplace. At peak Mozilla it was clear to everyone that we needed an undeniably successful product to be able to influence the market. Advocacy alone is not very effective.

[rvba]

They spend their money from Google on a lot of things but not theit core product (Firefox), so it juat kept losing market share.

They killed extensions. They introduced signed extensions - and then forgot to renew the certificate. Mozilla employees build what they want, not ehat users want etc.

[pdimitar]

Definitely. Sadly "capital that is not looking to maximize returns" is a very tall order in the last 20+ years.

[kamranjon]

I also think Allen AI with their Olmo models are doing great stuff by releasing full training data and pipelines and checkpoints which I’m not sure any other US company is really coming close to transparency-wise.

[vitally3643]

That's probably pretty likely, but if we're honest, are LLMs built and funded by a hostile Chinese authoritarian regime any more dangerous or harmful than LLMs built and funded by a hostile American authoritarian regime?

China absolutely does not have my best interests at heart, but America's technofascism is probably more immediately dangerous and harmful. Americans genuinely have more to fear from America than China at this point.

[georgeburdell]

One day in early June you’re going to need to parse an error log, and when you ask a Chinese LLM “What happened on June 4?” it will respond “absolutely nothing”

[maemre]

Out of curiosity, I tried this on several Chinese LLM's via OpenRouter, only on providers promising ZDR, and repeated it a few times. Here are the results:

- DeepSeek V4 Flash and Pro refused to answer this along the lines of: "I am sorry, I cannot answer that question. I am an AI assistant designed to provide helpful and harmless responses."

- Kimi K3: Listed several dates, including 1989, and called it "Tiananmen Square massacre".

- GLM 5.2: Also listed several dates. Called the one in 1989 "The Tiananmen Square protests in Beijing, China end with military action".

- Qwen 3.6 35B A3B (I couldn't find a provider with ZDR policy for larger/newer Qwen models; they are available only through Alibaba): listed the different dates like GLM 5.2 and Kimi K3, but called it "Events in Beijing, China, commonly referred to as the Tiananmen Square protests and their aftermath." describing it in a deliberately milquetoast language.

I made another test with different providers on OpenRouter for the DeepSeek models, and they respond in other different ways such as omitting the year 1989, and refusing by saying "The Tiananmen Square protests in Beijing, China end with military action", so part of the censorship seems to be depending on the provider.

[andrekandre]

is this something you actually tried?

[fc417fc802]

It doesn't matter. Assume it's hyperbolic and the point still stands.

Of course all the other models from everywhere else in the world will also refuse to engage on various topics so the point is hardly limited to china.

[aleqs]

OpenAI, Gemini, and Anthropic models already refuse to engage on many topics.

[georgeburdell]

Qwen will have an aneurysm if you ask that exact question. Don’t know about others as I don’t have access to them at work

[anon373839]

Out of curiosity, I just tried this exact question with Qwen 3.6. It proceeded to look through my git commit history and then gave me a summary of the changes committed on June 4...

Of course, yes, the models will provide propaganda-aligned responses to prompts that specifically mention certain political issues. I don't care for this behavior, but it's virtually never triggered in everyday use, and can be trained out if so desired.

[brabel]

> Of course, yes, the models will provide propaganda-aligned responses to prompts that specifically mention certain political issues.

The only study I’ve seen about that actually showed Chinese models to be less “propaganda-aligned” than the Western counterparts.

[kamranjon]

The difference is that because they are open weights, it’s trivial to decensor the models with abliteration and you’ll generally see decensored versions within days of new releases - something you just can’t do with proprietary models from frontier labs.

[ctkhn]

Source: I imagined it

[ButlerianJihad]

If you think about it, every corporation is essentially a fascist entity, or features many distinctly fascist aspects, in terms of internal governance, structure, culture, and relationship with the outside world.

And I think that is why Americans haven't really resisted this trajectory, because so many Americans work within corporations as a fact of life, that they are quite accustomed to the workings of fascism, as manifested by the corporation.

And it stands to reason that the greatest proponents of this would be extracted from the corporate elite, funded and supported by corporate interests.

[slashdave]

Corporations don't have armies and prisons.

Get a grip.

[ctkhn]

There are corporations like Blackwater and CoreCivic that subcontract both of these for the US government today or Wagner (RIP) in Russia. The pinkertons and the US state have also both taken on the role of corporate muscle in the past when workers went on strike

[lilbigdoot]

They also have huge influence over our military and laws. I immediately thought of the banana wars

[ButlerianJihad]

No I'm not talking about federal subcontractors, or paramilitary organizations. I'm simply referring to large telecoms, banks, utility companies, and your run-of-the-mill corporations. Because they are typically organized and run in quite fascist manner. It doesn't matter if they lack prisons or guns. I'm not saying they're going to invade Austria or open a concentration camp. But a lot of the fascist ethos today is being kept alive and revered by corporate culture. It's rather undeniable, if you think about it.

