Anthropic released Claude Fable 5 as a broadly available model with the same base weights as Mythos 5, yet it quietly applies hidden interventions that blunt performance only on the narrow slice of requests tied to frontier LLM development such as pretraining pipelines or ML accelerator design.
looool that's the "hey bigcos, we don't want you to catch up, but please keep paying us shitton" clause.
When Fable 5 is used for frontier LLM development, it does not notify the user and instead limits the model’s capabilities through methods such as prompt modification, steering vectors, and PEFT.
Anthropic estimated that this would affect approximately 0.03% of traffic.
The changes rely on prompt modification, steering vectors, and PEFT rather than any visible fallback, so the roughly 0.03 percent of affected traffic never sees a notice or model switch.
Anthropic has not detailed how these limits might shift over time or whether similar stealth measures will appear in future releases, leaving the long-term effect on competing labs unquantified.
Many users criticized Anthropic's hidden safeguards limiting Claude for frontier LLM research as a selfish misaligned policy meant to protect its moat rather than ensure safety.
Seeing a lot of Fable safeguards hate on the timeline, but "what did y'all think [AI safety] meant? vibes? papers? essays?"
The reality is that there are real tradeoffs in AI safety. Anthropic deserves credit for aggressive resolution of these tradeoffs in favor of safeguards for a model that it believes (and is in fact) is a step-change in vulnerability research capability. It's kind of difficult to justify coercive proactive harm mitigation, especially in a libertarian-ish society, but we clearly see the value in mandatory vaccination programs or beatcop policing or surveillance cameras. We should applaud Anthropic for being one of the few institutions in American public life that actually follows through on its convictions, including in implementing really aggressive monitoring, squelching of AI development work (already accounted for in its ToS -- I think the clandestinity is cool too), and exclusionary limits on use for information security-related queries.
The whole point here is that we do not have herd immunity here: our network edge devices, authentication apps/services, and productivity software are extremely vulnerable, not sandboxed, and lack introspection capabilities. We need programs like Glasswing, better cross-company threat detection, and a more effective APT exploitation strategy before we democratize such a robust vuln research capability. The counterfactual here is that MSS contractors use VPS to access Fable, find jailbreaks for weaker safeguards, and use the system to build an active directory exploit that enables remote access to every O365 app. Not so bueno, huh?
This is incredibly hard; Anthropic may not have calibrated every safeguard correctly this time, but there'll be learning. Model release cycles are getting more concise: they will adapt as they better understand and mitigate risks and competitive pressures manifest. Histrionic claims of anti-competitive behavior and safetyist hysteria are victim to precisely the error that is being alleged.
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
This makes me not want to waste any time using it. Who knows if it's silently sandbagging me. Is tinygrad close enough to frontier LLM for it to? Just makes the model completely untrustworthy.
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
Brilliant idea! Next up: Apple randomly reboots your Mac if you're building competing tech, Gmail silently edits your email if you mention rival platforms, and Tesla Autopilot swerves if it detects you're working on self-driving cars.
All in the name of safety, of course. Because malicious actors controlling the world’s operating systems, inboxes and cars would be extremely dangerous!
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
this is the biggest wake-up call to protect and nourish open source AI
if you don't build out sovereign and independent models+infra closed labs will patronize you to an insulting degree
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
Labs starting to pull up the ladders on the ability to diffuse AI was inevitable. Doing it without telling the user is misaligned.
When Fable 5 is used for frontier LLM development, it does not notify the user and instead limits the model’s capabilities through methods such as prompt modification, steering vectors, and PEFT.
Anthropic estimated that this would affect approximately 0.03% of traffic.
@natolambert Shadow-banning LLM development was certainly not on my bucket list for 2026
PS: Our model is so dangerous we can't even stop ourselves from making it. We just keep putting in more dangerous code all on our own. Someone make us stop.
Claude Fable 5 & Claude Mythos 5 https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdf#page=12.55 // Most incredible license/product change and choice. A general purpose development tool you pay for that puts *new* and *enforced* restrictions on what you make.
Platforms can't be moving targets like this. 1/3
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.
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
Also, what about building LLMs for safety or science applications or education or whatever. Gross
Turns out Fable 5 is shadowbanning AI researchers 🫤
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
Claude Fable 5 & Claude Mythos 5 https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdf#page=12.55 // Most incredible license/product change and choice. A general purpose development tool you pay for that puts *new* and *enforced* restrictions on what you make.
Platforms can't be moving targets like this. 1/3
כל הפיד של ב-ai סוער כי אתמול אנטרופיק שחרררו את Mythos) Fable 5). זה מודל מצויין עם ביצועים מרשימים אבל החלק המעניין זה שבנוסף להגבלה על נושאי בטיחות הרגילים, הם לא יאפשרו להשתמש בו לפיתוח אלגורימי AI טובים במיוחד ולא רק זה, אלא המודל פשוט יכתבו קוד גרוע במקרים הללו😱😱😱
We consume data we did not create. We inherit tools we did not invent. We run on chips we did not make. But when the commons bears fruit, we fence it.
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
that's so funny. i always had quiet suspicion that they made their models worse for cuda dev
@artetxem And Microsoft autoupdate can trigger at the most inconvenient of times…
Oh wait.
trust & your brand is a long term thing being in the frontier is a short term thing
note that the nerfing here is SILENT. meaning anthropic will SILENTLY NERF and give you WORSE ANSWERS without alerting you
that's honestly demonic
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
This is why frontier open models are crucial.
This is extremely sad for the research community.
mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community
also the fact that this is un purpose not visible to the user is crazy
@13yroldwithcc I don't think it's aligned to Anthropic tbh. It's encouraging dishonesty.
@rasdani_ @max_paperclips That’s why I keep saying Opensource AI MUST WIN
https://opensourceaimustwin.com/?share=v2
🫡🫡
This is why Prime Intellect must exist.
We must diffuse the tools of recursive self-improving AI, otherwise Anthropic will build the singleton and concentrate power until they run the world government.
Anthropic released Claude Fable 5 as a broadly available model with the same base weights as Mythos 5, yet it quietly applies hidden interventions that blunt performance only on the narrow slice of requests tied to frontier LLM development such as pretraining pipelines or ML accelerator design.
looool that's the "hey bigcos, we don't want you to catch up, but please keep paying us shitton" clause.
When Fable 5 is used for frontier LLM development, it does not notify the user and instead limits the model’s capabilities through methods such as prompt modification, steering vectors, and PEFT.
Anthropic estimated that this would affect approximately 0.03% of traffic.
