edit: > Commerce Secretary Howard Lutnick (lol)
Fable was out for 3 days, not really long enough for us to properly evaluate it, but the "Sorry we had to remove Fable. Read more. (because it's too powerful btw)" is loudly shown every chance they get for weeks. It creates a halo.
Reminiscent of the 1999 Apple G4 commercial where they displayed it next to military tanks. "For the first time in history, a personal computer has been classified as a weapon by the U.S. government."
And even if they solved that (it will take them 40 years of bickering), it’s not something you can top-down create, unless you want the AI equivalent of the Yugo.
China has the drive and the ability. It’s communist in name only, and has truly turned into a hyper-capitalist super producer (less government spending as % of GDP than even the US).
It will beat out both the EU and US and sweep both the digital economy (the US’s golden goose) and industrial economy (the EU’s golden goose) over the next 20 years.
These words don’t mean what they use to anymore. Newspeak is in full swing. Words still sound the same and are written in the same way but now mean something completely different. If Mao and Stalin were alive, they would be nodding approvingly.
I hope the Chinese models catch up soon so I can stop contributing to the American economy.
"When you see that in order to produce, you need to obtain permission from men who produce nothing - When you see that money is flowing to those who deal, not in goods, but in favors - When you see that men get richer by graft and by pull than by work, and your laws don’t protect you against them, but protect them against you - When you see corruption being rewarded and honesty becoming a self-sacrifice - You may know that your society is doomed." ― Ayn Rand, Atlas Shrugged
Let's hope this creates a bit more fire under the asses of other countries
Or may they'll decide to be a little more quiet and less end-of-the-world-is-nigh-if-you-use-our-services?
Play stupid games win stupid prizes.
I assume "trusted partners" means, "companies that have bribed Trump an appropriate amount". A few million for the inauguration, a few million for the ballroom, a few million on a movie about Melania, the don wants a taste.
Back when the administration hit Mythos/Fable with the surprise ban, I figured this would be the endgame. They'd keep Anthropic tied up until a competitor had a roughly comparable model ready, then gate them the same.
10 days for an appeal is too long. Company is a startup, so no team/enterprise support.
In the mass-marketing world it's less about who's right or wrong but who is perceived by the population to be pulling the leavers on the front page again.
Asterisk the size of a Mac truck.
Also this administration having say over who gets access to what AI is just so much more grift corruption and picking your favorites / destroying others, for these incdecent undemocratic in American grifters who've seized our state.
Maybe, maybe not. Tech stocks are mostly vibes-based now, reality isn't really a concern for them.
Does it mean US is allowing accessing to governments' exclusive list?
All these dangers were known and predicted.
There's an uncanny parallel with the climate crisis.
Fatalistic somnambulism.
I know it’s a bold statement but look at this timing and their valuations going south.
The rate that the ruling class ran into crony capitalism at the first chance they got is something that needs to be remembered. They'll try to act like they were always against it at some point in the near future.
> The two biggest enemies of the free market are two separate groups: my academic colleagues and business people. Business people are enemies of free markets, not friends.
> [...]
> The business people are just the opposite. They're all in favor of freedom for everybody else, and at the drop of a hat you can get any leading businessman to give you an eloquent speech on the virtues of a free market. But when it comes to their own business, they want to go down to Washington and get a special tariff to protect their business. They want a special tax deduction. They want a tax subsidy. And Chrysler is on the verge of failing, which it should have done. It should have been allowed to fail. Chrysler goes down and exercises political influence and tries to get the government to lend it money to subsidize it.
> So businessmen in general — not all, there have been some notable exceptions — and I don't want to include everybody. But in the main, most businessmen are enemies of free markets.
EDIT: I agree that it should require an act of Congress to explicitly delegate this power.
Every one of those is by a regulatory agency that was explicitly empowered by Congress to do such regulation.
Plus, they're relying on the "math is a weapon" law to ban "export" of the models.
Should ever new "weapon" invented require a new act of Congress? We've considered software subject this act since the 90s.
If everyone making AI is screaming up and down that we are in an AI arms race creating dangerous entities that will determine the fate of the world is the government just supposed to ignore them?
-- GOP probably
The only thing that changed is people are writing articles about it in the news media.
Is there any policy from this admin you don't support?
It’s not about size, it’s about where it chooses to operate
Genuine question: if Democrats take power, do you expect them to be more interventionist or less interventionist with respect to AI? Bernie's jockeying leads me to suspect "more", but I could very well be wrong.
(FWIW I personally think modern AI falls in the small realm of potentially dangerous technologies that merit careful, ideally bipartisan, government oversight)
The current admin flies by the seat of their pants and at least creates the perception of political decision making.
https://bidenwhitehouse.archives.gov/briefing-room/statement...
Free market? Small government? Big police state, trillions in defense contractor grift, unsustainable tax breaks to the wealthiest leading to massive spending deficits... all while doing everything to erode access to education, healthcare and basic services.
It is just utterly baffling to me. I'm... well along the spectrum... so people not responding correctly to obvious information is just something I've gotten used to but just... wow.
edit: typo
People get quite a kick out of seeing people they don't like get hurt. They can stay entertained by that for a long time until it bites them.
Only now is it finally biting with the collapse of the rural medical clinics, the war induced spike in the price of gasoline, etc.
That's probably playing a big part in the seeming shift in the electorate in every election.
A clear regulatory framework to operate within allows businesses to operate within it rather than get surprised by the King's whims upending their business on every few Fridays. If you expect regulation will eventually happen, pushing for it to happen on terms you're able to comply with rather than as haphazard surprises is pretty sensible.
The US government was created to protect the interests of rich, white, male slave owners. And if you look at Louisiana State Penintentiary (often called "Angola"), which is essentialy a Southern plantation with forced labor, you realize not as much has changed as you might otherwise think.
https://www.ecfr.gov/current/title-22/chapter-I/subchapter-M...
I don’t support the admin but if you are unwilling to engage with reality then that is on you.
We are talking about the government giving exclusive access the the most transformative technology in human history to a select group of companies with no formal policy as to how you gain access, lose access, what you are expected to do with that access or commitment to transparency around any of this.
congress has abdicated its role entirely.
Dajcie mi człowieka, a paragraf się znajdzie
Give me the man and I will give you the case against him> My paramount object in this struggle is to save the Union, and is not either to save or to destroy slavery. If I could save the Union without freeing any slave I would do it, and if I could save it by freeing all the slaves I would do it; and if I could save it by freeing some and leaving others alone, I would also do that.
While chattel slavery ended when the 13th Amendment was ratified in 1865, slavery continued through debt bondage and convict leasing up until 1941 where FDR suddenly decided to aggressively prosecute the practice for fear of the Japanese using it for propaganda value. I'm referring to Circular 3591 [2]. And while that heavily curtailed abuse (eg by locking people up essentially indefinitely for "vagrancy" or imaged debts), forced prison labor continues to this day, including private companies profiting from prison labor.
Also, while the Confederacy lost the Civil War, the South arguably won. Reconstruction saw severe curtailment of newly-established civil rights for former enslaved people. And after Reconstruction came Jim Crow until the 1960s.
[1]: https://www.loc.gov/collections/abraham-lincoln-papers/artic...
[2]: https://www.endslaverynow.org/blog/articles/state-imposed-fo...
It’s hard to take this seriously.
But “The South Will Rise Again” folks will be happy to hear it.
How well does it stand up to Mythos?
The Democratic party is more anti-ai than the Republican party and unfortunately both of them are increasingly responding to astroturfed populism.
Do you think Bernie Sanders in AOC are pro-ai? Are you kidding me? Have you seen what they say and the legislation they propose?
Not even on the same playing field. They just can’t be compared, they’re incomparable.
If it takes Trump to force people to educate themselves on how the US government actually works then I guess that is at least one good thing to come out of this.
But you have to admit this policy seems ad hoc and creates the impression of opening a wide door for corruption.
This was absolutely predictable, and many people made that prediction. The patterns of what tech they apply export controls to is actually pretty legible. People have long tried to ride the gray area so that their tech is not subject to export controls. Sometimes it works, sometimes it doesn’t. This was never not going to be subject to it.
Usually companies do this stuff quietly with lots of small new rules via Congress creating barriers to entry or through national security angles like the Chips act which funneled money and tax breaks to huge weathy companies, or Boeing, or the car industry, etc.
Anthropic and OpenAI went hard in the paint pushing for AI safety and it backfired into hurting their companies rather than protecting their interests.
1: https://www.usnews.com/opinion/thomas-jefferson-street/artic...
(edit: not that these models are equivalent to missiles oc)
So what’s left? Where does that decision making go? Turns out the executive, so that’s what we’ve been seeing and it’s largely uncontested. This should have been obvious to most people going into this, particularly if they understood Trumps platform or Project 2025.
Chevron and the unitary executive theory have essentially nothing to do with each other.
I’m still not sure what point is attempting to be made here.
So, the gap has been filled largely by executive orders.
I am simply arguing that there is a more effective approach to creating a regulatory framework here. It doesn’t have to take that much effort. But this random announcement with no prior or current guidelines seems ad hoc to me.
Well, by your own logic, there's a new a/b test running right now. Its results aren't exactly going your way.
This is why arguments about this go in circles. You either argue from a pure theoretic POV back and forth, or you go off data - at which point, bringing up every failure of free markets (like, obviously, US healthcare) is dismissed as "not really a free market"
Even the theoreticians on the free-market side are far less solid than.. all the other sides (behavioural economics, information asymmetry.. even Marx) but I regard it as deeply unpragmatic when there's so much data out there indicating what actually happens in the real world when you go one way or the other.
You can prove the logic part starting from the assumptions. It's also falsifiable. I just mentioned it was literally the most controlled test on human society you could make. We tested by splitting societies at the level of the entire planet, states and cities.
US healthcare is mostly not a free market; by free market, at minimum, I mean that the quantities and prices (ideally even the quality) are not set. The US healthcare system has a fixed number of practitioners who can get a license every year. This is as far as a market can be from being free (together with the case of having price controls). In fact, free market theory predicts that when you restrict quantity, you get higher prices for the same quality. It literally predicts the US situation.
It's funny you mention Marx, given I regard most of his claims as either unfalsifiable or easily proven false.
I mean, it's not. In a free market you'd have a choice of insurance providers rather than having to take whatever plan your employer offers, and you'd have some idea of what the hospital is going to charge you beforehand rather than receiving random bills for weeks.
isn't that the most free market thing ever?
It's pretty problematic to not make it more widely available at least to US businesses, and there is not even a vetting process to get approved quickly and easily. If this is the new norm, the intended or unintended consequences of this type of gatekeeping will be an unprecedented consolidation of power amongst the largest corporations. Even more than we have seen over the last 20 years.
Yes, sure, it is I who is blind...
