Anthropic races to overturn US export ban on Mythos 5
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic received a US export control directive at 5:21 PM Friday ordering it to suspend access to its Mythos 5 and Fable 5 models by any foreign national inside or outside the US, including its own foreign-national employees.
- The Trump administration gave Anthropic a 90-minute ultimatum around 1 PM ET Friday to shut down access or face Commerce Department export controls; CEO Dario Amodei spoke with Treasury Secretary Scott Bessent, Commerce Secretary Howard Lutnick, and National Cyber Director Sean Cairncross within hours.
- Anthropic said the flagged jailbreak was a "narrow, non-universal" issue reported by an unnamed entity, and that the same capability was "widely available from other models (including OpenAI's GPT-5.5)."
- Amazon researchers red-teamed Fable 5 and their work was explicitly cited in government conversations, with some reports identifying CEO Andy Jassy as the person who flagged concerns to the US government.
- Alex Stamos, chief product officer at Corridor, organized a public letter signed by tech and cybersecurity executives calling for the restrictions to be repealed, warning that American labs are only "something like six months ahead" of Chinese models.
- OpenAI, Google, and Microsoft have all released comparable products to Anthropic's Mythos with similar cybersecurity claims, meaning a sustained ban could pressure the administration to restrict competitors' models too.
- Anthropic had banked on Mythos to help recover from months of high-profile clashes with the US Department of Defense over autonomous weapons, making the timing of the directive especially damaging.
Why it matters: If the export order holds, Anthropic's flagship model is effectively frozen while OpenAI, Google, and Microsoft continue selling comparable products, forcing the administration to either broaden restrictions industry-wide or cede ground to Chinese labs that Stamos estimates are only six months behind. For Anthropic, the directive lands on top of months of Pentagon friction over military AI use, threatening both its commercial recovery and its standing as a US AI champion.
