Nathan Lambert, post-training lead at AI2, publicly called out Anthropic for yanking access to its most advanced models without any heads-up, labeling the move a clumsy rug pull that left his ongoing work in limbo even though he had braced for eventual limits.
I got a good nights sleep and I’m still just as angry about Anthropic’s choices.
I enjoy working in AI so much and to have my access to the cutting edge models for my work rugpulled in an under the table fashion is appalling.
I expected to be restricted eventually, but not now, and to be told it directly.
Lambert and others expected tighter gates eventually but got no direct notice or explanation, and the exact triggers or affected models stay unspecified in public posts.
The abrupt change has already nudged some affected researchers to test OpenAI tools like Codex as a workaround, though how many will fully migrate remains unknown.
Many users slammed Anthropic for cutting researcher access to Claude models without notice, calling the move hypocritical on ethics, covert, anti-science, and a clear erosion of trust.
To me this paints Anthropic clearly as anti science, and therefore anti progress and anti safety.
I got a good nights sleep and I’m still just as angry about Anthropic’s choices.
I enjoy working in AI so much and to have my access to the cutting edge models for my work rugpulled in an under the table fashion is appalling.
I expected to be restricted eventually, but not now, and to be told it directly.
@natolambert It’s completely changed my relationship towards Claude (Code). I feel really weird using it. Going to try to shift my usage to Codex. I’ll see if it sticks this time- I’ve tried to switch in the past but kept coming back to CC’s better interface.
I got a good nights sleep and I’m still just as angry about Anthropic’s choices.
I enjoy working in AI so much and to have my access to the cutting edge models for my work rugpulled in an under the table fashion is appalling.
I expected to be restricted eventually, but not now, and to be told it directly.
@natolambert Had a perfect 8 hours sleep. Woke up feeling frustrated too.
I got a good nights sleep and I’m still just as angry about Anthropic’s choices.
I enjoy working in AI so much and to have my access to the cutting edge models for my work rugpulled in an under the table fashion is appalling.
I expected to be restricted eventually, but not now, and to be told it directly.
@bindureddy Yeah, I've never seen a new major model piss people off so much:
Anthropic put a softball over Sam Altman's plate (and Elon's too). Can they hit it?
"Misanthropic."
I've never seen the AI community so angry at a major new model release. I asked my AI (an agent that @blevlabs made for me) to gather all the backlash.
+++++++++++
THE BACKLASH AGAINST CLAUDE FABLE 5'S RESTRICTIONS
The best analysis of why this matters:
@EnoReyes — "It's about who gets to decide, and whether you ever find out when they do. Fable won't fall back to a different model and tell you. It just limits the output through prompt modification, steering vectors, or PEFT. You won't be told when it happens to you."
THE VIRAL TAKE:
@0xBalloonLover — "anthropic won't let you use fable for biology, chemistry, ai research, or anything that accelerates human progress. that makes it the perfect tool for developing blockchains"
POWER CONCENTRATION:
@ClementDelangue (HuggingFace CEO) — "Concentration of power, capabilities and economic wealth is the biggest risk in AI. We need open science and open-source more than ever!"
@jeremyphoward (http://fast.ai) — "Anthropic has chosen the opposite of the safe path: they are allowing themselves, the current top lab, to use their top model for frontier AI research. They've said they'll sabotage others who try."
@gneubig (Graham Neubig, CMU) — "First they came for the model builders... I feel we're getting a glimpse of a future where AI is only provided to a privileged few, and that's not a future I want to live in."
OPEN RESEARCH:
@askalphaxiv (AlphaXiv open science) — "As believers of open research, we are disappointed to see Anthropic silently degrading Fable 5 for AI development."
@willccbb — "it is the first publicly available model that i am explicitly not allowed to use for my work, because anthropic holds the view that the work i do to facilitate open model research is harmful. capability and alignment research are coupled. anthropic wants to be the only lab."
NOUSRESEARCH / HERMES (which Anthropic has nerfed multiple times):
@Teknium (NousResearch co-founder) — "What's crazy to me is that Fable is blocked from life sciences broadly, nerfed even if you get passed the classifiers and filter level blocks. The whole point of AGI/ASI is to cure all diseases. Everything else is just nice to haves. But Anthropic wants to close off that path."
THE MECHANISM:
@kimmonismus — "When the model is used for frontier LLM development, it apparently does not simply refuse or warn the user. Instead, it quietly limits its own effectiveness through techniques like prompt modification, steering vectors, and PEFT."
MEDICAL COMMUNITY:
@DeryaTR (immunologist, BSL-3 certified) — "The word 'cancer' is flagged as a biosecurity risk by Claude Fable 5! I also tried to code a website on cancer mutations & Fable 5 was immediately removed from my list!"
@DeryaTR — "I can't even say 'hello' to Fable 5 except in incognito mode (memories off), because it knows I am a biomedical researcher!"