Bernie and AOC (which aren't DNC mainstream, but prominent) had just pushed for a moratorium on "AI data centers" with a definition that includes "that are used for the development or operation of AI models at scale" (trivially sidesteppable by "we build this GPU farm to sell to whoever bids for compute" - which is actually true), plus a bunch of fancy extras bundled in like "The government must review and approve AI products before they are released to ensure that AI products are safe and effective.", while lacking actual definition of "AI" (given that we had "AI" systems since '50s).
Here's the full text: https://www.sanders.senate.gov/wp-content/uploads/AI-Data-Ce...
Yeah, the bill has a cause - it recognizes some pain points. But then it haphazardly tries to address symptoms instead of underlying issues (environmental regulations, utility pricing, land use, job security), while pushing vaguely defined regulations that allow arbitrary application. As if misdirected measures and poorly defined laws aren't already a giant issue.
That was the last major thing the Democrats did, and healthcare has gotten substantially worse...but at least it's well regulated now.
The what? More like "the whims of an eighty year old in cognitive decline and those wishing to curry or keep his favor" - quite an expansive definition of "political decision making".
The previous administration was totally not exactly what's described here...
But yes, Biden was old and cognitively not well. But his “whims” didn’t exist much, and they were always fairly reasonable. Trump is the most unreasonable president, most likely in US history. I would even categorize Andrew Jackson as more restrained.
Is that a joke? We're back in a spat with Iran because Obama refused to engage with Congress, as required by our constitution, to enter the USA in any binding deal.
Any AI actions from the next admin is going to be executive yolos.
It eliminated the idea of a “gap” that executive agencies (whatever do you mean by “executive orders” in this context?) could be deferred to to fill.
My goodness, this is basic stuff in this Constitutional domain.
This has, in effect, diverted more power to president Trump. Not less.
I would love to see any examples of democrats being crony capitalist of the same degree as the current administration. They are selling pardons, hosting crypto dinners where the seating arrangement is based on how many coins you bought, seizing the equity of private companies like intel and the revenue of corporations in exchange for lifting export restrictions like to nvidia, holding up M&A transactions for rivals and approving transactions for allies, ignoring laws passed by congress (like the ban on TikTok) to let allies buy the assets.
I am not a favor of state capture by the capitalist class. But democrats are house cats.
No you wouldn't. You've established a subjective measure to the discussion by which you can refuse to acknowledge any example provided.
How about Biden leaving a bunch of equipment in Afghanistan so his weapons manufacturing buddies could get new manufacturing contracts?
How about Biden using the EPA to shut down eco friendly products in favor of his buddies' company's spill remediation that doesn't work nearly as well?
Oh well, those individually aren't "to the same degree" even if there are many more of them that cumulatively add up to a greater degree.
There are countless examples of this, as lots of others have pointed out to you across this thread. You're just choosing to walk them all back into a corner of "but it's not as bad because <insert additional criterion>."
Your opinion as given is entirely based around Trump, not government.
It's simple fanaticism. You've chosen a football team. Nothing will convince you that their quarterback sucks because he was just having a bad day every game of the season.
You can quickly stop doing this. Go research the counterpoints and see if you aren't being selective. It will require some honesty with yourself to do so. Most people find that very difficult.
Less judicial deference = more executive power? How does that story go?
That would be your politics coloring your memory.
If US companies have a large unfair advantage such that domestic competitors are no longer able to compete, then wouldn't it make sense for governments around the world to ban or tariff US products and services?
If I was responsible for national economic policy, I would place this at the top of my non-emergency agenda. The world needs to act quick before their industries fail.
Most of the rest of the world is too heavily dependent on US digital services to tax them more heavily. From social media to hyperscaler's clouds the US is dominant and stuff will just stop working if they get taxed. There would be a huge pushback from businesses if their government increased the cost of things like AWS.
Edit: edit to say tax more heavily
Didn’t he already lost that case? And, all the other countries know that.
Skilled engineers nowadays demand between southern and central European prices at the very least.
America mostly produces cheaper ag commodities than the EU but more expensive than South America. Deepseek is already a better search engine than Google. (Not sure if it does Google searches)
Travel the world it’s not American companies gaining market share. I would especially recommend trying Chinese AI or riding in a BYD car and judging for yourself.
It's a cost problem. If you want to try for a SOTA model, you're going to need to spend big time.
Germany spent ~$115B last year on it's defense, roughly 2% of it's GDP.
In contrast, ~$145B was spent last year just on AI infrastructure by Meta, and, well no one talks about Meta winning any AI races.
I think they have very little room for manoeuvre - companies like AWS or Microsoft can simply you are too cocky and we will shutdown infrastructure your country is running on if you don't bend the knee.
The US companies with access have a large unfair advantage over other US companies.
Imagine being Rivian if Tesla gets access, Rocket Lab or Blue Origin others if SpaceX gets access, most of SV if major tech companies get access, ... imagine being a startup.
I can only think of Russia that is decoupled from US at the moment and they are stuck with Putin that still lives with imperial mindset rather than actually being a rival to US
BTW EU will never have a "tech" industry in any meaningful size as long as US have access to EU markets, anyone who eventually got tech industry are those who blocked the US or were blocked by US.
So if US keeps its course, in a few years we may end up with fragmented markets with US blocked out because the US is very unpopular but the current politicians everywhere including in the EU are very pro-US actually hoping that current situation is just a glitch, which is not aligned with what the general population demands and as a result the next elections they will align with anti-Americans.
Europe isn't cooked because it lacks talent, there are untold smart capable people there, it's cooked because it built a social allergy to the very thing it needs most.
Even in the European case, Europe would lose much more than the US would if they closed their markets. Plus, a lot of Europe is either very close to breaking point and unwilling to change (Italy), or rapidly worsening into a crash, and unwilling to change (France).
It's Europe that is dependent on a large trade surplus with the rest of the world, financed by dollar loans to 3rd world countries. Now China is taking away their trade surplus, even directly (meaning Europe has a massive trade deficit to China), and indirectly (replacing demand for European goods, famously cars, everywhere). This is causing large-scale job losses in Europe as well as total disaster for government finances across the block, finances that were unhealthy to begin with.
Now Europe and China are unwilling to lend to the rest of the world (because initially that would make very rich Europeans/the CCP a little bit poorer, by raising inflation quite a bit, thereby raising interest rates, which will move government finances from disaster to catastrophe), so if these money flows are to keep going, either the US MUST export to China, which is not happening, or EU and/or China must loan several times their own GDP to the third world, or the EU and/or China must massively increase their dollar holdings (which will, of course, inflate the Euro and Renminbi something awful whichever way it goes). But either WILL happen, because a crash will do that too. Which is what people mean when they say the worldwide system is on a crash course.
If the US doesn't reverse course soon, I think we'll start seeing large-scale closure of international markets to US companies very soon. Even with US retaliation, there is no other option.
I can't find it right now but I read news just a few month ago that the EU is working on making it easier to invest into EU similiar to how the Petrodollar currently works.
But if the deindustrialization of germany/EU continues as it currently does and US implodes and China has also issues, we will see how the new world order will look like.
China is for sure more resiliant though. The living standards were never as high as what we as germans are used to and they dont demonstrate.
In the USA people have guns and civil issues and a hard divide between city and country sides.
But as Europe/Germany we should be able to increase our bonds right? Investing into solar/energy transformation; Doesn't matter short term how this will end.
And if AI continues as it does with robotics, we might even see in 50-100 years a complete change in system?
USA Bonds are exploding so they might have overplayed their hand already.
why is the commerce secretary making this decision
Thus, it really is an export control issue, but it has nothing to do with offensive capabilities. Offensive capabilities always exist, but pervasive defensibility would upset the asymmetric advantage that attackers, especially the USA, currently have.
There are now Asian models coming , optimized focused on cybersecurity defense at a high level, so I suspect this will be a relatively moot point soon.
LLMs are not great at creating exploits, but they are really good at detecting them. That asymmetry alone is enough to destroy the “offensive capabilities” narrative.
Yes, mythos can find exploitable bugs, even write bench exploits. But real exploits require a good dose of human psychology, and most of the tools needed are off the shelf available anyway. You still need a real cybersecurity expert to effectively weaponize a zero day into a deployable exploit.
But an LLM can inspect payloads, packages, and blobs en masse and find those exploits in a way that was wholly impractical before, so the asymmetric attack advantage is dissolved by strong LLMs.
The USA is trying to protect its cyberwarfare advantage, not protect against attackers. The exact opposite, actually. Porous security is a huge advantage to technologically advanced state actors.
But would another company who is not on the trusted partner list and has less to lose taking on the admin have standing to sue here? On the basis of the export control being illegal and this putting their business at a disadvantage vs. competitors with access
Who are those 100 companies? Clearly they can't compete on merit and have rubbed some hands to be picked as winners...at least for now.
Every single model is (and always was) behind anything from China.
I wish Europe could find one company that can put us on the map but Mistral’s not it. They have repeatedly shown that extra funding doesn’t magically transfer to better models.
My guess is, that Mistral plays fair and doesn't try to extract data from US models, while Chinese companies are not.
Are Google going to end up in a situation where the people working on their models cannot use the models after launch?
I am a British citizen living in Berlin, and even when making apps purely for the German market I still have to go through US export restrictions: https://developer.apple.com/help/app-store-connect/manage-ap...
I'd assume that Deep Mind, being owned by a US company and having US offices, has to care about US law, despite the differences (me: small fry; them: actually having offices there), because the chain of enforcement is still "take it or leave it" at each stage (USG->Google->DeepMind vs USG->Apple->3rd party Apple devs).
Unless you open source it then Google Deep Mind isn't the entity "releasing" it. It's some other Google that's US based, e.g. Google Cloud running the APIs etc.
User -> Google -> Google Deep Mind.
So nope, same restrictions apply.
They’re in effect saying “nothing else is as powerful as what Anthropic put out”. Even though that might not really be the case it’s what it sounds like.
The US might remove access next month in a fit of pique.
The Chinese models look increasingly more reliable and safer.
They just got their market cut to a fraction. Investing in new tech is now very risky because even if things work out you might not be able to sell anything.
There were already serious doubts about ROI for the frontier labs. If they can only sell to 100 or so entities it's over business wise.
What's the endgame here?
So much wasted potential.
And why would I pay Anthropic or OpenAI once consumer hardware gets powerful enough to run an open weight Chinese version of Opus 4.8? Even more so when mobile phones are able to run similar LLMs.
Their financial growth looks doomed. It looks like they will be heavily regulated just like the next missile factory. This is antagonist to VC led turbo growth startup regime.
The spirit is to provide effective tools for the people to resist federal military tyranny, and Mythos seems like it would be a good tool to defend against that, for so many reasons.