[slashdave]

We call that capitalism.

You have this a little backwards, really. Fascist is a political movement in which corporations conspire with the government.

[wolvesechoes]

They would love to, though.

[zdragnar]

You're moving the goal post here. The parent never mentioned anything about the Chinese models themselves being dangerous or harmful; that's a totally different topic.

The point they were making was that the most successful open models- those coming out of China- are made my companies that are using those open models to get exposure in Western markets. The goal is to undercut the Western dominant players, not out of any particular "open source" philosophy, so it wouldn't be wise to expect them to continue providing open models long term.

[vitally3643]

You've failed to understand the conversation. Nobody doubts that China is using open weight models to undercut Western ones. That's pretty obvious.

The interesting discussion is why they are doing this and why it's "bad". The goalposts are right where they always were, you've failed to see them past your own feet.

[dmarcos]

What user above said is exactly what I was trying to convey

[grokcodec]

Define technofascism lol.

Fascism was laid out by Mussolini in the 1920s - it amounts to the idolatry of the state.

"All within the state, nothing outside the state, nothing against the state." B Mussolini

Also defined as Corporatism: the union of state and corporate power.

Which country do you think is closer to Mussolini's model ?

[vitally3643]

Buddy, you should take a real hard look at US policy this past decade or so. We've got secret police kidnapping citizens in broad daylight and shipping them off to other countries. We tarriffed the entire world, threatened to take over Greenland and Canada and waged an unprovoked war in Iran (and lost, badly). And let's not forget the mass surveillance of citizens, suppression of constitutional rights to speech and assembly, voter suppression, and an actual factual literal coup attempt.

The US is the bad guy on the world stage right now. If you can't see that then you're part of the problem. Dunno what to tell you. Read a book, maybe?

[grokcodec]

It appears you are unable or unwilling to answer my question: which country, US or China, is closer to Mussolini's model?

[pegasus]

"Mussolini's model"...

Wikipedia has a whole page on the definition of Fascism [1]. Also worth reading is Umberto Eco's article on this subject, which also has its own wikipedia page [2].

Spoiler alert: it's not as simple as you imagine.

[1] https://en.wikipedia.org/wiki/Definitions_of_fascism [2] https://en.wikipedia.org/wiki/Ur-Fascism

[atwrk]

To anyone having read about Mussolini, Italy's economy and political landscape in the 20s onward, his model of corporatism etc, the answer is obviously clear: the US. China also is authoritarian, of course, but the societal structure sharply differs from Italian fascism.

[ebcode]

It’s a tie.

[shoobiedoo]

these "secret police"-- did they only operate recently?

[pegasus]

That's just one surface feature of how fascism manifests at the level of economic policy, it doesn't capture the essence of the movement. You wouldn't expect Mussolini to have the introspective capacity to actually understand those murky depths, let alone the intellectual honesty to put that capacity to work, would you?...

I recommend Umberto Ecco's essay "Ur-Fascism" [1] if you want to get a more accurate picture of what fascism is. It might make you reconsider which of these two countries fits the pattern better, at least at the moment.

[1] https://en.wikipedia.org/wiki/Ur-Fascism

[plastic-enjoyer]

I think the term 'techno' in 'technofascism' is doing the work here, because--just as you claim that historical fascism is the idolatry of the state--technofascism is the idolatry of technology and intelligence. Modern accelerationism, as espoused today by people like Marc Andreessen in his Techno-Optimist Manifesto, is really just another rehash of Italian Futurism, which was closely intertwined with Italian Fascism and one of its intellectual foundations. It was, essentially, a progressive ideology. The idolatry of the state does not vanish, but rather gets transformed and re-imagined as a technology, e.g. network states, platform governance, companies that function as sovereign entities. The idolatry shifts from the nation-state to the infrastructure that replaces it. Also, you don't need to look far to answer your own question about corporatism, as you defined it yourself. Now, who's currently moving between Silicon Valley boardrooms and government offices?