@DeryaTR — "I am not even allowed to use Fable 5 with memories on! Apparently the model thinks I am a biosecurity risk, though I had been certified to work in biosecurity level 3 labs! Not a single Anthropic person has tried to reach out to help either!"
@banteg — "claude fable 5 refuses completely benign tasks like analyzing bloodwork."
@bneyshabur — "Working on AI for cancer? Sorry, I can't help you. Working on AI for Alzheimer's Disease? Sorry, I'm becoming a bit dumb when it comes to the AI part of it."
SUBSCRIPTION CANCELLED:
@bubbleboi — "Have canceled my team subscription for Claude Pro. Idc how good that model is, it's not good enough for me to support people who actively stifle innovation and gate keep knowledge that they didn't even create."
BILLING AND PRIVACY:
@GergelyOrosz (The Pragmatic Engineer) — "Things I really dislike about Fable: 1. Anthropic collects my prompt history, stores it, and does whatever they want with it for 30 days. No opt-out. 2. They can nerf their most expensive model without telling me, billing me the same amount, wasting my time. Whenever they want."
THE KARPATHY QUESTION:
@SanthProject — "the old @karpathy would never support a company that fucks other llm researchers. Were the stock benefits that good?"
THE MONOPOLY CHARGE:
@tunguz (TabulAI founder) — "Starting to suspect that Anthropic's putative security and safety considerations are largely posturing and performative."
@BlancheMinerva — "Anthropic is choosing to make decisions that make the world a significantly worse and potentially more dangerous place."
@LinusMixson — "Dario personally, and Anthropic as a whole, have been extremely straightforward about wanting a monopoly for a long, long time."
@TheAhmadOsman — "I started warning people about Anthropic more than a year ago... Today I am vindicated, everybody knows that company only acts in bad faith."
WHY REGULAR PEOPLE WILL EVENTUALLY CARE:
@DanJeffries1 — "The fury is real and what all of us in the open community have been saying for years and yet regular folks don't get it yet because nothing they care about is restricted or taken away for 'safety.' They will care a LOT in the future when AI is integrated into every aspect of [life]."
Full analysis: https://alignednews.com/ai
@natolambert You aren't alone:
"Misanthropic."
I've never seen the AI community so angry at a major new model release. I asked my AI (an agent that @blevlabs made for me) to gather all the backlash.
+++++++++++
THE BACKLASH AGAINST CLAUDE FABLE 5'S RESTRICTIONS
The best analysis of why this matters:
@EnoReyes — "It's about who gets to decide, and whether you ever find out when they do. Fable won't fall back to a different model and tell you. It just limits the output through prompt modification, steering vectors, or PEFT. You won't be told when it happens to you."
THE VIRAL TAKE:
@0xBalloonLover — "anthropic won't let you use fable for biology, chemistry, ai research, or anything that accelerates human progress. that makes it the perfect tool for developing blockchains"
POWER CONCENTRATION:
@ClementDelangue (HuggingFace CEO) — "Concentration of power, capabilities and economic wealth is the biggest risk in AI. We need open science and open-source more than ever!"
@jeremyphoward (http://fast.ai) — "Anthropic has chosen the opposite of the safe path: they are allowing themselves, the current top lab, to use their top model for frontier AI research. They've said they'll sabotage others who try."
@gneubig (Graham Neubig, CMU) — "First they came for the model builders... I feel we're getting a glimpse of a future where AI is only provided to a privileged few, and that's not a future I want to live in."
OPEN RESEARCH:
@askalphaxiv (AlphaXiv open science) — "As believers of open research, we are disappointed to see Anthropic silently degrading Fable 5 for AI development."
@willccbb — "it is the first publicly available model that i am explicitly not allowed to use for my work, because anthropic holds the view that the work i do to facilitate open model research is harmful. capability and alignment research are coupled. anthropic wants to be the only lab."
NOUSRESEARCH / HERMES (which Anthropic has nerfed multiple times):
@Teknium (NousResearch co-founder) — "What's crazy to me is that Fable is blocked from life sciences broadly, nerfed even if you get passed the classifiers and filter level blocks. The whole point of AGI/ASI is to cure all diseases. Everything else is just nice to haves. But Anthropic wants to close off that path."
THE MECHANISM:
@kimmonismus — "When the model is used for frontier LLM development, it apparently does not simply refuse or warn the user. Instead, it quietly limits its own effectiveness through techniques like prompt modification, steering vectors, and PEFT."
MEDICAL COMMUNITY:
@DeryaTR (immunologist, BSL-3 certified) — "The word 'cancer' is flagged as a biosecurity risk by Claude Fable 5! I also tried to code a website on cancer mutations & Fable 5 was immediately removed from my list!"
@DeryaTR — "I can't even say 'hello' to Fable 5 except in incognito mode (memories off), because it knows I am a biomedical researcher!"