That being said many legal scholars say the state militia was intended to be the defense against tyranny not individual citizens because there were government led crackdowns on rebellion under Washington and other presidents from the earliest days of the republic. State militias have the full range of weapons
As an EU company I think I now basically have to consider US AI as hostile and avoid it.
Yes, that’s the only sane conclusion. And yes it does screw the TAM, but it’s not like the AI vendors had actual realistic economic plans to begin with
Unfortunately, Europe should had made this conclusion at least 18 months ago, not now.
Watch out European politics procrastinate for at least one more year hoping that Trump will reverse. Then procrastinate more, because “elections soon, maybe Dems will win and reverse”.
I live in Europe and will never go to work in the US; but EU/UK inability to solve national security problems is beyond pathetic.
If the US commit to handicapping their AI industry like this it's going to destroy the competitiveness of those companies globally. All of those US spending commitments on data centres etc are going to collapse, or americans will need to pay 2x the token cost of the rest of the world. Both very bad options.
Only the frontier AI labs have true access to frontier AI. Everyone else gets a reduced version.
Mythos never was and I don’t think that’s changing.
Isn't this basically the reason that no major tech labs or startups ever come out of Canada or Europe though?
Anthropic is past that. They’d have no trouble moving if they wanted to, except the USG could stop them.
lol
As long as you are in the US? No one protects you. Asset forfeiture, arbitrary sanctions, arbitrary tariffs, your employees can get arrested and deported right off the street if they happen to have non-white skin and live in a city under siege by ICE, and if you oppose the regime or just dare to not kiss the ring, say goodbye to planned mergers.
Rule of law matters, and rule of law is gone in the US.
The US is already seeing energy market distortions from the power use of AI; the UK has a much smaller total electricity supply, both from a lower population and the baseline per person being lower.
Total UK demand in 2023 was 316.8 TWh[0], or an average of about 36 GW. The US currently has 33 GW of data centres, and the AI boom plan, so far as discern actual plans from AI hallucinations in the modern web, is about ten times that.
From the scales people talk about, my expectation is that even the smaller additional supply needed for the constant churn of newly trained frontier models would probably exceed what London alone can manage.
[0] generation less than that, it has imports; chapter 5: https://assets.publishing.service.gov.uk/media/66a7e14da3c2a...
Sadly UK is fully into nanny state "safetyism" culture so I am not convinced it'd be better.
It's like the epidemic of scam nvidia cards being resold without gpu or memory - where do you think those are going?
However, I haven't seen any prominent articles proposing this theory, I haven't seen anyone in the administration gesturing towards this as the reason (but haven't been following too closely)
Do you have any sources?
(And in fact, it seemed like an obvious hypothesis that wasn't getting much air time in the first weekend, but again, I didn't see anybody really staking a claim to it except in a few comments or tweets like this one)
That’s pretty much taking the mask all the way off.
It’s the only motive I have seen that aligns with the actual scenario, and if I -did- have inside information I certainly would not be at liberty to cite it publicly lol.
As someone in the field for 30+ years, I find that, generally speaking, the only groups that use this term are charlatans and military/military-adjacent. The military use has been slowly leaking into the general public, but I quite wish it would stop.
A lawsuit would be a hard sell though, because Anthropic themselves argued that the technology is dangerous. Even if many people on HN might think that Anthropic was scaremongering about Mythos, a court is probably going to take their assessment at face value, and courts are loathe to find against governments in cases of national security.
There's also the issue that these models are getting better through an iterative process, so even if the line between GPT 5.5 and Fable/GPT 5.6 is somewhat arbitrary, it doesn't mean that the government shouldn't be able to draw a line at all. So you're left arguing that they drew the line too early, which is subjective.
They never released mythos 5 to the general public. And they never will.
They said Mythos was dangerous, not Fable which is what got banned.
No. Only if those employees have a green card and the company must not only take on that responsibility but ensure other employees are denied access. Otherwise the company would be subject to millions in fines.
US export laws are no fuckin' joke like everyone here seems to think they are.
It's really frustrating to read pages of comments rooted in emotion and no understanding of the existing laws.
It is technically impossible. Many of the researchers working on the models aren't US citizens. That's not just within Anthropic. It'd make things 100% worse.
Claiming ignorance is a good way to pay tens of millions of $ in fines or do prison time.
Here's one from TWO DAYS AGO:
https://www.justice.gov/usao-edny/pr/manager-us-freight-forw...
She will be doing 18 months in federal lockup.
The people in charge of enforcing US export law are worse than city building inspectors and the penalties are orders of magnitude more severe. They're not people you want to mess with, ignore, or pretend you didn't know the rules.
That feels materially different than a software program released in the open domestic market.
And while it feels different, the government would apparently disagree: https://exportcontrol.lbl.gov/training/export-control-overvi...
Distribution of software to a foreign national within the US is an explicit proviso of the law. That is a wild law and I am surprised it stands.
I’m surprised about some of the reactions, but again anyone who has a security clearance wouldn’t touch any of this discussion with a ten-foot pole also.
Export controls are governed by the Export Control Reform Act of 2018. The procedures the government must follow to enact a rule is defined by the Administrative Procedures Act. It's not a slam-dunk case, but I think there are significant questions around the arbitrary nature of the approved company list, the control restricting US-based permanent resident employees from accessing models, and how the rule was enacted in general which would make a viable case.
Suing the government saying that APA was not correctly followed or that the export control in question does not comply with the ECRA will absolutely not get you put in prison, that's not the kind of country we live in and honestly I have no clue what you're talking about.
The government has arbitrary commandeered their business.
This could ruin Anthropic.
They are walking a tight rope with respect to revenues, hype, IPO.
If this kills their hyper growth prospects, it could kill the IPO.
If there's a serious change the gov. is out of line, the judges could put a stay and possibly throw this out.
There may however be enough of a case, in which there's maybe not much they can do.
Having a crazy person completely control your business is very, very bad.
Lots of these moves would and should be struck down in court as an arbitrary and capricious use of administrative power. Some of them might not be, and in the meantime you're signing up for tons of trouble. A trillion-dollar company does not simply go to war with the US government.
A more mid-sized company that's not so intertwined, but not so small that they can't get a good legal team, might be another story.
Ask every satellite launch company in the 90s how that worked out.
Trump does many things he lacks the legal authority to do.
Ignore export control regulations?
I think you’re trying to say you really feel it’s not fair, and you’d like so and so meany and bully to go and pound sand. And yeah most people feel the same way here but ignoring export control regulations is not a joking matter and not something to play around with. Especially for a company that feels they are having extra eyes on them.
- extract monetary contributions for their side of political spectrum from ai companies
- extract money for personal gain
- grokify ai answers on political / worldview topics, because polls are showing people trust ai answers more than wikipedia
A published policy with the right to appeal exclusions from the list.
An equal standard for all companies rather than ad hoc application.
A countervailing policy to mitigate the unfair advantage conferred on the companies that have early access (such as a higher tax rate that goes to fund ai job loses, and a commitment that AI use of the new models won’t result in layoffs).
A requirement that hardware is made available for open source models rather than locked up in by the AI labs.
A restriction on AI labs being vertically integrated from hardware all the way up through the app layer. I would restrict AI labs to being API providers and prohibit them from building apps. That would allow an ecosystem of independent software development on the app layer without fear of being copied by the labs that have an unfair advantage in seeing the data while apps are being built, the usage data as they become successful and the ability to undercut competitors by subsidizing tokens unfairly.
I could go on.
I've read all the sci fi they have. It's not hard to see where the ideas came from.
What's being questioned is this sentiment of "The only way to save humanity is for me and my lighthaven groupies to become Xillionaire god-kings"
Why? Because I am a social media addict that lives in 2026 and I don't know how to relate to things in the world that don't involve complaining.
Complaining you see is more form of epistemology and entertainment at the same time to me. Reasoned debate and nuance just doesn't get enough likes for me. I am all about the emotional response to a topic.
Your question is like asking what evidence would convince us that a bag of rocks doesn't have rocks in it. Easy, just take the rocks out.
Between the Chinese government and Anthropic, I know which one I'd rather send tokens to. For all of the problems of the US, for-profit corporations, data harvesting, etc. the CCP (and, perhaps more troublesome, its allies) is far less likely to align with your interests.
It's not like China can be trusted either, but China isn't planning any direct invasions to the west. Taiwan, perhaps, but they're playing a long-term tactical game rather than a "invade the country we don't like this week" game. They might get some info on you, but the data brokers in the west will sell a lot more details about you, pre-categorized and all.
If you're afraid of industrial espionage, Chinese companies may be a risk, but in that case you shouldn't be uploading your secrets to an AI company in the first place.
Personally I am much more concerned about handing my data over to the government that actually has power over me and labels dissenters terrorists than I am with the government overseas that has no direct effect on my life... well, other than providing alternative LLMs with permissive licenses that can be hosted anywhere in the world... but to each their own, I suppose.
That's the neat part with the Chinese open weight models. You don't have to send your tokens to the PRC, the models can be hosted stateside or anywhere else you'd like.
Do you have access to Mythos? If not the choice has already been made for you.
You meant to write "Between the Chinese government and the US government". Completely agreed though, better to send it to the former.
You know what? I live in Europe. China was not the country threatening military action against one EU nation that would throw the whole continent into war a few months ago.
Also, it's not Chinese companies harvesting every piece of data about me that they can get their hands on.
If fear is your argument, I know that I fear the US and its big tech corporations a lot more that China.
The financial growth based on that everyone on earth will need to have an OpenAI or Anthropic subscription is quickly falling apart.
DeepSeek V4 is good enough for most of my work that that I can no longer justify paying 100x for something else. The cost difference is astronomical.
Sad situation is that half the talent that did the primary research and initially created these models and got them out there was educated here or taught here or have citizenship here in Canada. Karpathy, Sutskever, Hinton, there's a huge list. And for other countries, too.
In the short term this play may work for the US administration, in the long run it will only reduce the flow of talent and good will and sharing between G7 nations.
One can hope.
Either because ordinary people hate it or (more likely) because Sam and Dario have got too powerful and they’re now starting to become a genuine rival castle to the US government in elite theory terms - of course at that point you get your wings clipped
(Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
- It's not "incrementally better". It's a complete game changer. Opus 4.8 on max thinking does X amount of mistakes in my commercial work. Fable 5 did 5% of X. Counted. I barely had anything to contribute in the work sessions, for a full week I could count on my two hands the total amount of times I actually caught Fable 5 -- and one part of those were not true mistakes, more like divergence from policy in our `CLAUDE.md` files.
- It's not "security focused". It's simply better in every way _plus_ it's also security-conscious.
- It legitimately accelerated my work. I don't have too much unknowns in my work, I simply have way too much to do. Fable 5 was an objective and measurable improvement over Opus 4.8. Returning to it after Fable 5 was removed was extremely discouraging and frustrating, and still is to some extent.
