Anthropic has sent senior technical staff to Washington to negotiate with the Trump administration over new export controls on its AI models, *Mythos* and *Fable*, after safety concerns triggered federal restrictions, a source confirmed to Axios. The company aims to address the administration's allegations of non-compliance, which led to sweeping limitations on its most advanced systems.

The dispute escalated when Anthropic's models were placed under strict export rules, prompting the company to mobilize its technical team for urgent talks. Sources indicate virtual meetings between Anthropic engineers and White House officials began after the administration's initial outreach last Friday. Both sides claim progress, though tensions persist over compliance protocols and model accessibility.

The situation highlights the growing friction between AI developers and regulators as safety measures intensify. Anthropic's rapid response underscores the high stakes of balancing innovation with national security concerns, particularly as the U.S. tightens oversight on cutting-edge AI technologies.

This move signals a critical test for the AI industry: can companies like Anthropic navigate complex regulatory landscapes while maintaining technological leadership? The outcome could set precedents for future AI governance frameworks.