But the strength of differential privacy is that you can now make this tradeoff explicit and quantify it. I always liked it because it offers a mathematical solution to a policy problem, but then of course it's up to us to decide what parameters and tradeoff to choose. Also, some data might just not get published at all if the privacy implications are too problematic, so differential privacy might buy you more signal!
Yeah, the main issue with differential privacy is that you need competent government officials making decisions who understand math beyond a high school level.
This has dampened my opinion on Anthropic quite a bit. It's difficult to take their marketing for AI as an empowering technology seriously when they are quite clear in their new deployments that they do not mean empowering for you, but empowering for them and organizations that are in their (or the US government's, despite Anthropics performative disagreements with the administration) good graces. You are allowed to vibe code some dashboards, a web app or let it drive Excel, but anything more interesting than that is forbidden.
If it was just plain monetary concerns and sabotage of competitors I'd almost be fine with it, but it seems they actively want to monopolize most of human progress in their enlightened hands, lest the mob does something undesirable with these powers.
Don't forget their push for full regulatory capture in the name of "safety" as well so they can pull the ladder up behind them before anyone else has an equally capable model and releases it without the anti-competitive safeguards, while also pushing to completely ban open weight models, or any model trained on a certain level of compute without "rigorous" government testing and validation (which I'm sure, they'll conveniently provide the framework for).
Dampened opinion on Anthropic is an understatement.
i wonder if some lawyer may see a consumer protection class action here. In my view the Stuxnet that Anthropic pulled over its customers isn't much different from say those unauthorized extra accounts by Wells Fargo.
> asking for domestic safety testing of frontier models only is not regulatory capture
It very much is regulatory capture. The goal is to make it so only the handful of heavily capitalized tech giants and frontier labs can afford the legal and compliance rigamarole to meet the new standards. It's an effort to crowd out open source development and smaller competitors (and foreign competitors which threaten whatever moat they may have). They define safety through some speculative catastrophic threat to prevent new upstarts instead of focusing on the very real, localized harm they are causing right now.
Its also shifting the definition of safety away from their current operations and toward purely speculative future scenarios.
What backward logic is this? PRC doesn't give a fuck about how US regulates AI companies. Pushing more regulation would ensure that Chinese companies catch up sooner. If you think otherwise you need to think harder.
It's a good thing you weren't in charge of nuclear arsenals during the Cold War, sounds like your approach would have been unchecked proliferation.
Fortunately developing frontier models takes immense amounts of specific resources and knowledge. There are only a handful of companies capable of developing new cutting edge models. This is an area a few governments absolutely could coordinate on and regulate, if they were so inclined.
Obviously the current US administration is completely lacking both the will and competence to actually negotiate an agreement like that with China, and who knows if Xi would even be interested. But with different leadership we actually could be reducing our existential risks in this area much more than we are. Just like having a few thousands nukes across several countries isn't totally safe, but it's a heck of a lot safer than hundreds of thousands of nukes spread across a hundred countries.
> It's a good thing you weren't in charge of nuclear arsenals during the Cold War
You know how many nukes Soviet had right at its peak? Hint: much more than the US by the time. Non proliferation didn't stop Soviet from building more nukes at all. And it's not going to stop China from pouring more computing power into AI. History is a really good lesson.
The whole point of non-proliferation is to ensure that big boys like the US and Soviet can bully smaller guys like Venezuela and Ukraine. In this regard, non-proliferation is the most successful foreign policy ever. But it didn't win the cold war and a similar policy over AI will definitely not win the AI race (if it's a race worth winning is another issue.)
The original topic was Anthropic's guardrails, which were meant in part to stop China from using Anthropic's models to bootstrap their own. I take it the logic of the comment was that pulling attention to Anthropic's stance on regulation is switching to the topic. But for what it's worth, I also think that people are way to quick to assume that strong regulations would only help China and thereby hurt safety. There are many reasons why the opposite may be true:
- reducing demand for Chinese models reduces the incentive for Chinese companies to make them
- if US companies can't use Chinese models, they won't have an incentive to help their development
- China may enact similar regulations if the US leads, either out of concern for US safety or for commercial reasons
Also, I think some similar things can be said about AI safety measures in China aside from regulation. Currently, the US leads in model safeguards, but it isn't like China has zero interest in AI safety. Even if the US and China are rivals, there are many points of common interest (biorisk and "sci-fi" scenarios like an AI takeover, to name just two).