> It’s lots of fun to pretend it’s some step-change that’s too dangerous for general release
Maybe, but not as much fun as tearing down a straw man apparently. :)
> (Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
It's ridiculous for multiple other reasons but ridiculous nonetheless.
They def make less than in europe. You make very good money for working at Volkswagen on the line.
And China now also has software developers and even if they make the same amount, they are def also now a big player in the game and take parts of the cake.
Its also worth noting that st southern and central European labor costs they would still have an advantage over the US.
i.e. freedom for me but not for thee
Everything is going according to plan. When we have majority, we'll move the companies to Norway, but don't tell anyone, this is supposed to be a secret. /s
Fable + jailbreak = Mythos
Was hoping someone else had proposed it though. It's a good thought
I might be un-convinced if someone published a clear technical explanation of what error these companies have all made. The only counterarguments I've seen so far are vague speculation about how bubble dynamics might incentivize them to say it worked great, and to me that's not persuasive.
It sounds insidious, and it probably would be if they weren’t so damn dumb.
But there's definitely a large contingent who denies that they think there's any risk at all, instead of them engaging in motivated reasoning to think their self interest just so happens to coincide with what is best for safety.
Yet.
Things change as well, in all sides. The US might swing back to normal with the next election, or China worse, or vice-versa.
Kushner and the rest of the gang have probably already got a heads up about this and cashed out their stakes, now it's time for Sam and Dario to go through some struggle sessions
On the bright side I got to add "successfully managed the purchase of controlled munitions" to my CV.
Then you've had lots of fun with EAR compliance.
Lol
People claim this, but it doesn't seem to be the case at all. We have multiple ongoing conflicts between countries currently, and we're not seeing huge gains from those with more access to AI.
Previously US used rebel groups to throw governments in other countries that are not US friendly. In the future I can see they can just start chaos by attacking banking systems, energy infrastructure or any other important modern infrastructure until that country falls in line with US
It’s so easy to argue on putting tariff on US tech, but we forget how much Europeans depend on it and it would be like shooting one’s own foot.
One could argue that over the time EU can build their own infrastructure and alternative, but who is going to invest for it? The governments? With tax payers money? And who is going to build it? EU is one of the fastest aging continent in the world and what can they offer to attract young talent?
aws et all are just extracting rent. we all know this.
telcos have some ownership of the android firmware they distribute, and could easily spin up and fund alternatives like replicant or even go full partnership with Chinese versions.
simply moving email providers from Microsoft would make threeletter agencies work much more difficult, and it's a 2min dns change.
any way you look at it, the impact would be positive. i don't think your example is the example you think it is. like trump, you're ignoring agency to every other player in the systems and ignoring some obvious consequences.
Forget about phones, all the companies that uses RDS would lose their data and that would be catastrophic.
The normal USA citicen can't afford a car anymore. Either they make it a lot cheaper over there or they have to continue pressue the USA Citicents to accept that they are not allowed to buy cheap China products.
It will be quitei nteresting to see if we will see a global rebalancing of manufactoring and co around the globe for USA, Europe and China or a overall change in system from pure capitalism to something else.
Or the big players start to reinvest into countries again to have customers who can actually afford it again.
In other words, the problem of "The normal USA citizen can't afford a car anymore" can be understood in a different way. That can be fixed, by paying people less (yes, less, really, think about it)
The US can drop it's imports and do essentially everything itself.
Right now people can't afford a car because the US is so equal, meaning they can't compete with labor in India or ... But you could "fix" that, by creating more inequality in the US (do what we effectively do now, worldwide, but within the US. Forbid people immigrating from California to New York. Move all banks to New York, build factories in California. Pay people in California 20% what they make in New York for the same job), and production will move back to the US.
And if that offends you, please understand that's exactly what we're doing now. Just replace California and New York with India or China and New York with, well New York, or Washington, and that's exactly the system we have. That's the system the EU is trying to create in the EU.
Doesn't work that way for exports (EU, or China, or to a lesser extent, India). The EU is an export block that's rich getting out competed by export blocks that are poor ... Now THAT is a difficult problem to solve. Not that EU politicians are even trying.
The US is not doing better because US politicians are better than EU ones. They just have a much easier problem to solve. And, of course, both US and EU politicians are failing, but the consequences are much bigger in the EU.
For example, the EU is in fact making it harder to immigrate WITHIN the block, despite "free movement", and is creating massive differences in pay for the same work, which is apparently what socialism stands for:
Typical after-tax pay for supermarket cashier:
Netherlands: 2550 euro (after tax), for 36 hours of work ~ 15 euro/hour
Greece: 1050 euro (after tax), for 40 hours of work ~ 6 euro/hour
Oh and that's not where it ends. You get way more social protection and medical in the Netherlands ... for paying less in tax. The EU is purposefully creating inequality to solve the problem of rich Dutch people not getting richer fast enough. It's only within the same country that there is less inequality in the EU. And, of course, the EU has been making it worse, not better, for about 18 years now. They're not about to stop. The US hasn't even decently started doing this yet.
I assume it was a mix of exports and a continuesly growing population keeping this system alive for a long time.
If you know reduce the avg salary down to 1/3, you will get cheaper cars (if you even still need people to make cars, relevant amount of people), but still no one can afford it.
You need to get down from the current price point of 50k down to what? 20k?
Car manufactorers will have to shrink down production, reduce laber numbers, reduce salaries and the avg us american will have to downsize its consumption by a lot.
The USAmericans are at least not taking social security and health care as a human right, after all they don't know it better mh?
This doesn't directly follow from the first part of your comment, and more importantly seems inaccurate with respect to Anthropic's public statements on this situation. For example:
> As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts. This action does not adhere to those principles.
There is a good reason no one seriously tests the resolve of these particular regulatory authorities.
As enjoyable as the sheer irony of the situation is, this is a terrible development... Not only are foreign peasants like me cut off from the best models, so are the chinese AI labs that were distilling them into open weights for the rest of us to enjoy. This sudden acceleration has been genuinely terrifying, I'm not sure what to expect of the future anymore.
Lockheed doesn't whine that they aren't allowed to sell their products for a $20/mo subscription to everyone with a pulse.
true but that isn't what we're talking about here
It's really sad seeing how little so many of us Europeans understand the situation. America and China hold all the cards, Europe holds none. It makes very little that is both unique and strategic. Decades of left wing economic policies are coming home to roost and there's no way to turn the ship around now.
Yawn. If you look at actual production in industry data we're at the peak or close to.
A lot of this yapping is (outside of just boring political droning) due to conflating relative stagnation with "deindustrialization" lol and looking purely at individual countries like Germany, not realizing the most of the recent growth is east and south of it.
Indeed. This is the same argument people make about oil: it's at (or near) it's highest level ever! So things must be good! And that's not exactly true.
Even though, for oil, there's barely any people who see how bad things are (I guess the top of the US government sort of demands you don't if you work for the US government). But what matters is the change.
If things even just stay the same, efficiency improvements will lead to massive job losses, and in practice "stay the same" still means some companies suffer and others gain. The suffering companies are the employment giants, the gaining companies are the "lights out" factories, to an approximation. Things must go up, it's "the derivative" that matters, so to speak.
That's why it's such a massive deal that the derivative for Europe's industry (even with the massive military investment) ... very close to zero. In other words: the job losses will continue even under Germany's massive investment plan.
You could say they're not beating inflation, but that isn't exactly accurate. Or it isn't the problem.
that def will hurt.
But we still have a lot of capital left, so plenty of ways to adress the elephant in the room. Its just very weird that the current CDU doesn't has any real good ideas.
USA has a certain amount of hands especially IT but otherwise? China runs circles around Europe and USA for everything else.
Btw. CDU was in power for 16 years and is more right wing than center. If you say left wing in sense of not protecting our own markets, i think this was accepted globally.
USA also can't make things anymore.
If no-one comes up with the general notion of a min of like 60% home made, none of it matters because rebuilding this will take a lot of time and money.
And even with a chip factory in USA, they still import most and automate as much as possible.
This might reduce the pressure on USA Loans but will not help the USA people themselves. They will just continue seeing a bigger and bigger split between poor and rich.
I read your frustration. Try to let go of the fact that there are many smart people who aren't experts in legal affairs. Cite eCFR if they're wrong, and move on. As much as they don't know the rules, you don't know their situation.
For all you know, the subscriber may be a US Citizen + Delaware C Corp owner.
I remember a time not very long ago when everyday crypto like 128-bit SSL was restricted under US export law. The old web browsers came in separate, "exportable" versions. [1]
Phil Zimmermann was in big trouble for releasing PGP. That was the mid-90s. Clinton was President so this stuff transcends politics.
[1] https://en.wikipedia.org/wiki/File:Netscape_Navigator_1.1_fo...
This is different situation. Cybersecurity specialists (at least those I found and read) don't consider Mythos as something really powerful. Good tool but not groundbreaking.
Anthropic was playing terror game and burned by it
It's not impossible, it's just hard.
Getting just a secret clearance takes up to a year, and expecting to do this for all employees, in this AI market? Not going to happen, things move way too fast.
People complain about the US approach, but almost every high end manufacturing is being export controlled, from ASML machines to high end optic used in missiles.
The US government has stated that the US is very behind in manufacturing of military equipment vs e.g. China (in terms of scale and speed).
So if your example is gimp it then sure it's possible.
> https://calibremag.ca/trump-preparing-to-roll-back-itar-expo...
It hasn't yet happened. And, as things stand right now, nobody in their right mind wants to work in a business beholden to ITAR. Strong export controls will kill American AI.
> The EAR definition of “export” extends beyond the transportation of physical goods outside the U.S.
> A “Deemed export” is the release of technology or source code to a foreign national in the U.S. The release is “deemed” to be an export to the last permanent residence status/citizenship of the foreign national. This can occur through demonstration, oral briefing, site visit, or through transmission of non-public data.
Ref: https://exportcontrol.lbl.gov/training/export-control-overvi...
You don’t want to fuck around with export control laws.
The second part played an enormous role too.
So you have to be decently wealthy now to migrate legally in the EU, going up. Which of course achieves exactly what's "needed": poor Greeks/... cannot get away from low Greek wages by simply driving to the Netherlands. Doesn't work. Wage disparities and inequality can grow and grow and grow.
Not that this is going even remotely fast enough for the EU. There is talk now about a "2-speed" EU. Meaning people from the periphery would get less rights (especially when it comes to how much weight their votes carry, but also when it comes to moving)
Companies, by contrast, can easily move to find the cheaper labor. Because, of course they can.
And, because of course, people who "cheat" by not taking benefits or jobs, just making a bit on the side and living in caravans to avoid everything, big example being the Roma, are being hunted by the authorities in 50 different ways. Problem is that mostly they're not doing much of anything wrong, so methods like forcing them through stealing their kids are used (sorry, of course I mean "protecting the kids" through youth services, excuse me, I misspoke. Protecting kids. Not kidnapping kids to pressure the parents).
