White House intervention followed the release of Fable 5
TechFlow reported on June 14 that the process behind the U.S. government’s export controls on Anthropic’s advanced AI models Fable 5 and Mythos 5 has recently come to light. According to the report, the measures were not presented as a simple product takedown. Instead, the situation escalated through a sequence involving corporate safety concerns, direct pressure from senior White House officials, and the eventual use of export control powers.
The report said that two days after Fable 5 was released, Andy Jassy raised concerns with the White House. His concern centered on the risk that the model’s safety guardrails could be bypassed, allowing users to obtain high-risk capabilities that should not have been made available. That warning then entered discussions within the U.S. government and led several senior White House officials to engage directly with Anthropic.
Anthropic refused the 90-minute ultimatum
Senior White House officials later issued an ultimatum to Anthropic CEO Dario Amodei, demanding that the company voluntarily take down the relevant models within 90 minutes. Anthropic refused to comply with the request, and the two sides failed to reach an agreement on how the models should be handled.
After the disagreement remained unresolved, the Trump administration ultimately invoked export control authority on national security grounds and imposed restrictions on Fable 5 and Mythos 5. The administrative move forced Anthropic to halt the related model services worldwide, shifting what had begun as a dispute over model safety boundaries into a compulsory government regulatory action.
Anthropic later issued a statement opposing the administrative intervention. The company argued that the action lacked transparency and had not been sufficiently discussed on the basis of technical facts and risk assessment. The report also noted that Amazon was not only the party that raised safety concerns with the White House, but also one of Anthropic’s largest investors. The episode was described as reflecting a change in the U.S. government’s regulatory posture toward frontier AI models, with national security factors playing a larger role in the release and deployment of advanced models.