I don't subscribe to the belief that regulations in the US will lead to China advancing further.
But I also don't buy into the "China bad" narrative that gets frequently spread in online circles and in political circles. Its the cold war all over again, but this time its China instead of the Soviet Union.
Regardless of that, the regulations being proposed by Anthropic recently are not focused on the current issues which is my problem with all the hype marketing around hypothetical AGI/ASI. What is being proposed to be put in place will further cement the current frontier labs in their marketing leading position, and work to block new entrants, and open source competitors. That is the problem.
The other problem is none of them are talking about the real, difficult issues we are experiencing right now in the present. We don't need to talk about a sci-fi future scenario to recognize that LLMs have already caused and are causing harm in the real world. "We should probably regulate future frontier models" does nothing to help the current issues.
Wake me up when Anthropic says "The government should immediately stop us from hoovering up data and selling it back to the public. They should immediately stop us and others from enabling misinformation at scale that is already negatively effecting our democratic process. They should immediately stop us from building out new data centers until we have a large scale switch to renewables in the country, shore up the grids, or force us to generate our own power only with renewables" so on and so forth. Notice how any time the labs propose regulations, its only for a future hypothetical super intelligent model. Its never about their current operational liabilities.
And why would any regulations put in place in the USA affect the PRC in anyway whatsoever? They wouldn't. China will continue to push forward and govern things in their own way, we have zero jurisdiction over China.
> asking for domestic safety testing of frontier models only is not regulatory capture.
Yeah, asking for additional state-provided barriers to a market entry to a valuable market a provider already is one of a narrow few dominating only for firms that are a competitive threat is exactly regulatory capture.
Nothing, they are just trying to scare monger the public and prime the pump for a massive bailout when it crashes out because apparently China are the big bad meanies.
It has nothing to do with being "fine" if the PRC or anyone else for that matter get to some speculative and hypothetical ASI first. There are zero US regulations that would be effective to prevent that.
US regulations apply to US companies and citizens, exclusively. Anthropic crowding out all future potential competitors in the US via regulatory capture has no weight on what the rest of the world does.
Unless you are proposing military action over a speculative sci-fi future
PRC labs reportedly aren't even thinking about getting to ASI, much less trying. They think of AI as a technology that can provide utility across the board even without anything like superhuman smarts.
How do rules that inhibit what AI can be sold on the US market (adding additional costs to trading in that market) do anything to inhibit a competing nation from reaching ASI first? Insofar as they inhibit anyone from reaching ASI, its firms whose primary commercial interest is selling AI services in the US market, not foreign threat actors except to the extent those two categories overlap.
No, because there is zero reason to think LLMs will lead to it but we do know that the massive LLM investment has a huge financial risk for the US. Not too mention it's exacerbating the climate crisis (you know the actual thing that might end civilization, not a fantasy delusion of AGI), giving citizens cancer that live next to data centers, the extreme decrease in quality of life, and the misallocation of capital while Americans lack healthcare, childcare, housing, and education.
Also don't believe China is actually a threat to the world. That's some cold war delusional think you got there.
All the companies seem to believe is that it's okay to immiserate a large percentage for the pursuit of money, you seem to believe the lies they're feeding you.
I didn't downvote, but HN probably remembers when Anthropic's competitor was a "charity" that cared deeply about AI safety whose marketing gimmick was GPT-2 being too dangerous to release.
Anthropic's founder wants you to buy into his vision for safety, but he also wants you to buy into his vision that in two years AI will be a "country of geniuses" that will update itself, and the IPO that will fund it...
The flawed premise is thinking that AGI is a real risk, and that they care about it more than making money, that is why HN does think it's simply regulatory capture.
Right now the PRC is looking like the adult in the room. They also have a view of how AI should work that's smaller and more worker centric rather than trying to create superintelligent worker replacements.
