Pentagon’s AI Strategy Faces Scrutiny Amid Anthropic Dispute and Official’s Financial Ties

The Pentagon’s relationship with artificial intelligence firm Anthropic has turn into increasingly fraught, marked by ethical disagreements over the leverage of AI for surveillance and autonomous weapons, and now complicated by questions surrounding potential conflicts of interest involving a key defense official. The dispute has escalated to the point of legal challenges, with a judge suggesting the Pentagon’s actions may be an attempt to “cripple” the AI company.

Ethical Clash Over AI Deployment

Defense Secretary Pete Hegseth directed the Department of War to designate Anthropic as a supply chain risk in February 2026, following stalled negotiations. The core of the disagreement centers on Anthropic’s refusal to allow its AI model, Claude, to be used for “mass domestic surveillance of Americans” and “fully autonomous weapons.” Anthropic maintains that current AI models are not reliable enough for autonomous weapons systems and that domestic surveillance raises fundamental rights concerns.

Potential Conflict of Interest

Adding another layer of complexity, reports indicate that Emil Michael, the Undersecretary of Defense for Research and Engineering and Chief Technology Officer, holds significant financial ties to Anthropic’s competitor, Perplexity. The Lever reported that Michael owns $2-$10 million in Perplexity stock and previously served on the company’s board. Even as Perplexity doesn’t have a direct contract with the Department of Defense, it has secured an agreement to deploy its AI search engine across federal agencies and is a contender to host government AI systems. Michael also has ties to Tools for Humanity, the company behind the eye-scanning verification technology, and its founder Sam Altman’s OpenAI, which is poised to potentially take over the Pentagon contract previously held by Anthropic.

Legal Challenges and Pentagon’s Continued Use of Anthropic’s Tech

Anthropic has filed lawsuits alleging retaliation from the Department of Defense. A judge overseeing one of these cases indicated the Pentagon’s designation of Anthropic as a supply chain risk appeared to be an attempt to harm the company. Despite the public dispute and accusations, reports suggest the Department of Defense has continued to utilize Anthropic’s Claude, even during the early stages of its recent actions in Iran.

Looking Ahead

The situation highlights the growing pains of integrating advanced AI technologies into national security frameworks. The ethical considerations surrounding AI deployment, coupled with potential conflicts of interest, demand careful scrutiny as the Pentagon navigates this evolving landscape. The outcome of the legal challenges and the resolution of the ethical concerns will likely set a precedent for future government-AI partnerships.