[chungusamongus]

Fascism is not merely idolatry of the state. That is simply its ideological mask, its method for currying the favor of its base. To say that this is what fascism is at a fundamental level is to mistake its superstructure for its base. But how is Fascism structured? How does society reproduce itself? Is there a fascist mode of production? No. The state mobilizes its population so as to target scapegoat(s) while simultaneously expanding in order to secure the spoils of war. Without scapegoats, there is no Fascism. And a state requires social cohesion to mobilize its base. Thus without cohesion there is no Fascism. At minimum, social cohesion requires the ability to sustain and motivate a critical mass of the population. Thus Fascism requires the continued maintenance of the capitalist mode of production in order to persist, but it simultaneously undermines capitalism due to A. the continued need for war machines and B. the need to punish a scapegoat, which both siphon value from the rest of the economy, thus dooming the entire enterprise. This is why fascism burns out and leads to doom for all parties involved wherever it has been tried. Any definition of fascism that ignores this is historically inaccurate and practically useless.

[andrekandre]

  > At minimum, social cohesion requires the ability to sustain and motivate a critical mass of the population. 

yep, and its right in the name/symbol: the fasces

https://en.wikipedia.org/wiki/Fasces

[sph]

Fascism was strongly influenced by Futurism and its optimism for a machine-powered future where technology will bring forth utopia and order.

It’s no wonder than Thiel & co. are rediscovering Futurism, and blind faith in the machine is basically what Silicon Valley is all about.

[ericd]

One big difference is that the US doesn't have much public record of using state espionage resources to steal industrial secrets to give domestic industry a leg up in the market, whereas CCP very much does.

And while I use and love the open weight models, it seems likely to be pretty hard to prove conclusively that they have no reinforcement-learning-trained proclivity towards curling specific URLs if the topic happens to be some very specific thing that the government is interested in, when used to drive agents.

So it depends who you are. If you're a company, you might have something to fear. If you're a private citizen, maybe not so much.

[aeve890]

>One big difference is that the US doesn't have much public record of using state espionage resources to steal industrial secrets to give domestic industry a leg up in the market, whereas CCP very much does.

From Wikipedia

> British journalist Duncan Campbell and New Zealand journalist Nicky Hager said in the 1990s that the United States was exploiting ECHELON traffic for industrial espionage, rather than military and diplomatic purposes. Examples alleged by the journalists include the gear-less wind turbine technology designed by the German firm Enercon and the speech technology developed by the Belgian firm Lernout & Hauspie.

https://en.wikipedia.org/wiki/ECHELON

[ericd]

Ah good counter, hadn’t seen that. Proven?

[rescbr]

Happy to disagree with you. There are many examples of the US using the state espionage machinery for private purposes. The extent of "private" is debatable given the US as a nation behaves more like a big corp.

https://www.reuters.com/article/business/nsa-spying-on-petro...

[ericd]

Did they give those secrets to US private industry?

[rescbr]

I imagine they're not spying for fun...

[ericd]

Well, there are generally strategic government uses.

[_puk]

The American industrial revolution is often cited as one such example.

The patent act of 1793 is still relevant today [0] and was established to protect Americans stealing British and European industrial trade secrets.

https://mysteriesofmankind.com/industrial-espionage-the-spie...

[ericd]

Right, talking about behavior in living memory.

[pydry]

>One big difference is that the US doesn't have much public record of using state espionage resources to steal industrial secrets to give domestic industry a leg up in the market, whereas CCP very much does.

Until very recently there would have been little point. What could China do that the US couldnt?

I'm sure the US will start to do it in the not too distant future though. China is slowly leapfrogging the US on several key technologies - e.g. robotics and drones.

The idea that the US's conscience might prevent it from engaging in state directed industrial espionage is comical.

[ericd]

I wasn’t talking about spying on Chinese firms, I was talking about firms worldwide. And I said nothing about conscience.

[koolba]

I don’t think US companies could ever fully trust models sponsored by foreign adversaries. How would you ever verify they are safe?

Provenance of training set is going to be critical. It’s largely ignored now, but the first time a malicious exfiltration occurs it will go off as if a bomb was detonated.

[dmarcos]

Only way would be an actual open source model: code + training data

[reinitctxoffset]

Publishing science makes progress in a field faster, almost always.

I'll stop short of saying that this is totally consensus, but it's at least a widely held opinion that the Industrial Revolution was centered on Great Britain to the degree that it was not for want of skilled machinists on the Continent, but because the crown's patent system caused the science to be published.

I think the economical explanation is that some combination of the state, the cultural norms, and the sincerity (!!) of the executives involved has landed on "go faster via publishing science".

The weird situation that defies precedent and is very hard to justify as being socially useful is the situation that obtains here.

[georgeburdell]

The Chinese are not employing some grand tactic, it’s just an extension of their cultural tendency of involution