The PRC (like any superpower) has done some bad shit, but if you're going to paint them as the bad guy keep in mind the USA has a long, long history of genocide, slavery, overthrowing foreign governments for corporate interests, unjust wars, political meddling, etc. The scales of righteousness don't tip in our favor TBH, we just have better PR and a nicer veneer over our brutality.
The US did the same thing. Environmentalist and workers rights movements date back to the 19th century. China's position on this is that the western nations that already developed are trying to pull the ladder they used up and wag a finger with false morality with the intent of maintaining global hegemony.
Yeah, I cancelled my Claude subscription yesterday after learning about their attitude of intentionally sabotaging their paying customers.
Especially after trying Fable yesterday for some benign projects and being unimpressive relative to opus.
Rolling it back is the right move, but I’m still not convinced that using them is in my best interest anymore, I’m investigating open source cloud providers now.
Models are spiky. In some narrow domains (cybersecurity, for instance) it will be a generation ahead. On the other hand a lot of people don't see a measurable difference between Opus ~4.5 and 4.6/7/8, because Anthropic taught it how to do some hard stuff better, but they didn't give it better taste or make it produce cleaner solutions to simpler problems.
Fable is very much an incremental development over Opus, and even more incremental when properly compared to its existing counterparts GPT-Pro and Gemini Deep Research.
I have a design for a really complex software I want to build and there were gaps I knew of in the design. Opus couldn’t identify them but Fable did. I’m just talking about it reviewing the design, not coding. But yeah, it’s insanely expensive. It does spin off sub agents so I suspect it might be cheaper if you had it create a bunch of plan files and then pointed deepseek at this plan files or something like that
Can you write a more specific question? I think the meaning of the comment is clear enough, but maybe you’re asking for more specifics? Ironically I can not understand what you are asking for with such a generic comment.
Looking at the comment thread you linked, this kinda looks like harassment by you rather than anything "confirmed". You seem to have an unhealthy fixation on this user, who may just be a Claude enthusiast rather than a shill as such.
Google has been doing the same thing for longer than Anthropic[0]. To protect their models from distillation attacks, they silently will downgrade the model's performance to essentially poison your training data without your knowledge.
A bit different than Anthropic refusing to assist with any AI development at all, but it's in the same vein and seems not widely known.
edit: reading the whole series of Google's AI Threat Tracker articles also provides some insight into threats Anthropic and others are dealing with
"Only I can save us". It's a classic tragedy and cautionary tale.
The idea Anthropic was going to speed run AI so they could control the usage and make it "safe" for humanity was never altruistic; it was a HUGE FUCKING RED FLAG.
They do, but only for specific definitions of "work". Like, benevolent dictators in Cuba 100% raised the literacy rate by an insane amount in just a few years (something like 20% => 80%").
If you define work as "literacy", they no doubt succeeded. But if you consider the people (and children) they tortured, raped, and murdered, suddenly literacy doesn't seem so important.
Correct, they should. If there are zero days out there, then they should be able to be found by everybody, instead of only being found by the select elite that this model is available to. Though, I very much question the truth of said ability.
And? Now all the zero days, if thats true, get discovered and patched instead of being exclusively hoarded by the select few governments and Israeli spyware companies.
Corporation cannot help but act this way. They are too big. The pressures for profit are all that matters. That is the priority. It doesn't matter what colorful words they put on the paper to make you feel better. Look at the "green" movement 20 years ago. All talk and no action.
Stop supporting organizations that don't put humans first. Don't believe a word that anyone says. Lip service is free
Yeah I'd say this has been a big concern ever since it turned out immensely expensive training methods could create effective frontier models. So far at least, open source models have kept up better than I expected, but they definitely lag the top ones and there's no guarantee the gap doesn't widen further.
Imagine the software world if Linux never existed as an effective OS and Microsoft + Apple had completely controlled computer platforms for the past decades. I think it's almost certain that both companies would be even more profitable, and the tech industry would be vastly less free and more dysfunctional .
Yes, that is basically the plan. It's based on the belief that unfettered AI would let anyone be a supervillain and destroy the world. There are enough would-be supervillains out there, but they rarely get far because they can't get teams of smart people to build doomsday machines for them. So the AI has to not let anyone do evil with it.
Unfortunately, that won't feel very much like freedom.
It sounds like you might not agree with that belief.
