Anthropic Refused to Fix Fable 5 Jailbreak Before US Export Control Order, Sacks Says
David Sacks said the export-control order that forced Anthropic to disable Fable 5 and Mythos 5, its most advanced AI models, for all customers came after the company refused to fix or de-deploy a jailbreak of Fable 5's guardrails. Sacks said the administration hopes the controls can be lifted if Anthropic remediates the safety issue; the order formally suspended access by foreign nationals, but Anthropic said it had to shut the models off for everyone to ensure compliance.
Anthropic has disputed the government's assessment, saying it was given only verbal evidence of a narrow, non-universal jailbreak and that the cited capability is already available in other models, including OpenAI's GPT-5.5. The shutdown also fueled interest in decentralized-AI alternatives, with Bittensor's TAO rising 17.2% and Venice and Morpheus tokens climbing as traders pushed a "permissionless AI" pitch.
From the sources (25 posts)
@ylecunRT @askalphaxiv: As believers of open research, we are disappointed to see Anthropic silently degrading Fable 5 for AI development "Any to…
@zeffmaxNEW: Anthropic is walking back Claude Fable 5's policy to covertly degrade performance for competing AI researchers, after facing fierce backlash. “We’re changing Fable 5’s safeguards for frontier LLM development to make them visible,” Ant
@zeffmaxHere's the new policy: "Starting this week, flagged requests will visibly fall back to Opus 4.8. On the API, any flagged requests will return a reason for their refusal. You will see this every time it happens."
@zeffmaxAnthropic says it implemented these safeguards to limit foreign adversaries, and that making them hidden was a way to make them more narrow. However, the company now says it made the wrong tradeoff.
@tftc21Anthropic is walking back a policy in Claude Fable 5 that would have covertly degraded the model's performance when it detected users were doing frontier AI research. The move would have invisibly sabotaged researchers trying to use Claud
@techmemeAnthropic backtracks on a policy limiting Claude Fable 5's ability to develop other AI models, after significant backlash from the AI research community (@zeffmax / Wired) (Visit Techmeme dot com for the link and full context!)
@eigengenderRT @ZeffMax: NEW: Anthropic is walking back Claude Fable 5's policy to covertly degrade performance for competing AI researchers, after fac…
@simonwVery pleased to hear Anthropic have walked back this policy
@emollickThis was perhaps the most controversial aspect of the guardrails around Fable, now being rolled back.
@mtsliveSITUATION UPDATE: Anthropic is reversing its Fable 5 policy of covertly degrading performance for competing AI researchers, per Wired.
@andrewcurran_New policy: 'Starting this week, flagged requests will visibly fall back to Opus 4.8. On the API, any flagged requests will return a reason for their refusal. You will see this every time it happens.'
@lcastricatoRT @ZeffMax: NEW: Anthropic is walking back Claude Fable 5's policy to covertly degrade performance for competing AI researchers, after fac…
@zacharynadoRT @ZeffMax: Here's the new policy: "Starting this week, flagged requests will visibly fall back to Opus 4.8. On the API, any flagged requ…
@jekbradburyRT @simonw: Very pleased to hear Anthropic have walked back this policy
@tradfi*ANTHROPIC FACES EXPORT CONTROLS BLOCKING FOREIGN ACCESS TO ADVANCED MYTHOS 5, FABLE 5 MODELS *ANTHROPIC FACES POTENTIAL FINANCIAL AND CIVIL PENALTIES FOR FAILURE TO COMPLY WITH NEW MANDATES
@financialjuiceTrump administration blocks overseas access to Anthropic's leading AI - Axios
@financialjuiceTrump administration blocks foreign access to Anthropic’s most powerful AI: Axios
@axiosScoop: Trump admin blocks foreign access to Anthropic's most powerful AI
@phildstewartTRUMP ADMIN BLOCKS FOREIGN ACCESS TO ANTHROPIC'S MOST POWERFUL AI - AXIOS
@techmemeSource: Howard Lutnick told Anthropic that Mythos 5 and Fable 5 would be subject to export controls after a company claimed it was able to jailbreak Mythos (Axios) (Visit Techmeme dot com for the link and full context!)
@polymarketJUST IN: Trump administration moves to block foreign governments, companies & individuals from accessing Anthropic's Fable & Mythos models.
@mtsliveSITUATION DETECTED: The Trump administration is placing export controls on Anthropic’s Mythos and Fable 5 models, blocking foreign governments, companies, and individuals from access, per Axios.
@deanwballIf this is true, it is just baffling. An administration whose posture is that we *should* export advanced AI chips to China, which also wants to ban… Britain (and every other non-American on Earth)… from using our best models? I have no wor
@andrewcurran_The Trump administration has placed Anthropic's Mythos 5 and Fable 5 under export controls. Commerce Secretary Howard Lutnick sent a letter to Dario Amodei tonight stating that foreign governments, companies, and individuals will no longer
@andrewcurran_This allegedly was triggered by a successful jailbreak of Mythos by an unnamed group who reported it to the government. We will hopefully get details soon.