Pentagon CTO rules out resolution with Anthropic, calls Mythos a broader ‘cyber moment’
Pentagon CTO Emil Michael ruled out any reconciliation with Anthropic, calling the company’s exclusion from military work a necessary stance amid broader cybersecurity concerns. The release of Anthropic’s cybersecurity model Mythos has sparked discussions with the White House, though the Pentagon remains firm on its ban, citing integration risks and past disputes over surveillance and autonomous weapons use.
The Pentagon has no plans to resolve its conflict with Anthropic, despite the White House adopting a softer stance toward the AI company. Emil Michael, under secretary of Defense for research and engineering, stated during a fireside chat at the AI+ Expo in Washington that the Department of Defense (DOD) would not reconsider its ban on Anthropic’s models for classified networks. He emphasized the Pentagon’s refusal to rely on a single AI vendor, directly referencing Anthropic’s exclusion. The ban stems from a dispute over potential misuse of Anthropic’s AI for domestic surveillance or fully autonomous military attacks. The Pentagon blacklisted Anthropic over two months ago, a decision later extended to civilian agencies under former President Trump’s administration. Michael highlighted the complexity of integrating and later removing AI vendors from classified systems, calling the process highly technical and resource-intensive. While the White House, including Trump, has praised Anthropic’s team as ‘high-IQ’ and expressed openness to cooperation, the Pentagon remains steadfast. Trump noted that Anthropic’s capabilities could be valuable, but Michael framed the conflict as a broader cybersecurity issue rather than a company-specific dispute. The release of Anthropic’s new cybersecurity model, Mythos, which detects decades-old vulnerabilities in software and infrastructure, has intensified discussions within the administration. Officials including Treasury Secretary Scott Bessent, Federal Reserve Chair Jerome Powell, and White House Chief of Staff Susie Wiles have engaged with Anthropic and other tech firms about Mythos’ risks and benefits. Michael described the model’s emergence as part of a larger ‘cyber moment,’ questioning how the U.S. government should address evolving cyber threats. He stressed that the Pentagon’s stance reflects a strategic shift toward securing classified networks against advanced AI-driven risks, regardless of diplomatic or political pressure.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.