While I don't agree with their actions here, I do think there's sufficient reason to hold that belief.
On some fronts (e.g. security, on which you've experienced more than me), I think there are surmountable challenges. But on other fronts (e.g. bio), a single errant actor could reasonably kill millions or billions of people with sufficiently powerful AI. We don't have good defenses here, and those actors do exist.
I still don't agree with these actions, but I do think I agree with their assumptions.
The model release cards for Opus have repeatedly and consistently stressed that the model doesn't have the fiddly know-how that's required to provide meaningful assistance in possibly dangerous subfields of biology. Mythos (Fable without the overly strict guardrails) has shown improvements in things like drug design, but even then the situation isn't really that different. This risk is ridiculously overblown, and the way to manage it sensibly is to introduce meaningful oversight for actors that seek to order the actual specialized materials involved (especially any synthetically generated genes/proteins/whatever).
No, Anthropic's model cards have claimed that the models don't show considerably more uplift than previous ASL-3 models, which already showed material uplift.
I participated in the internal bioweapons uplift test for Sonnet 3.7, and even then, one non-expert got huge uplift from the model [1]. I'd consider evals a lower bound of capabilities that can be elicited from a model.
The team behind Biomni, a biomedical agent that's widely used by researchers, has continued to find consistent gains between models [2]. I trust them, because I visited them to build their HPC tool [3], which the model is quite capable of using – moreso than most grad students. The Biomni team cares a lot about about real usability for real researchers, so they have a great pulse on capabilties.
SecureBio also has some public evals [4], which have continued to show increasing uplift.
And while synthesis monitoring is a part of the solution, I think you might underestimate how much goes under the radar. See the Reedley lab incident for an example [5].
Is Anthropic still effectively throttling beneficial biomedical research? Yes! And so is OpenAI. But the underlying capability is still actually dual use.
> No, Anthropic's model cards have claimed that the models don't show considerably more uplift than previous ASL-3 models, which already showed material uplift.
Doesn't this simply amount to disagreeing about what counts as "meaningful" from a bio-safety POV? Also, even the ASL-3 deployment safeguards for Opus 4 and higher were always adopted as a mere matter of caution; it's not clear that even Anthropic believed at any point that this reflected any genuine "threshold crossing" event. So it's just not obvious how much weight we're supposed to place on that particular stance.
In normal bio, there are standardized biosafety levels, because without it there would be no standard agreement on what "meaningful" safety is. So yes, I do think there's ambiguity here.
But I don't think I've found any domain expert who thinks granting everyone raw access to the most capable models wouldn't meaningfully increase risk. OpenAI recently staffed a biological threat modeler to help quantify this risk.
(Edit: just saw your edit, this includes at Anthropic. ASL tiers were "rule-out" to exclude rather than "rule-in", so exact thresholds were murkier, but I think it's clear that models have passed that threshold by now.)
That said, there are clear steps and requirements to set up a BSL-2 or BSL-3 lab, and I think there should be similarly clear rules around model capabilties and access. The process for Anthropic and OpenAI is murky and still implictly gated on spend, which I think is holding back research.
For example, anyone who has access to a BSL-3 lab should have a clear and low-cost path to a model with corresponding capabilities, as long as they set up corresponding precautions for model access.
I think it would be a bad outcome for only frontier labs and a select few groups they choose to have access to the most capable models – which is sadly the precedent that's currently being set.
> But I don't think I've found anyone who is a domain expert who thinks granting everyone access to raw modes wouldn't meaningfully increase risk.
It depends how capable these raw models are. Biology as a field depends most on real-world knowledge, which is an expensive capability for open models targeting widespread deployment. It's quite plausible that even Opus 4 would be a lot more capable in these domains than the best universally accessible "raw models" today, quite unlike other domains such as coding or pure math. The securebio.org benchmark has spotty representation of openly available models, but it does show Kimi 2.5 being no more capable than GPT 5 mini, and clearly below o4-mini and Opus 4.0; which may be a plausible summary of where things stand today.
That's a good clarification. I've updated my comment to the "most capable models" to refer to the most recent releases.
