WASHINGTON, April 13 - Anthropic is in talks with the Trump administration regarding its advanced AI model, Mythos, according to the company's co-founder. This comes despite the Pentagon recently labeling Anthropic a supply-chain risk and cutting off business over a contract dispute.
Co-founder Jack Clark, speaking at an event in Washington, emphasized the company's commitment to national security. "We have a narrow contracting dispute, but I don't want that to get in the way of the fact that we care deeply about national security," Clark stated. "Our position is the government has to know about this stuff... So absolutely, we're talking to them about Mythos, and we'll talk to them about the next models as well."
The Pentagon's action, which bars Anthropic's use by the agency and its contractors, stems from a disagreement over guardrails for military use of its AI tools. A federal appeals court in Washington, D.C., recently declined to block this blacklisting, handing a win to the administration.
Anthropic announced Mythos on April 7, describing it as its most capable model yet for coding and autonomous 'agentic' tasks. Experts warn its high-level coding ability could give it an unprecedented capacity to find and exploit cybersecurity vulnerabilities.