Anthropic sent senior technical staff to Washington over the weekend to negotiate with government officials after the Trump administration issued export controls that forced the company to shut down its two most powerful AI models.
The company dispatched senior security researchers to meet with government officials after the Trump administration forced its two most powerful models offline
Anthropic CEO Dario Amodei (Krisztian Bocsi/Bloomberg via Getty Images)
Anthropic sent senior technical staff to Washington over the weekend to negotiate with government officials after the Trump administration issued export controls that forced the company to shut down its two most powerful AI models.
The company dispatched senior security researcher Nicholas Carlini, Logan Graham, who leads the team that evaluates models for risks, and Dave Orr, the company's head of safeguards, according to The Wall Street Journal. They are expected to meet with government security researchers in an effort to de-escalate the dispute.
Friday brought an executive action from the Trump administration prohibiting Anthropic's Fable 5 and Mythos 5 models from being accessed by foreign governments, companies, and individuals. Citing the breadth of the restriction — which swept in foreign-born Anthropic employees alongside overseas governments and businesses — the company determined that universal suspension was the only workable path to compliance, and pulled access for all users. Access to other Anthropic models was not affected.
Representing the government in Saturday's calls were Commerce Secretary Howard Lutnick and National Cyber Director Sean Cairncross; Anthropic sent co-founder and chief compute officer Tom Brown alongside public policy chief Sarah Heck. Both parties are interested in resolving the situation, but the shape of a potential agreement remains unclear, the Journal said.
The restrictions stem from research by Amazon $AMZN scientists who found that Fable 5's safeguards could be partially bypassed, alarming White House officials. The research was brought to the attention of U.S. officials, including Treasury Secretary Scott Bessent, by Amazon CEO Andy Jassy. Trump personally authorized the export restriction and tapped Lutnick to coordinate the administration's handling of the matter.
Anthropic disputes the severity of the finding. The company said it reviewed what it believes is the underlying report and concluded that the level of capability demonstrated is available from other publicly deployed models, including OpenAI's GPT-5.5, and is used by cybersecurity defenders routinely. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," Anthropic said in a statement.
A letter circulated among cybersecurity researchers on Sunday urged the administration to reverse course, with signatories writing that the controls had "taken the best models away from defenders, created market uncertainty, and risked America's AI leadership without any real risk to justify it."
The shutdown comes at a difficult moment for Anthropic, which filed a confidential IPO prospectus with the Securities and Exchange Commission earlier this month, disclosing a revenue run rate of $47 billion and a valuation of $965 billion. The company has faced prior friction with Washington: the Department of Defense previously designated Anthropic a supply chain risk following the collapse of talks between the two sides, and a lawsuit Anthropic filed to challenge that designation remains active.
Join 500,000+ readers who start their day with Quartz.
By subscribing, you agree to our Terms of Service and Privacy Policy.