And sure, and I love open models – I spent much of the past couple months doing additional RL on Qwen 3.6 35B A3B, Gemma 4, Kimi K2.6, and GLM 5.1. Without these open models, I'd be forced to do my research inside a frontier lab.
There's a balance to strike here, but I don't think the biological risk is overplayed. It would be very easy to accidentally cross the threshold of "meaningful" without adequate safeguards, and then be unable to undo what you've released to the world.
Do they? We don't even have single errant actors who go and kill 1000 people. I don't believe human motivations support the idea of killing so many people unrelated to you.
Even with them making those guardrails visible, it's a bit ridiculous in my eyes. I have been experimenting with smaller models, will Claude assume I'm some Chinese or Russian agent trying to distill their secrets and bar me from learning? Because that's insane. What if I discover a more efficient way to build models with Claude? Well, we'll never know now. What if someone else entirely could discover a breakthrough in how we design and build LLMs.
The whole shtick is to get you addicted whilst reducing your ability to go without, acquire power over you, jack up the prices whilst manipulating the quality of the tokens/output available to you.
Cant believe how stupid people are. You couldnt see this coming? Shame on you.
I already made up my mind, I'm not using that model if its sending proprietary code over to Anthropic, they can kiss my rear. If every frontier model winds up doing this, I will stop using them. There's plenty of employers / jobs where this is not okay behavior from an LLM.
First time? They've always been misanthropic, ironically. They seem to hate their users and think that their AI is so dangerous it'll destroy the world and not to be trusted, I mean Anthropic was literally started because people at OpenAI thought the latter was too forgiving on "safety."
> If it was just plain monetary concerns and sabotage of competitors I'd almost be fine with it, but it seems they actively want to monopolize most of human progress in their enlightened hands
But that is “plain monetary concerns and sabotage of competitors”, they are just more ambitious than most people doing sabotage of competitors in the fields they hope to dominate by that tactic.
That's why Dario's advocating for making open weight models illegal and also saying we should stop the clock on model development amongst the large labs.
>"but it seems they actively want to monopolize most of human progress in their enlightened hands, lest the mob does something undesirable with these powers"
Americans continuing to act shocked they're being cucked by corporations dampens trust and makes it difficult to buy into memes Americans are "exceptional" and "gritty", "educated", "world leaders".
Seriously the world is watching the American public get porked by grandpa and reconsidering putting their trust in not just US government as that's clearly failed, but the people themselves.
Occasional weekend warrior protest while our government destabilizes their lives? That's all the effort ya got for global allies and partners, eh?
how did you read it this way? Distill is such a big problem that distill attempts consist a significant share of their revenue(!).
A distill model with easy jailbreak can easily be used to coordinate terrorist attacks, or hostile government attacks. Read russia, north korea etc.
A distilled model can be used to rob your grandma in a very effective way. It's no longer about placing a few business logic requirements in js + css on your website. wake up .
At least Anthropic weren't lying when they said only a week ago or so "No one has figured out guardrails yet", because they apparently haven't either and Fable simply flat out rejects anything remotely connected to biology or security, no matter how trivial.
Anthropic owns the TOS... "If we think your involved in criminal activity were turning all your history over to the FBI/CIA/NSA/Local police". Then if their tooling was so good offering the same agency analysis tools to aid their experts in making some sort of decision.
But their detection isnt that good, and their analysis isnt either... this is pure theater, to create buzz (no such thing as bad press) and make their tool look far better than it is.
The reality is that, they arent even looking for the vectors that pose some of the largest risks in the modern era. And when someone uses it to do something terrible, they did not think of they are going to look dumb.
Fortunately I can't use Fable anyway, since their hyperactive content flaggers do not let you work on anything remotely biological or medical related (i.e. parse a CSV with some medical content, nope, you're probably a bioterrorist) and you get downgraded to Opus immediately.
I'm not even working on anything biological/medical, almost all PyTorch work is getting flagged (not even a safety notice and a downgrade, just an outright refusal with "this is against our ToS").
They're pissed about distillation "attacks" and locking down transformer based work to prevent that, would be my guess. Its how they'll protect "their" IP (Model Weights and other features) now that they've plundered the rest of the world's.
