Pentagon and Anthropic Clash Over AI Access and Safeguards

A dispute between the U.S. Department of War and Anthropic, a leading artificial intelligence company, has escalated over access to its AI model, Claude, and the safeguards Anthropic has put in place. The conflict centers on the Pentagon’s desire for “unfettered access” to Claude for military applications, whereas Anthropic is hesitant to remove restrictions related to AI-controlled weapons and mass domestic surveillance.

Rising Tensions and a Friday Deadline

Defense Secretary Pete Hegseth issued a deadline to Anthropic CEO Dario Amodei, initially on Friday, February 26, 2026, to comply with demands to lift restrictions on Claude. Axios reported on this initial deadline. Failure to comply risked the termination of a $200 million contract and potential blacklisting of the company. Hegseth has since indicated a willingness to invoke the Defense Production Act to compel Anthropic’s cooperation if necessary. CNN details this threat, stating Hegseth would too label Anthropic a supply chain risk.

Anthropic’s Concerns and Prior Stance

Anthropic has expressed concerns about two specific use cases: the development of AI-controlled weapons and the implementation of mass domestic surveillance. The company believes current AI technology is not reliable enough for autonomous weapons systems and that there is a lack of legal and regulatory frameworks governing the use of AI in mass surveillance. Anthropic’s statement, released February 26, 2026, clarifies that these two use cases have never been included in their contracts with the Department of War and should remain excluded.

Department of War’s Extensive Use of Claude

Despite the current dispute, the Department of War already extensively utilizes Claude across various mission-critical applications. These include intelligence analysis, modeling and simulation, operational planning, and cyber operations. Anthropic was the first frontier AI company to deploy its models in the US government’s classified networks and at the National Laboratories, and has provided custom models for national security customers. According to Anthropic, they have proactively worked to deploy their models to the Department of War and the intelligence community.

Legislative Response and Calls for Negotiation

The escalating conflict has drawn attention from lawmakers. Top senators involved in defense policy have urged Secretary Hegseth and CEO Amodei to extend negotiations and find a resolution. POLITICO reports that senators are seeking a ceasefire in the dispute.

Anthropic’s Previous Actions to Protect National Security

Anthropic has previously taken steps to safeguard national security interests, even at a financial cost. The company has reportedly forgone revenue by cutting off access to Claude for firms linked to the Chinese Communist Party, disrupted CCP-sponsored cyberattacks targeting Claude, and advocated for strong export controls on chips to maintain a democratic advantage. As stated by Dario Amodei, these actions demonstrate a commitment to defending America’s lead in AI.

Looking Ahead

The outcome of this dispute will likely have significant implications for the future of AI development and deployment within the defense sector. It raises critical questions about the balance between national security needs and ethical considerations surrounding AI technology. Continued negotiation and a clear understanding of the limitations and risks associated with AI are crucial to ensuring its responsible use.