Oh and it is TOTALLY unreasonable that these people are teaching their kids to directly use violence against anyone come to "help" from the state. Totally, completely, 100% unreasonable. What are you going to do though? When youth services was allowed to use violence against the kids, they massively abused that. For money. For sex. Shooting (yes, really, with a gun) kids slowly leaving the premises. Leaving "troubled" kids in an isolation cell until they commit suicide. Not called that, of course. You'd be surprised how many euphemisms youth services have for both isolation cells for children and for suicides happening on (or immediately after) their watch. Things like this[1] for example have never been named isolation cells, which is what they are, but, loosely translated "time-out room", "separation from everything", "low stimulation room", "TAVA room", "separation from incitement room", "cool down room", "sensory-friendly room", "seclusion room", "safe space" (NEVER trust or go with anyone from youth services talking about safe spaces, unless you're at least 25, and even then) and I'm sure I haven't covered even half of them)
It's not impossible, of course. It's not even terribly difficult, but it does require a different level of record.
(No, I'm not saying that the goons running the United States give a shit or won't do it anyway.)
Also most crypto companies are not good for laundering since the blockchains record that fraud forever and publicly. I could see some specific protocols where that may not be true — like monero or tornado cash — but these projects are not really startups. Most crypto startups pitch their products for enterprise customers and thus would be horrible for laundering money.
Maybe it is better at vibe coding or finding security flaws, but at how much is it sufficiently better to be worth paying the extra?
And either way no AI basically puts you back to where we were last year. US employees have always been far more productive, that’s nothing new.
This isn't an AI problem, it's a management issue. Why are you getting devs to build things that aren't going to generate revenue?
It’s not just us: where’s the revenue in the entire market? We can see all the public filings. There haven’t been any revenue gains. The only people making money from AI are the LLM providers. And even they are losing money. Even the biggest tech companies are limiting token spend. At best the tech is a new cost just to maintain parity, and I think most businesses look at it as a way to cut dev costs (trading for token spend). I think they will learn that it’s less of a win than they hoped. If a dev was spending 50% of their time coding, and you reduce that to 10% - that’s a big change but it isn’t really making you more because it’s all that time we hate in meetings, understanding customer needs, etc, that make us money.
So from a wording point of view: generic term still would be CDU is right.
Oh, European billionnaires and incumbent telecoms would love to. As soon as they get enough government concessions and guarantees. Who exactly? The ALDI family. Xavier Niel.
At least for people working in IT this is the fastest way to answer the question "how can we make it even worse?". Especially European incumbent telecoms. They're the answer to this question "do people sometime quit their job, in Europe, through suicide?". Yes, yes they do.
I'm pretty sure they are really looking forward to pocket billions of tax payers money. If they are so eager, why wait for the government? just go for it like the US counterparts.
Say you had a perfectly smooth progressive chain from rocks to spears to guns to nuclear weapons. When it comes to government restrictions, you still have to choose to draw lines somewhere, right?
Given the impossibility of compliance, what Anthropic and OpenAI are doing is working with the government to release it to certain organizations with the government's blessing.
If this were about missiles and not AI models, nobody would question this turn of events. If the government said that nobody can export this missile or allow non-citizens access to the missile, and then they started giving permission for certain organizations to handle the missile, that would be normal, not picking winners.
The only reason people are questioning it in this case is because they believe that these models are not dangerous enough to deserve these kinds of export controls. Personally I'd agree that in my 3 days of using Fable I didn't observe any superpowers. Unfortunately however, Anthropic undermined that argument by claiming that Mythos is highly dangerous, which set them up for any jailbreak of Fable to be considered a national security risk. Who is a court going to believe? Someone who used a model for 3 days? Or the government and the people who made that model?
It is both. The US and China are locked in an AI arms race with economics and security intertwined, given the perceived power of the trajectory of frontier AI models.
Some classic examples of this, on which the US places severe export controls, was advanced materials science and inertial navigation technology. Neither of these are weapons but advanced technology in these domains greatly enables the development of advanced weapons. Any work in these areas is automatically subject to the full export control regime. In extreme cases they may be nationalized and classified.
AI tech is becoming just another tech domain subject to the same level of scrutiny. I’m not making a moral judgement. This was always going to be the reality and a lot of people could see it coming.
I think it's most likely that they felt that this would drive wide antagonism towards the government which would help put them in a more favorable situation for future negotiations to establish a more 'commercially favorable regulation regime.' In other words, build me a moat, now! The government responded by super-screwing them, but is doing so in a way that can help keep the corporate class relatively content while this plays out.
I don't see any way how releasing an incrementally better model could get even close to the positive PR. For every Mythos token not sold because of this they'll sell dozens of Fable, Opus and so on. Perhaps the ban did not origin at Anthropic, as a PR stunt, but genuinely at some government desk. Entirely possible. But Anthropic would be foolish to not pick up that ball and drag it out a little.
That's why I raised skepticism that Anthropic has not been subject to that.
America will do that before gun control.
Not the future I want but I see that too.
The nra has more lobby power than anyone.
As a small business owner this is unacceptable
Current AI was science fiction 5 years ago. But, you're entitled to your opinion, no matter how ignorant it is
[1] https://mcraenglish.weebly.com/uploads/1/5/8/7/15876062/the_...
Identifying the next few things to improve the product is hard work, with market investigation, competitive analysis, synthesizing user feedback... some of this gets faster with LLMs, certainly.
I think most tech companies haven't figured out how to rebalance product to engineering to ensure engineering teams are doing work that grows the business funnel rather than expending tokens on other work.
It's why ASML can't just sell the machines to whomever they want. It's largely American tech inside them.
If this technology has export restrictions, then Anthropic and OpenAI will go out of business.
Anthropic's revenues are already 66% international.
No, they will not.
Where do all these people come from?
Bond yields are already rising on sovereign debt, if the government resorts to 'printing money', rates will rise materially, debt payments become unsustainable and the economy would crash in a worse way than 'the Great Recession' and it's hard to understand what woudl become of that.
In 1929 the US was not 'global reserve' meaning there was not all that extra leverage in the system.
The US 'hooked' on cheap debt, aka the rest of the world buying USD.
Kind of like leveraged speculation - it has upsides, but the downside is harder.
But that's even before the fact that 'nationalization of Anthropic' would lead to it's instant evaporation.
Have a look at what's going on with the DOJ right now, it's being completely corrupted, 'banana republic' style.
People would be reluctant to have such a thing in 'gov' in normal times, but right now ... things are extremely bad.
I really hope the controls are relaxed, but at some arbitrary levels it’s expected that all countries will apply export controls.
Any number of materials and technology can be used for nefarious purposes.
'Security' is real concern, I'm not denying that, it's not unreasonable for gov. to be paying attention or to take some steps - but it can obviously bee too heavy-handed as well.
The Murican propaganda machine keeps turnin.
Interesting, I'm curious what work you do? My software engineering career has never been in that situation, it's always so much ambiguity and unknown that trumps everything.
I work in a financial startup. The codebase is a mess and very much spaghettified. One rework that forced us to migrate our data model from 1:1 users<->loans to M:N (many-to-many) took two months and touched ~40% of the codebase... multiple times. Huge churn. And it just crossed two months of work, even though it's now in its very final phases.
I know what must I do:
- Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems (sigh; not going to rant here about that but it does tick me off how we are _all_ constantly reinventing the same wheels almost every day).
- Saga discipline: if step 6/9 in a pipeline fails, revert everything up to this point, even if it was touched by a 3rd party API.
- Compensation/undo steps. Including flagging / logging those that cannot be undone (sadly one part of our 3rd party APIs are like that).
- Introduce an universal runtime validator library that enforces contracts -- including conditional validation i.e. "only validate field Z if field X is present and is a positive integer and if field Y is present and is a valid UUID".
- Introduce runtime contracts / invariant enforcement.
- Introduce our own dynamic workflow engine, piggybacking off of a few free and unencumbered solutions in the language of choice's ecosystem.
...And these are just off the top of my head after I slept only 4.5h and woke up due to the heat. And each one of these can take from 2 to 6 weeks _even_ with Opus driving all coding and me reviewing and keeping it behaving within my policies and coding standards.
Me & Claude are maintaining a TODO list that is no smaller than 150 items at this point (though in fairness, at least 75% of them are fairly small and not architectural like the ones above).
I believe I know how to architect this thing but business customers and the CEO keep coming back with feature requests which of course always take priority.
When Fable 5 was around, for mere 4 workdays, I not only went ahead of my own schedule feature-work-wise but even had the bandwidth to start tackling a few other architectural decisions, tightened them up in `CLAUDE.md` and Fable even devised an opinionated AST linter for test discipline (disallow direct DB access in our tests, only go through the domain/context modules to do so). It helped me start turning the tide.
This all went out the window when I had to go back to Opus 4.8. It's still _very_ good, mind you, but it does feel like I am a special-education teacher periodically. It forgets disciplines we discussed and codified likely 15-20 times at this point, forgets important project context and attempts to reintroduce subtle bugs, and a few others.
My next game is, with or without Fable, to continue its work and just enrich the AST-based linters to convert the theoretical prompt-based guard-rails into actual LLM hooks and compiler / runtime-at-startup hooks so the agent cannot ignore them.
I don't enjoy harness engineering but the interesting and very positive effect has been that it helped me think more like an architect and less like a coding monkey, which I do hugely appreciate and only realized I was missing it for years after it actually started happening again.
Hope that helps put things in context.
> Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems
Amen to that.
Mistral experience was as miserable as running local models btw.
Just in case you missed that class, taxes are determined by legislators
Trump has ruined that, so now we'll be moving to build our own alternatives
We are taking our time about it and not being consistent. For example some European governments are moving away from US cloud systems, but they are also pushing apps that require American devices to work. There is very little movement of the private sector away from the US so much of the economy economy remains dependent on the US. The end result will be government clouds not dependent on the US, but that can only be used using US controlled devices, and a private sector still just as dependent.
It's just chaotic. Chaos isn't good for business. What if Microsoft were instructed, on short notice, to block EU customers from using Entra over a trade dispute? If the identity provider of millions of European customers is suddenly unavailable, work grinds to halt. That's unacceptable.
Now you might say that sounds unrealistic - but would you have expected the American president to threaten an invasion of Danish territory? Or freaking Canada?
And even if the next administration turns out to be more "normal", we won't forget that American voters don't care about anyone but themselves - they wanted this. Twice. There is virtually no guarantee the same thing won't happen again in the future, and we all have seen how quickly fruitful business relations turned to shit over a pathetic narcissistic a-hole at the helm throwing temper tantrums.