Yeah... I've got downgraded to Opus 4.8 in a purely theoretical discussion of a secure permission model for agent tool calls. So classifier is very broad, indeed
My 2 cents is that doctors people with lots of money and very specific needs who generally don't really go for tech jobs, so they're probably planning to create a separate monetization tier.
That, or alternatively, Mythos is so good at medical stuff, that it cam replace a lot of physician work 90% of the time, pissing off doctors, while the remaining 10% would result in very expensive lawsuits.
Third alternative: Mythos is so catastrophically bad at medical tasks that attempting to use it for medical research would instead create bioweapons. ;)
> That, or alternatively, Mythos is so good at medical stuff, that it cam replace a lot of physician work 90% of the time, pissing off doctors
Well they definitely don’t give a teaspoon of shit about putting people out of work by hawking munged-up versions of those people’s data, which was involuntarily ‘ingested’ for the benefit of society (in a way that happened to fuel a centabillion dollar industry.) So it’s prolly not that one.
>To release the model both safely and quickly, we’ve tuned these safeguards conservatively—they’ll sometimes catch harmless requests, though they trigger, on average, in less than 5% of sessions. With more capable models arriving in the coming months, we’re working to improve our safeguards and reduce false positives as quickly as we can.
Sure. IMO a lot of people will not touch fable again. The risk is to high. If they don't want the model to be good in some field they shouldn't train it on it.
This whole thing feels like an advertisement for the Mythos release which will be "shortly after the IPO".
They don't want the real risk of someone using it to make biological or genetically targeted weapons, and they don't want the social risk of someone asking it a bunch of leading questions in order to 'prove' some racist thesis or to 'prove' Mythos is woke if it declines to along with their performative inquiry.
Let's face it, if some rando comes up to and asks if you have a few minutes to talk about population biology there's a good chance they're a kook.
Surprising how little appetite for changing norms exists here on HN. Yes, the transition to agentic coding will be difficult, but to me this is mostly exciting. Despite my AI enthusiasm, I also run into shortcomings that the agents have very often, but that's a more interesting learning experience than the status quo without AI would have been!
We'll have more such disruptions and we'll learn to live with it.
The less NIMBYs are involved or aware of any infrastructure project, the better. Is that even a question? They ruined enough opportunities and wealth in the last decades.
Yes, it is a question. Obviously a balance must be struck but where the line is drawn will depend. The point about "ruining wealth opportunities" will likely rub a lot of people the wrong way even on a place like YCombinator forum
The EU's attitude to American tech firms is weird. On the one hand, they have extrajudicial private entities they outsource censorship requests to ("trusted flaggers"), which the companies have to follow at the threat of massive fines and which therefore creates the incentive to ban quickly.
On the other hand, there is stuff like this where they created another arbitrary "voluntary" mechanism to punish the companies for banning too much. I think ultimately the EU just wants a set of rules to use as a pretense to levy fines on big tech.
I think that premise is wrong - there are many interest groups, and by luck/lobbying/reaching critical mass/... they manage to put one of their interests into a law.
I think the censorship framing is quite manipulative. It is removal of unlawful content.
Is removing CSAM censorship? What about snuff?
If no, then where do you draw the line? Why can't our democratically elected governments decide what is and isn't lawful? Why should foreign Big Capital be allowed to decide instead?
Restricting the distribution of any material on content-based criteria by persons other than sender or intended receiver is censorship.
Whether it is desirable censorship or not is generally a separate issue from whether or not it is censorship, unless, for example, you have previously adopted a rule that the particular actor committing the censorship shall not engage in censorship at all, in which case they are, of course, inherently the same question. (Where this gets hairy is when one likes to pretend that one has such a rule for a particular actor, but actually really would prefer that actor to censor certain things, which sometimes occurs with modern liberal democratic regimes, and especially frequently occurs with a particular North American one which has what superficially looks like a very strong restriction in that area in its Constitution.)
>I think the censorship framing is quite manipulative. It is removal of unlawful content.
Yeah, that's called censorship. It's exactly the same thing everyone you accuse of censorship does. There is exactly zero difference beyond your support of the views/people being censored (and sometimes not even that).
>Is removing CSAM censorship? What about snuff?
Yes. Yes.