@DeryaTR — "I am not even allowed to use Fable 5 with memories on! Apparently the model thinks I am a biosecurity risk, though I had been certified to work in biosecurity level 3 labs! Not a single Anthropic person has tried to reach out to help either!"
@banteg — "claude fable 5 refuses completely benign tasks like analyzing bloodwork."
@bneyshabur — "Working on AI for cancer? Sorry, I can't help you. Working on AI for Alzheimer's Disease? Sorry, I'm becoming a bit dumb when it comes to the AI part of it."
SUBSCRIPTION CANCELLED:
@bubbleboi — "Have canceled my team subscription for Claude Pro. Idc how good that model is, it's not good enough for me to support people who actively stifle innovation and gate keep knowledge that they didn't even create."
BILLING AND PRIVACY:
@GergelyOrosz (The Pragmatic Engineer) — "Things I really dislike about Fable: 1. Anthropic collects my prompt history, stores it, and does whatever they want with it for 30 days. No opt-out. 2. They can nerf their most expensive model without telling me, billing me the same amount, wasting my time. Whenever they want."
THE KARPATHY QUESTION:
@SanthProject — "the old @karpathy would never support a company that fucks other llm researchers. Were the stock benefits that good?"
THE MONOPOLY CHARGE:
@tunguz (TabulAI founder) — "Starting to suspect that Anthropic's putative security and safety considerations are largely posturing and performative."
@BlancheMinerva — "Anthropic is choosing to make decisions that make the world a significantly worse and potentially more dangerous place."
@LinusMixson — "Dario personally, and Anthropic as a whole, have been extremely straightforward about wanting a monopoly for a long, long time."
@TheAhmadOsman — "I started warning people about Anthropic more than a year ago... Today I am vindicated, everybody knows that company only acts in bad faith."
WHY REGULAR PEOPLE WILL EVENTUALLY CARE:
@DanJeffries1 — "The fury is real and what all of us in the open community have been saying for years and yet regular folks don't get it yet because nothing they care about is restricted or taken away for 'safety.' They will care a LOT in the future when AI is integrated into every aspect of [life]."
Full analysis: https://alignednews.com/ai
yeah exactly the same, it's been quite a weird day for me so far (EU time), there is little things like codex being fast which make my brain telling me the output will be bad 😂
oh and also i miss quite a lot claude for reading paper i miss a lot claude, i guess could still use it but it just feels super weird :(
@natolambert Was this not obvious already? It's the most woefully out of touch group of individuals in the field. They genuinely believe they are doing the right thing because they are better than you and will make more responsible decisions, just like this one.
@natolambert Disappointed yes, but surprised? Surely not
@matanSF I’m a bit surprised still, mostly in the method of how they did it.
@natolambert I understand why you’re frustrated, but refusing to help the CCP or unvetted third parties does not make them anti-science or anti safety. Given their reasoning, they are acting quite rationally, and it seems to me that this is the best way to proceed.
@finbarrtimbers @natolambert I feel like I had hired a contractor, and they came out and said they wouldn't send their best employees because the firm didn't agree with my work.
How can I trust whoever they do send? opus / sonnet?
@finbarrtimbers I feel like I can’t use it. Need to figure out a new long term solution for my workflows which pretty much all discuss frontier ai of some form
@natolambert Do you think it’s plausible that other model providers have similar policies, even if undisclosed?
I remember I had some fear about openai being the only ones to have plans 'forever'.
I tell myself that it's temporary, This lead won't last. We my not like some of their personality traits but I am happy Elon and Sam, such ferocious competitive people lead competing enterprises to anthropic.
I am also happy China does what it does, and Nvidia is in the open model game.
@finbarrtimbers @natolambert Codex is actually lowkenkinuinely better than opus
@Ymirpcl @natolambert Again the issue is secretly sandbagging, they are ofc within their rights to restrict access generally
@finbarrtimbers @natolambert Use the Codex app, is so much better
@natolambert @MaximeRivest Couldn’t use Fable 5 for even a basic tasks in Healthcare. Had to fallback on Opus 4.8, but I am worried this becomes a thing! What if Anthropic extends these limits to the next Opus model!
@natolambert Wtf @FTC what are we even doing here? For what other product or service would it be acceptable to bait and switch like this???? Why isn’t @SenWarren making a big stink this instead? What a joke.
@Ymirpcl @natolambert It is evil to secretly sandbag answers, that’s my opinion it’s okay if you disagree
Nathan Lambert, post-training lead at AI2, publicly called out Anthropic for yanking access to its most advanced models without any heads-up, labeling the move a clumsy rug pull that left his ongoing work in limbo even though he had braced for eventual limits.
I got a good nights sleep and I’m still just as angry about Anthropic’s choices.
I enjoy working in AI so much and to have my access to the cutting edge models for my work rugpulled in an under the table fashion is appalling.
I expected to be restricted eventually, but not now, and to be told it directly.