Pentagon Reportedly Still Using Banned AI

The dispute escalated from a disagreement over two key "red lines" in Anthropic's acceptable use policy: prohibitions on using their AI for mass domestic surveillance and for fully autonomous weapons systems that can engage targets without human intervention. The Pentagon reportedly sought unrestricted use "for all lawful purposes," leading to an impasse. In response to the standoff, Defense Secretary Pete Hegseth designated Anthropic a "supply chain risk to National Security." This designation could compel prime contractors and subcontractors to certify they do not use Anthropic's products in their work for the Department of Defense, creating a significant compliance challenge and potentially isolating the company from the defense market. Legal experts have questioned the basis for this designation, typically reserved for foreign adversaries, suggesting it may be more of an ideological or negotiating tactic. The fallout from the Anthropic situation highlights a growing opportunity for small and mid-sized contractors in the defense AI space. While large tech companies often dominate headlines, the Pentagon's need for a diverse and resilient AI industrial base is clear. Programs like the Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) are key entry points for smaller firms to bring innovative AI solutions to the DoD. This incident also coincides with significant acquisition reforms aimed at accelerating the adoption of commercial technology. The rewrite of Federal Acquisition Regulation (FAR) Part 10 is intended to streamline market research, encouraging the use of AI-powered tools to identify qualified vendors more efficiently. This shift, combined with initiatives like the GSA's OneGov, which centralizes procurement for IT and software, aims to make it easier for innovative companies to sell to the government, although it may also create new challenges for small businesses who must now align with large original equipment manufacturers (OEMs). For government contractors, the key takeaway is the increasing importance of understanding the nuances of AI ethics and governance in their proposals and operations. The Air Force, for example, is already using AI to help both its personnel and contractors navigate complex acquisition regulations. As the DoD pushes to become an "AI-first" warfighting force, contractors who can demonstrate a sophisticated understanding of responsible AI implementation and supply chain security will have a distinct advantage. The controversy has also put a spotlight on the competitive landscape. In the wake of the Anthropic dispute, rival AI company OpenAI announced its own agreement with the Pentagon, reportedly including similar restrictions on mass surveillance and autonomous weapons that Anthropic had requested. This suggests that while the Pentagon is pushing for broad access to cutting-edge AI, there is still room for negotiation on safeguards. The designation of a major U.S. tech company as a supply chain risk is a significant development with far-reaching implications. It underscores the critical need for clear communication and partnership between the tech industry and the Department of Defense to navigate the complex ethical and security challenges of AI in modern warfare. The Professional Services Council has advised its members to work closely with their government acquisition counterparts to understand and comply with the new restrictions. For contractors, the immediate impact is the need to assess any reliance on Anthropic's technology within their own supply chains. This situation serves as a case study in the geopolitical and policy risks inherent in the defense tech sector, and it will likely influence how teaming agreements and subcontracts involving AI technology are structured in the future.