>If no, then where do you draw the line? Why can't our democratically elected governments decide what is and isn't lawful? Why should foreign Big Capital be allowed to decide instead?
Well in my country that line has been drawn. It's just recurring and persistently ignored by the state, the justice system, and private entities.
When a constitution says explicitly "no type or form of censorship is permitted", that's pretty clear what it means. You ignoring it doesn't make that line less clear.
Nice motte-and-bailey. Removing CSAM is good, therefore so is removing “we need to take back our country”:
@JudiciaryGOP has been investigating European censorship since the EU tried to silence President Trump last August. In February, we subpoenaed tech companies for their communications with foreign censors in Europe and around the world. Today, we’re releasing a report with preliminary findings about the EU’s censorship regime. The EU’s censorship law, the Digital Services Act (DSA), requires platforms to censor so-called “misinformation” and “hate speech,” even when the content “is not illegal.” New documents obtained by the Committee show that European regulators distort these terms to require censorship of legitimate political discourse that is neither harmful nor illegal. In May, the EU hosted a DSA-focused “workshop” where platforms were asked to consider hypothetical “scenarios” involving hate speech online. Unlike other EU “workshops” with tech, the public was NOT allowed to watch this one. And the EU told the platforms to NOT share the “scenarios” with the public. What were they trying to hide? Turns out, the EU wanted to hide what it defines as “illegal hate speech” that must be censored: tweeting ordinary political rhetoric like “we need to take back our country.”
If removing CSAM is good, then there is clearly a line somewhere.
The only question is "who gets to draw it"?
Why should it be foreign oligarchs and not our own democratic representatives?
To do so, ironically, constitutes "taking back our country" from autocrats abroad.
The arrogance of the House GOP trying to tell our democracy what is and isn't harmful is grating. You're not our colonial administrators. Abide by our rules or get out.
No, it constitutes the taking of the country by the autocrats at home (or in Brussels). It is ridiculous to call them "democratic representatives" when they are trying to keep secret how they are censoring the people they rule, sorry, "represent".
It's also ridiculous to claim that censoring the very desire to reclaim one's country, is reclaiming one's country, by virtue of who is doing the secret censoring.
> Who do House Republicans think they are that they can tell our democracy what is and isn't harmful? Abide by our rules or get out.
As a European I am very grateful to the Republicans for exposing how my own supranational rulers are controlling me, something which, again, they tried to keep secret, while preaching about "democracy".
Edit as reply:
> So your definition of "taking back our country" is... Americans telling us what we can and can't do in the best interest of their owners?
They're not telling us what we can and can't do. They're telling us what our rulers are keeping secret from us.
> Should the EU actually overstep, tech companies can take it to the open, public courts
I repeated how they kept the nature of censorship secret three times. You don't get to talk about "open, public courts".
> If you don't like the DSA, vote for MEPs that are opposed.
I will. But many won't, because they are being deceived by the EU regarding how the legislation works. How democratic is it when governments lie, and people vote based on those lies? And for some strange reason, none of the many misinformation watchdogs picked up on this misinformation.
So your definition of "taking back our country" is... Americans telling us what we can and can't do in the best interest of their owners?
Should the EU actually overstep, tech companies can take it to the open, public courts, thanks to our actually functioning separation of powers. That is the beauty of democracy.
If you don't like the DSA, vote for MEPs that are opposed. You're never going to be able to control what's going on in your country if you give away power to wannabe colonialists abroad.
The problem is that the EU politicians do not want the think about the problem. They want to outsource it and make someone else to pay it.
But there is no solution. Any censorship is always subjective.
Also trusted flaggers bans cannot be disputed easily. Meta rather just takes whatever trusted flaggers have flagged, real or not, and it is not their problem. It is a problem between the EU citizen and anonymous trusted flagged with no accountability.
The next step is of course corrupted trusted flaggers who can take down business pages, whatever, for a payment.
It's almost as though they want to be the government making the rules, rather than sitting back and letting the likes of META do whatever they want. The imperfect approach comes down to the reality of politics.
The EU fines are not enough to get the US tech companies to change, or even leave completely. But they are enough to continually fund the EU regulatory bureaucracy itself. So this arm of the government really only exists to preserve itself.