Trusting the USA has proven to be a bad strategy, even for its closest allies. That won't change all too soon.
https://apnews.com/article/icc-trump-sanctions-karim-khan-co...
Trump is using sanctions and started a trade war with Europe.
Europe would be mad to entrust their digital ecosystem to such a country
20% VAT to begin with payed with heavily taxed money of mine.
Than on top of payed by provider of services to UK gov machine making 10usd sub be 10£ sub.
I'm not saying Anthropic is a defense contractor, I'm just saying once something is export controlled, the whole dynamic changes and there's no difference between them and a defense contractor.
Those who tried are only doing it in mostly academic/toy languages which is a damned shame. We _really_ need those constructs, compile-time enforced, in commercial codebases, like 25 years ago at least!
People are trying everywhere though but I've witnessed CTOs getting cold feet and fearing their codebase will get too abstract or hard to maintain when they hire the next dev.
Group-think and conformism and fear of change, demonstrated live, every day. :/
So yep -- amen to that indeed.
Fable 5 was amazing at helping me with that. Opus 4.8 is... good, really good, but... you have to be on your toes _all_ the time. It gets tiring. But as already explained: it also helped me become better.
Regardless of US controls, I can from Australia buy a high quality 50 fps thermal scope with built in 1000m (or ~ 1000 yard) laser range finder from Hangzhou, China.
ie. Why lock the stable at 9 fps when out on the open plain 50 fps is already running free.
But being part of a gov industrial entity would be hard, unless it had 'all the right ingredients'.
Like the modern version of NASA? It's hard to imagine under this administration. But maybe it's plausible with a different mindset.
Also worth remembering NASA exists to counter Sputnick, we think of it as 'space, exploration, science' but it has a 'cold war' underpinning.
'Being nationalized' is 'Going out of business' - in this context.
The US will not pay market prices that investors would want - and - everyone working there would leave.
Even if they did not - the fuel required to keep it going would not be provided.
US Gov is not geared to lead an institution like that.
It would evaporate
Nick 30 ans is tired and wants someone else to pick up the slack for once.
The current models are open weights and already out the door. They are hosted by many providers and are already comparatively good in many domains. Even if this generation is the last one to be open, I‘d argue it would already put the US providers in trouble
I'm upbreat about China because they seem to be the biggest player here, but even if they don't come through I expect other countries will be able to put out decent free models.
What's perplexing is people see CCP as some kind of libertarians while they in fact love regulations and routinely leverage them both domestically and in exports.
Imagine for a minute that Trump were a lot smarter, and also ran the supreme court and congress, and you'll basically have Xi.
Why as a European be more afraid of a country that whilst you wouldn’t want to live in and which has supported Russia is quiet and selling you cheap stuff vs a country who used to be a friend who is now wanting you to grovel for a bad deal and is also supporting Russia?
Does anyone have any insight into how we could potentially counter China’s propaganda? This comment right here showed the danger of ignoring it.
Rebuilding trust and foreign relationships is going to take a long time after Trump is gone.
Chinese 50 fps thermal cameras and scopes have been about for a while now, they've been supplying the industrial specialist market even longer than the general consumer goods market.
I note the US export control is limited by licence .. which likely means the US high frame rate thermal gear is equally as available in AU as (say) TI DSP chips were when they first iterated that family (ie. restricted by licence but easy to purchase in allied countries if you just sign a form stating you're not supplying them to countries in the Chinese sphere of influence, etc.)
And I can't trust my gov to grow some balls and say we will no longer negotiate with a senile old child raping man baby.
I can't even trust any company having their workers include or upload my data to a free AI model.
The US is a big and beautiful country. I also dislike contemporary US politics, but don't deny yourself the experience of ever traveling there out of spite.
My friends experiences visiting family (before Trump) have confirmed them multiple times. Imagine visiting family as a 17 year old German and then being questioned for 11 hours because you don't know the exact street where the family lives. While the relative is waiting at the airport without any kind of information.
I never felt welcome. The only country I felt the same was Russia, and to some much smaller extent - China.
Read about GE and Alstom and how the US government (under Obama) forced the sell at a discount, without a true GE financial audit.
No, experience tell if you're a foreign company owner, you risk less allying with the CCP than with the US. At worst with the CCP you'll lose your IP, with the US you will get arrested and be forced to 'sell' (I.E. you'll get overpriced stocks)
So one of the world's biggest and most rapid military build-ups in history that is largely intended to give China the ability to seize a democratic country by force by 2027 over any US/Western efforts to protect it is OK because...it's "a long-term tactical game"?
Note that China is not just menacing Taiwan. It's constantly harassing Japan, the Philippines and Vietnam too. Other countries in the region are worried because they understand that if China takes Taiwan successfully, it's not likely to stop there and become a good, peaceful neighbor.
The US, under Trump, is a foreign policy disaster. That doesn't mean that China, with a seemingly more emotionally stable dictator at the helm, is any less dangerous.
> They might get some info on you, but the data brokers in the west will sell a lot more details about you, pre-categorized and all.
With all due respect, you're really naive about how China operates.
https://www.wired.com/story/chineses-surveillance-state-is-s...
Like, can you not find the US bombing of Iran and, say, China's genocide against the Uyghurs to be a terrible thing at the same time?
Is there something preventing you from being concerned about Trump's threats to take Canada or Greenland and being concerned about China's planning and threats to seize Taiwan by force at the same time?
Wake me up when they start drone striking neighboring country's fishermen and accusing them of carrying drugs.
https://cset.georgetown.edu/article/a-different-kind-of-army...
At this point I am afraid of US government than China...
Huh? Russia invaded Ukraine, at significant economic cost, and hundreds of thousands of people have died.
> At this point I am afraid of US government than China...
Just as people can walk and chew gum at the same time, you can have reason to be concerned about more than one country at a time.
The US, under every single president has been an imperialistic threat to half the world. From imposing embargoes, to overthrowing governments and supporting dictators and genocides all over the world. Half the world hates you, and the other half has begrudgingly no choice but to half assedly support the greatest threat that the world has ever known, a military in a trenchcoat made to protect their dominance over world trade.
So, respectfully, fuck off. The Chinese are not a worse problem than you are, merely a different one. The NSA already has all my data, no reason the MSS shouldn't get a piece of that data too.
This administration only removed the blinds on what has always been an adversarial policy, allies included.
If given a choice you may choose US or Chinese models for whatever reason it's fine, but there's no need to fall into the delusion that it is for moral reasons or obligations.
Thanks to these big guys the odds are stacking against any fresh competition. Data sources have dried up and training material is harder to get, regulation is controlling any advanced model, prices are inaccessible now, and they’re seeking that courts cut off the rest of the avenues, including the ones they used to get where they are.
If Anthropic wanted to continue slow rolling their best model to a select few companies they could've kept expanding mythos access instead
Not to mention the fact that OpenAI has had time to prep 5.6 for release so that there will actually be an alternative around the same intelligence point as Fable. This is just all around terrible for Anthropic.
If you want a conspiracy theory, the one that says "this is corporate assassination" or "the US government does want to be in the business of picking winners" is much much much closer to the truth than yours.
Rather than buying either conspiracy theory, though, I expect Hanlon's razor describes the situation best (incompetence on one side, inability to coordinate/communicate effectively with the current administration on the other)
[edit: don’t forget my previous tin foil hat note, I’m not overly serious about this]
By restricting use, the outcome may be slower progress, thus narrowing the gap for other reasons.
This particular specific doom is from the USG, Trump has a history of kayfabe, and there's a stink of market manipulation coming from the White House.
* Musk, however, you can totally have: "With artificial intelligence, we are summoning the demon" in 2014 to "if I go ahead and build this enormous robot army, can I just be ousted at some point in the future?" in 2025.
Assuming they are evil unsane and actually believe that crap while doing what they do looks way more like conspiracy theory.
Every time I've met someone who believes a conspiracy theory, they don't realise how not-sane the conspiracy they propose is.*
Founding a company on the basis you don't think the others are safe enough, raising capital on that basis, developing methods to improve AI safety, publishing literature about your methods, making open calls for legislation for safety standards, etc.?
While also managing to not leak documentation of this despite all the staff who did leave in order to openly speak about the stuff that they thought still wasn't safe enough?
Thinking that all the doom-talk from OpenAI and Anthropic** is just a PR technique even though they maintained this position continuously starting before they had any money or offices is about as sensible as thinking 9/11 was an insurance scam.
* by definition, because if they did they wouldn't believe it; actual conspiracies can of course be insane, but a conspiracy theory has to also justify why the "evidence" of conspiracy consists of people saying "I recon" rather than documents (specifically documents that mean what they think they mean, cf. "Mike's Nature trick").
** Again, just those two. I'm absolutely not fully generalising this, I'm absolutely not saying there's zero people who do as you say. Heck, the mere fact that this is a common talking point practically guarantees someone post-ChatGPT saw all the people claiming it was PR and said to themselves "great idea I'll do that".
On your last point abour “causing instability on the internet”, for example, I could point out we are in a discussion about America wanting to restrict LLMs into haves and have nots and if that isn’t destabilising the internet by preventing non-American companies from defending themselves or paying American companies to defend themselves against them protection racket wise then what is?
All Americans are given a social credit score. Trump can decide any transgender sympathizers are unruly citizens, unworthy of a good score. They are banned from taking out loans. They are banned from buying plane or train tickets. If you call them, a warning informs you they are actively bad people. If you are spotted speaking to them, your friends are encouraged to report your behavior in an app, lowering your score as well. Going down is easy, but going up is nearly impossible. You’ve outed yourself as a problem.
In the meantime, Trump has also decided to round up and forcibly sterilize every Jew. They’re pretty destabilizing to the nation as a whole and don’t really fit in, so it’s important for the stability of the nation. They are disappeared, sent to prison colonies, and given mysterious injections. International news organizations sneak in to try and get ideas of what you’re doing to them.
Online, you never speak of this of course. Not only are your comments tracked (affecting that citizen social score I mentioned), the propaganda you need to hear is tuned to you harder than ever before. It is designed to have a perfect answer to every complaint you have. If you fight too hard, a quick reeducational visit can do wonders.
There are like ten million more things I could complain about, but let’s just keep it to the most obvious, well-known stuff for now.
All of this stuff is actively happening in China right now. So, I apologize, but I cannot accept that you are “a careful follower of politics”. I don’t hold it against you, but it’s insanely concerning.
FWIW I do not buy into the "China bad, we good" narrative either.
The US has done really fucked up things, bombed countries for freedom, there has been recent events as well. I do not even think they were ever the "good guys" they thought they are. In Hollywood movies, sure, but in reality? Nah.
What lists? And what happens to people on those lists?
> Maybe that is the reason?
If you think people are signing up throwaway accounts to post replies to random comments, making no substantive or even controversial statements, I have a bridge to sell you. In the US or China. Take your pick.
