US Used Banned AI for Iran Strikes

The dispute between Anthropic and the White House stems from the company's refusal to remove safeguards against using its AI for mass domestic surveillance and fully autonomous weapons. Anthropic CEO Dario Amodei stated the company has "red lines" and will not move on them, arguing that current AI is not reliable enough for such applications and could increase civilian casualties. The Pentagon, however, insisted on the ability to use the technology for "all lawful purposes." This public clash escalated when the White House ordered all federal agencies to "immediately cease all use" of Anthropic's technology, with a six-month phase-out for the Department of Defense. Defense Secretary Pete Hegseth went further, designating Anthropic a "supply chain risk to national security," a label typically reserved for foreign adversaries, which bars defense contractors from working with the company. This move was described by Anthropic's CEO as "retaliatory and punitive." The timeline of these events is notable, as the ban was announced just hours before the U.S. military reportedly used Claude AI to help select targets and process intelligence for strikes in Iran. This highlights the deep integration of AI into military operations and the complexity of removing such tools once they are embedded. Prior to the ban, Anthropic's technology was widely used across the DoD and intelligence community for tasks like intelligence analysis, operational planning, and cyber operations. The termination of Anthropic's contract is a significant development for the OneGov platform, a GSA initiative designed to streamline AI procurement for federal agencies. Just months earlier, in August 2025, OneGov had announced a deal to provide all federal branches with access to Anthropic's "Claude for Government" for a nominal one-dollar fee for a year. This program was intended to accelerate AI adoption and position the U.S. as a global leader in government AI. This situation unfolds as the DoD continues to shape its AI ethics and governance frameworks. In 2020, the department adopted five core principles for the ethical use of AI: responsible, equitable, traceable, reliable, and governable. These principles are meant to guide all AI initiatives and ensure human accountability, particularly in high-stakes military environments. The current dispute with Anthropic brings these principles into sharp focus, questioning how they are applied when a vendor's ethical red lines conflict with military operational desires. For government contractors, this rapid shift in vendor relationships underscores the volatility of the defense tech landscape. It also coincides with major acquisition reform efforts, such as the "Revolutionary FAR Overhaul," aimed at simplifying procurement. Proposed changes include significantly raising the Simplified Acquisition Threshold, which could streamline processes for smaller contracts and make it easier for new tech companies to enter the defense market. The Department of Defense invests approximately $1.8 billion annually in early-stage technology development through the Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs. These programs are a key avenue for tech startups to partner with the DoD, offering funding for feasibility studies (Phase I) and prototype development (Phase II). Small businesses retain the intellectual property rights for technologies developed under these programs. The U.S. Air Force has been actively experimenting with AI for target recognition and decision-making acceleration. In a recent exercise, AI software developed under the Maven Smart System was used to ingest data, prioritize targets, and provide real-time recommendations to targeting teams in a simulated combat scenario. The goal is to create a more "resilient, data-driven, and automated kill chain" while reducing the cognitive load on human operators.