Senior technical staff from Anthropic traveled to Washington, D.C., over the weekend to engage with White House officials in an effort to challenge recent export controls imposed on the company’s Mythos and Fable AI models. The move comes after the Biden administration implemented the restrictions last Friday, citing national security concerns.

The White House’s decision followed intelligence reportedly provided by Amazon CEO Andy Jassy, indicating that the safety mechanisms, or guardrails, built into the AI models could be circumvented, potentially posing security risks. Anthropic has maintained that AI safety remains a central priority for the company, with CEO Dario Amodei and other executives publicly emphasizing their commitment to robust safeguards.

Despite these assurances, sources say White House officials expressed frustration over what they perceived as Anthropic’s downplaying of the vulnerabilities. Since the controls were announced, the company has engaged in multiple virtual discussions with members of the administration in an attempt to address these concerns and seek reconsideration.

The export controls restrict the international distribution of the Mythos and Fable AI systems, which are known for their advanced capabilities in natural language processing. These restrictions align with broader efforts by the U.S. government to regulate the spread of advanced artificial intelligence technologies amid rising geopolitical and security considerations.

Anthropic’s recent visit and ongoing discussions underscore the tensions between emerging AI companies and government authorities aiming to balance innovation with safety and national security priorities. As these talks continue, the outcome could influence the regulatory landscape for AI deployments both domestically and abroad.