I would be interested to see how many EU government jobs the US tech fines are supporting. Maybe Meta or Google is indirectly the largest employer in Brussels?
Incorrect. Plenty of changes made because of that, they just apply only to EU member countries coz they still want to sell the data about other country's citizens
> But they are enough to continually fund the EU regulatory bureaucracy itself
Why do people keep repeating this falsehood? The EU budget is hundreds of billions of Euros. Even the largest fines so far would be a drop in the bucket compared to that. But more importantly it's the individual member states that invoke those fines (like Ireland with the €1.2B GDPR fine on Meta) not the EU itself.
Another way to read this is the tech companies are stupidly bankrolling the EU by not complying with the laws of the lands (EU states voluntarily enact their own implementations of EU directions in return for a slice). That’s far too good a cash cow to pass up. Keep it coming.
> The EU's attitude to American tech firms is weird.
At the core, what the difference is is how "free speech" is interpreted. In the US? Unless you literally fake-call "fire" or call for Luigi'ing someone, it's protected free speech, even if it's a bunch of Nazis holding tiki torches and showing the Hitler salute [1].
In Europe? We have been through 1933-1945. Almost every country in continental Europe was either occupied by Hitler's Germany or lorded over by one of his allies (e.g. Spain's Franco, Ante Pavelic in what would be Yugoslavia or Mussolini in Italy). An awful, awful lot of people died thanks to these regimes, and our collective learning from that history is to either ban such speech entirely or at the very least ostracize its followers. The Eastern Bloc countries who had been occupied by authoritarian pseudo Communism in the decades after WW2 added the Communist symbols (e.g. Red Star) for the same reason. Generally, we do not want a repeat of these eras.
Now with social media, we saw Americans (and people of other non European countries, but mostly Americans) openly post symbols we see as symbols of hate, we saw them call for a repeat of what happened between 1933 and the 1990s, and US platforms did barely anything against that. We tried the soft way as we almost always do, announced "hey we don't like that", platforms didn't get it under control (and some, like post-Musk Twitter, openly announced they DGAF)... and similarly to GDPR or USB-C, we acted.
What are you smoking? Any fines against these megacorps are at best a tiny fraction of their quarterly profits. They get booked as cost of doing business.
As an EU citizen I am glad there's at least some overwatch and control over these companies. I don't trust them at all. In the US unregulated capitalism is fine: everything and his mother are for sale. Here, not so much.
Ironically if there is a single population that will immensely benefit from socialism, is the United States. Yet they are raised in fear of the very doctrine that would save them.
I'll take our #1 companies, #1 tech scene, #1 innovative companies and #1 stock market over.... whatever you have, any day.
We aren't raised in "fear" of anything. We are raised to reach our potential, not depend on a nanny state or the USA. (which the EU does, you conveniently omit)
EU is an organization of bureaucrats. They want rules, first and foremost. Rules that they can lord over you and that justify their continued existence. What those rules are about is a secondary matter.
Very interesting. Often I only perceive the stock market as existing equity changing hands and the stock value of the company not being immediately relevant for its success (it's just third parties trading ownership around, after all), but I rarely heard of cash raises for the company after the initial IPO - of course only because I didn't pay attention and mostly IPOs make the news.
It's insightful to put such documents into Claude and see how they use many different financial mechanisms to raise the money. $15B sold directly to the big banks, $40B sold to the market (but also facilitated by these banks), a direct investment (PIPE) from Berkshire. Pretty cool how financial markets do these things.
excluding IPO proceeds, existing public companies in the US raise about $200B a year through selling shares on stock markets. This DOES NOT include stock-based comp, which is simply another form of funding operations post-IPO using public markets.
Stock based comp is another $350B a year in US markets alone. So if you think about public markets as an avenue for companies to raise capital, post-IPO firms are doing it to the tune of more than half a trillion a year.
Seems like a very bad precedent if that were to become the legal interpretation. I can understand if there were requirements for AI companies to document their efforts to reduce harms in their model reports, but ultimately this is a general intelligence (to which degree you can debate) and it's part of its purpose and utility to be able to converse naturally.
Of course it should steer people away from harmful thoughts like any sensible human would, but that's all you can do, really.
reply