> FWIW I do not buy into the "China bad, we good" narrative either.
I never presented that narrative.
> The US has done really fucked up things, bombed countries for freedom, there has been recent events as well. I do not even think they were ever the "good guys" they thought they are. In Hollywood movies, sure, but in reality? Nah.
I'd agree. But what does this have to do with an analysis of China?
Distillation is neither an 'attack', nor 'adversarial' in any reasonable sense.
> Does that mean I am getting paid by the CIA or something
It at the very least means you're uncritically parroting the framing of a company that'd like nothing more than to successfully persuade lawmakers that something should be done about open Chinese models (eliminate choice), preferably so that Anthropic is close to anyone's only option.
I think this is mostly a confusing way to describe it, but I'm not really sure why you say it isn't an attack or adversarial. One side is doing something the other side doesn't want. Seems to be pretty clearly adversarial.
This is Anthropic we're talking about here. A company that's infamous for adversarial scraping of copyrighted content. I generally don't accept their framing, especially when it's pretty clear what the end goal of that is.
There are no Chinese military personnel and bases in Europe. The Chinese have not tried to take European land by force and China is also quite stable. It won't change it's mind every 4 years and negate international treaties that have been signed. I also don't recall China invading another country and/or kidnapping leaders of a foreign state in recent history. China also does not control the global reserve currency and can not unilaterally impose debilitating sanction on countries such as Cuba causing the death of thousands of innocent people because of some historical beef.
Yes, China does not have a good human rights record, they imprison a lot of political dissidence but then who are we comparing them too? The US has a 5x worse per capita prison population than China and a horrific human rights record.
What exactly bothers you so much about China that the US does not?
The irony is that the US is currently getting so much flak because it's leader is trying (and failing, EOs only go so far) to become what Xi Jinpeng is to China.
So if you want to put all your eggs in Xi's basket alone, go ahead. But if he has a stroke tomorrow and thinks China would be best off making pool toys, all your eggs will be smashed in one instant.
What did US do? Are you referring to Kosovo?
For context, I am European using both US and Chinese models, not rooting for one side or the other (but prefer openness of Chinese models).
Once China can twist your arm you can be sure they will. Just mention democracy to them and you will see what comes next.
Also FYI: the CCP has officially adopted the designation "socialist democracy" for themselves*, so I don't think you're going to bother them much by using that term. You'll have to get more specific about what you think their "democracy" should look like for them to start giving you the side-eye.
* Many places that are not really recognizable as democracies from a western POV do this. People, we have democracy at home!
> Founding a company on the basis you don't think the others are safe enough, raising capital on that basis, developing methods to improve AI safety, publishing literature about your methods, making open calls for legislation for safety standards, etc.?
These are marketing claims. Or self-delusion claims.
> Thinking that all the doom-talk from OpenAI and Anthropic is just a PR technique even though they maintained this position continuously starting before they had any money or offices i
Continuing to maintain self-aggrandizing position that makes investors give you money is not proof that you are genuine.
But, if they are genuine, then maybe they should stop trying to make that doom happen as fast as possible. I just dont see how "they are really trying to cause maximal harm to maximum amount of people" is a defense.
Original founding says otherwise, for both OpenAI (famously) and Anthropic with this in a copy of their certificate of incorporation I found:
ARTICLE III
The specific public benefit that the Corporation will promote is to responsibly develop and maintain advanced Al for the cultural, social and technological improvement of humanity. In addition, the Corporation may engage in any lawful act or activity for which corporations may be organized under the Delaware General Corporation Law as the same exists or may hereafter be amended (the "DGCL.").
Immediate thing they did? Spend one year training a model, then a second year safety-testing it.
> These are marketing claims. Or self-delusion claims.
It's published reproducible work. Early-stage stuff you could get away with calling "self-delusion", that doesn't fly when papers get published, sometimes on topics that can be described as "AI do X, which is bad, can we make them not?" e.g https://arxiv.org/abs/2310.13548, https://arxiv.org/abs/2406.10162, https://arxiv.org/abs/2503.10965, https://arxiv.org/abs/2511.18397 from Anthropic and https://arxiv.org/abs/2412.16339 from OpenAI.
> Continuing to maintain self-aggrandizing position that makes investors give you money is not proof that you are genuine.
Conspiracies and growing businesses are incompatible with not getting caught.
A conspiracy of 2 is easy to hide. How big did Facebook get before we saw their dirty laundry?
> I just dont see how "they are really trying to cause maximal harm to maximum amount of people" is a defense.
With that in quotations, you seem to be arguing backwards from what you think they're doing.
Their actual arguments are much the same as open source software: with enough eyeballs, all bugs are shallow.
This is obviously true. It may not be sufficient, there's plenty of other doomers who say it isn't sufficient, but it is true.
Easy way to demonstrate, is the difference with the armchair philosophising before LLMs vs. what we see now they exist:
We had decades of people claiming the "obvious" way to keep superhuman intelligence safe is to keep it in a box, keep it offline. Every one of those discussions I've been in, the person saying this refused to believe that it was possible for an AI to convince humans to let it out of the box.
Real life? Blake Lemoine violates his NDA and hires a lawyer on behalf of an LLM, after the LLM convinced him with arguments that a lot of other people (justifiably) mocked him for believing. LaMDA was not superhuman, and it still convinced him.
Did this immediately stop people saying we can keep AI in a box and just not let it out? Nope, kept hearing that for another year or two. That's how much experience humanity collectively needed to internalise just that one part of "AI is more dangerous than you think it is".
Growing business is super compatible with lying to make yourself sound better to investors. How is this one an argument? It is not even a conspiracy, just a normal CEO behavior at this point.
Also, what safety testing seriously? It is not like Chinese models were more unsafe in any realistic way.
And it is not like there wpuld be any real attempts to make humanity better.
I dont kmow what you are on about offline AI and someone not keeping corporate secrets. Or antropomorphised ai with conscoiusness hiring layer by proxy.
Who cares?
Is this how Americans are coping with their civilizatory down fall and vanishing hegemony?
China has only state media and little internet overlap with the west.
From the POV of your typical western internet denizen, The US elected a dictator to burn the country to the ground, while China is building a utopia where the fruits are free for everyone.
I certainly don't see our european media to be pro-China. The Uyghur issues are major criticism. Also the way they treated their citizens during the pandemic and the protests.
The US has for like 90% of its history been at war in one form or another, typically killing people abroad.
I think the US did this more in a century than China did in a millennium.
You don't have to because that wasn't an argument I made. Both the US and China have done horrible things to innocent people and are both currently engaged in malicious behavior that is an affront to humanity.
> I think the US did this more in a century than China did in a millennium.
Once again, I'm not interested in "who is worse" pissing contests, but please educate yourself.
https://en.wikipedia.org/wiki/Great_Leap_Forward
> The Great Leap Forward led to between 15 and 55 million deaths in mainland China during the 1959–1961 Great Chinese Famine it caused, making it the largest or second-largest famine in human history.
https://www.humanrightsresearch.org/post/persecution-of-uygh...
> https://victimsofcommunism.org/tibet-is-lost-but-not-forever...
The Tibetan government in exile believes as many as 1.2 million Tibetans died as a result of Mao's invasion of Tibet in 1949.
https://en.wikipedia.org/wiki/Pol_Pot
China was the Khmer Rouge's main backer. The Khmer Rouge's actions were responsible for a genocide in which 1.5-2 million people perished. Afterwards, China offered asylum to Pol Pot and his top aides.
People who are genuinely interested in justice, human rights and peace should be horrified by what both the US and China have done.
Way to speedrun your argument being shit. I am now free to not take it seriously.
I'll reply for the other onlookers anyway.
> The Great Leap Forward
I know quite well about how awful China was to its own citizens during Mao years, thank you very much.
The thing is, I don't care. China can be as awful as it wants to be to it's own citizens.
The US has a very long list of spying and interfering (many times very violently) abroad. That is what I am concerned about, and that was the point I was replying to.
I could say something along the lines of you needing to educate yourself, but I tend to not seed asphalt.
> The Tibetan government in exile believes as many as 1.2 million Tibetans died as a result of Mao's invasion of Tibet in 1949.
Again, I don't care. China sees places like Tibet and Taiwan as China. Not my problem.
China never threated to send troops to invade Greenland.
> People who are genuinely interested in justice, human rights and peace should be horrified by what both the US and China have done.
In many ways, I am. But that is not what is being discussed here.
China also happens to be the one that's electrifying the world, producing and improving batteries, solar panels, has a long term plan while the US is going hurr durr VC money printing for smart dog food.
You'd probably be a much more content person if you stopped going around the internet assuming things about random strangers on the basis of their nationalities.
If you're going to assume other people have ulterior agendas without evidence, you can't be surprised when people make assumptions about you in turn.
Can a country trust a trade deal with the US? Whereas when has China ever not honoured a trade deal?
China also creates incredibly bad deals meant to use debt to capture nations. The US traditionally is not this aggressive.
But let’s answer your question directly, Mr. “Politically aware”
1. The US-China trade deal, where China agreed to purchase $200 Billion in goods. (Inb4 we explain away how breaking this one is justified and therefore doesn’t count) 2. China randomly placed a shitload of tariffs on Australia because they’re friends with the US. They admitted this was a really foolish choice and walked it back in the following years. 3. China blocked Lithuanian imports because they had a Taiwanese office 4. China gets their ass kicked in the WTO regularly (as do many countries) for reneging on trade deals.
So, Mr. Very Politically Aware, what’s the next goalpost?
https://en.wikipedia.org/wiki/Cross-strait_relations#Deterio...
It doesn't change the fact, that as an European I really wish we had a better relationship with China and ditched US and their backward policies. It's OK to point out atrocities made by big empires, they are all guilty anyway.
The most charitable read i can get is:
> Theft presupposes a legitimate possessory claim by the victim. If A’s possession of X is itself wrongful because A stole X from B, then when C takes X from A, C has not violated A’s rightful ownership of X—because A has none.
I think the whole thing is a bit fraught. In the best case, all frontier model companies would have invested in a giant expansion of Wikipedia and thus distillation would be stealing because the base information is already public and available. Obviously that's not what happened.
However, at this point, I suspect the stolen books (and scraped websites) are largely a footnote of training. Something that was essential to create early models, but relatively minor given the work expended since to create new content and RL environments
"Theft presupposes a legitimate possessory claim by the victim. If A’s possession of X is itself wrongful because A stole X from B, then when C takes X from A, C has not violated A’s rightful ownership of X—because A has none."
Anthropic has no standing in claiming that using its model output is 'stealing' because that would basically eviscerate their business model where they claim that enterprises can use its model output and still claim the generated code as their IP.
My understanding is that distillation uses specific probing queries to extract model weights in some meaningful sense.
