Anthropic Pulls New AI Models After Federal Directive
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic disabled access to its Fable 5 and Mythos 5 models on Friday, three days after release, after receiving a federal export control order requiring it to block foreign nationals from using the models.
- The Trump administration gave Anthropic 90 minutes to pull Fable 5, according to a source close to the company, with no prior indication of a national security threat.
- Amazon CEO Andy Jassy reportedly reached out to the administration Thursday to flag concerns about a potential jailbreaking method on Fable 5; Amazon is an investor in Anthropic.
- Anthropic pushed back publicly, arguing a "narrow potential jailbreak" shouldn't trigger a model pull and warning that if this standard were applied industry-wide it would "essentially halt all new model deployments."
- Anthropic staff met Monday with officials at the Commerce Department following the directive.
- AI policy advocates, including Brad Carson of Americans for Responsible Innovation, warned the move reflects an "ad hoc" executive approach that could cost America its AI lead, with critics noting it contradicts the White House's recent stance supporting only optional AI testing and oversight.
Why it matters: The 90-minute compliance window with no stated national security threat creates an ad hoc licensing regime for frontier AI, contradicting the White House's own stated preference for voluntary oversight. Anthropic warns this standard, applied industry-wide, would halt all new frontier model deployments, and Amazon's investor role raises competitive-dynamics questions about what triggered the directive.