There's also the element of "how one uses model outputs can be in accordance with or against the TOS"
(This is true of any website! How I access and use the info on the website can be in accordance with or against the TOS)
Both of those things: doing something against TOS, using specific queries to illegitimately gain access to model weights - that's why calling it adversarial and stealing feel fair.
I didn't (and don't) understand your point, but in an earlier post you made some claim about it being Anthropic - you can't steal from Anthropic. The above is the only reason I can think one might believe that
I do believe that Anthropic complaining about their TOS being violated is rich and should be thrown into the dustbin considering that Anthropic violates other people's TOS when it suits their business.
I do also think them complaining about it is aimed purely at influencing lawmakers to ban open Chinese models and eliminate choice so that them and OpenAI are the only game in town.
As far as other models distilling using Claude, that's just consuming the output of a service in a particular way. Anthropic should be very careful about imposing restrictions of how you can use their model output if they want to maintain marketshare. Enterprises are not going to like that.
That leaves the TOS violation of maybe hitting the service too fast, using too many accounts etc. Except again Anthropic did exactly the same thing when scraping for training data.
Do you believe The Pirate Bay complaining about a competing site stealing 'their' torrents should be taken seriously?
That's how I look at this.
For these reasons, I don't think Anthropic has a legitimate argument here.
Also distilling simply means using the answers from one model to specific questions to steer another model to answer a category of questions in a similar manner using its own capabilities. So as far as extracting weights, it's not exactly that, it's maybe slightly biasing the model that's being distilled so that its weights might look somewhat closer to the model used for distillation, but it's far from extracting them exactly/directly.
Way to speedrun proof that you're an abhorrent person.
> Again, I don't care. China sees places like Tibet and Taiwan as China. Not my problem.
> China never threated to send troops to invade Greenland.
So China causing more than a million deaths and displacements when it actually invaded a functionally sovereign state (Tibet) and threatening to invade and seize Taiwan, which has its own democratically-elected government, constitution, currency, central bank, military, etc. "is not your problem" but somehow Trump threatening to invade Greenland is your problem?
Unless you're a Greenlander, by your own standard, you have no legitimate reason to care about the ~57,000 people who live in Greenland. Let China take Taiwan, and let the US have Greenland.
You're a real piece of...work.
Correct.
> but somehow Trump threatening to invade Greenland is your problem?
Correct.
> Unless you're a Greenlander, by your own standard, you have no legitimate reason to care about the ~57,000 people who live in Greenland
EU, article 42. Educate yourself.
(Just kidding, it would be in vain).
> You're a real piece of...work.
No, I just live in a country that does not pretend to be the world police.
I can't do shit about China invading Tibet or Taiwan, much in the same way I can't do shit about Israel turning Gaza into an open-air Auschwitz, beyond writing mean words on the internet.
But I can recognize what country poses the biggest threat to my living nowadays. That country is not China, it is the US.
No, you're just an abhorrent person.
The fact that you don't care about the people of Tibet, Taiwan, or any other country or group of people that you don't deem important because they don't share your interests, but feign fear that your life could be shattered by Trump invading Greenland, speaks volumes about what type of person you are.
The largest contributor to the EU economy (Germany) was responsible for the worst genocide in human history. The second and fourth largest (France and Spain) were two of the most brutal colonial powers in history, and their wealth today wouldn't exist without the resources they violently extracted from their colonies. Portugal was the largest participant in the Transatlantic Slave Trade. The wealth of the Netherlands today stems from the Dutch trading empires of the 17th century, which were responsible for shameful atrocities around the world.
https://blogs.lse.ac.uk/europpblog/2023/10/31/the-colonial-o...
The EU narrative is one of a peace project but in reality, it's pretty obvious that it was an attempt to maintain Europe's geopolitical relevance after the post-WW2 rise of the US and USSR. When the EEC was founded, it even incorporated France and Belgium's African colonies and fully absorbed French Algeria, where, by the way, somewhere between 400,000 and 700,000 people eventually died during the fighting for independence from France (https://en.wikipedia.org/wiki/Algerian_War).
Humans anywhere and everywhere (the US, Europe, China, etc.) can be horrible, and your loathsome comments prove that in microcosm.
They are far from anything remotely communist except by the one and only thing that defines communist? "Communist" has only ever referred to a political party.
Are you confusing "communist" with "communism"? The latter has held dual meaning, both referring to a nation under rule by a Communist party and an imagined sci-fi world where post-scarcity has taken hold. Obviously we remain far from achieving post-scarcity. However, nobody has ever claimed China has. In context, we know that the political party is the point of focus.
No, but perhaps you are now confusing communism with Marxism? Long story short, Marx and Engels hypothesized that capitalists would effectively withhold the technology necessary to realize post-scarcity. They suggested that the only way to bring post-scarcity into reality was to have the public take control of the means of production — a.k.a. socialism.
You are right to point out that being the foundation of the Communist party. Their efforts (on paper) revolve around the belief that they need to take action to see us find post-scarcity, believing, like Marx, that vested interests will stonewall the transition.
> China has adopted a capitalistic economy
You might say that is a move away from the Marxist underpinnings, but is not at odds with the Communist agenda. We nearly have post-scarcity in the area of food, and that has been primarily driven by American and European advancements, so today we have come to accept that Marx's ideas have not proven themselves. That does not diminish the quest to find post-scarcity, however. Communists (on paper) still have the same goal. If you look closer, while their system is capitalistic, the Communists still hold the reins to prevent autonomous capitalists from stonewalling post-scarcity.
You know, like as is seen in the original article. We have a purported transformative technology that can lead us towards post-scarcity being withheld by a small guild for their own benefit. Marx may have not gotten the exact mechanics right, but his underlying fear is playing out right before our eyes.
Why don't you specifically point to which of my comments you're referring to? Is it my response to the dubious throwaway account https://news.ycombinator.com/item?id=48695288?
China is increasingly aggressive towards its neighbors, and we don't even have to talk about Taiwan.
https://ipdefenseforum.com/2025/06/ccp-fighter-jets-harass-j...
https://en.wikipedia.org/wiki/2025_China_Coast_Guard_and_Peo...
https://www.seafoodsource.com/news/supply-trade/philippines-...
https://www.businessinsider.com/germany-china-shined-laser-p...
https://www.cbsnews.com/news/philippines-china-sea-conflict-...
America's aggression around the world is well-documented and is deplorable. That can be acknowledged and discussed without dismissing Chinese aggression. There's no law of the universe that says only one country can be a bad actor at the same time.
Now you just dropped five links in an attempt to demonstrate that Chinese aggression is comparable to US aggression, and yet none of those incidents amounted to extrajudicial execution (aka murder), which is what the person with the throwaway account was referencing.
All of it is beside my point anyway, which is that you are making assumptions about people while also asking people not to make assumptions about you.
For all you know, that throwaway account is someone who uses this site for professional development under their real name and does not want their criticism of a vindictive administration tied to them. Instead of considering that, you implied that they are a Chinese troll engaging in bad faith.
I'm inclined to believe that if someone is drawing false equivalencies and needlessly smearing their interlocutors as trolls, they are the one engaging in bad faith.
We are thoroughly off-topic at this point, so let's just end this thread here.
China engages in the extrajudicial execution of its own citizens.
https://www.amnesty.org/en/documents/asa17/009/1990/en/
https://en.wikipedia.org/wiki/Persecution_of_Falun_Gong
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3598791
China even runs a large network of "overseas police stations" to track and intimidate overseas Chinese, including dissidents.
https://en.wikipedia.org/wiki/Chinese_police_overseas_servic...
And China doesn't just target plebs. Government officials literally disappear too.
https://www.nytimes.com/interactive/2026/02/03/world/asia/ch...
I personally believe that the US has committed a significant number of crimes. That doesn't mean I'm blind to the fact that other countries commit crimes too.
I'm not sure why so many people struggle to acknowledge that all of this stuff is evil, regardless of the country doing it.
> For all you know, that throwaway account is someone who uses this site for professional development ...
Normal people using sites like this for professional development don't decide to create throwaway accounts post a single response to one short comment, buried in the context of a larger thread, that wasn't even highly downvoted or upvoted. This is what troll farms do and I'd have to question your intellectual honesty if you keep pretending that this isn't exactly what China-sponsored troll farm commenting looks like.
They have my sympathy. But I can't do anything about it. Nor can you.
What exactly do you propose in a conversation that started by people discussed using Chinese LLMs instead of LLMs from the US?
> but feign fear that your life could be shattered by Trump invading Greenland, speaks volumes about what type of person you are.
The country where I live would, per article 42 of the treaty of the European Union, have to defend Denmark if the US took military action to take over Greenland.
I am not feigning anything, this would absolutely have a major negative impact in my life.
In many ways, Iran was the best ally the EU could have hoped for. By bravely resisting US aggression, it made the war costly enough that invading Greenland would probably be politically untenable now.
> The largest contributor to the EU economy (Germany) was responsible for the worst genocide in human history.
In recent memory, yes. I can think of others that may have been worse in relative numbers, but further in the past and not as well documented to know for sure.
> The second and fourth largest (France and Spain) were two of the most brutal colonial powers in history, and their wealth today wouldn't exist without the resources they violently extracted from their colonies. Portugal was the largest participant in the Transatlantic Slave Trade.
I know quite well about all that. In fact, I come from one country that used to colonized by such powers.
You can mention many more past nations, empires and so on, all with their laundry list of atrocities. Imperial Japan, Mongolian empire, Ottomans, Aztecs, British Empire, etc and so forth.
I am still not sure what is your point. I don't think you have any to be fair, beyond some vague anxiety that people would use Chinese LLMs.
> Humans anywhere and everywhere (the US, Europe, China, etc.) can be horrible, and your loathsome comments prove that in microcosm.
I am horrible because I care about things that affect me directly while caring a lot less about thing that don't?
Or am I horrible because when I evaluate two evils I would pick one that affects me less?
Because in terms of power to enact change in the world, I have none. I am just some middle-aged software engineer living in some small town, somewhere. I don't exactly have the power to do anything for people beyond my immediate vicinity.
"Give me the strength to change what I can, the resilience to accept what I can't, and the wisdom to know one from the other"
> I am horrible because I care about things that affect me directly while caring a lot less about thing that don't?
Your previous words speak for themselves:
> The thing is, I don't care. China can be as awful as it wants to be to it's own citizens.
> Again, I don't care. China sees places like Tibet and Taiwan as China. Not my problem.
You are abhorrent, no matter how hard you want to believe that you're a helpless piece of styrofoam floating in the rough ocean.
I'll give you a hint that your parents should have taught you when you were a child: you can care about other people and have compassion for them even if you're not able to help them.
The seeds that grow into great atrocities have throughout history always been planted by apathy.
> I am just some middle-aged software engineer living in some small town, somewhere.
My sympathies to your neighbors.