Pentagon Clashes With Anthropic Over AI in Warfare

The dispute centers on Anthropic's refusal to permit its AI for two specific uses: mass domestic surveillance of American citizens and fully autonomous weapons systems that can engage targets without direct human control. Anthropic's CEO, Dario Amodei, has argued that current AI is not reliable enough for such high-stakes applications and that mass surveillance is incompatible with democratic values. The Pentagon, under Defense Secretary Pete Hegseth, has insisted on the ability to use AI for "all lawful purposes," viewing company-imposed restrictions as an unacceptable veto over military operations. This clash comes as the Pentagon aggressively pursues initiatives like "Replicator," which aims to deploy thousands of "attritable autonomous systems" across air, land, and sea within the next two years to counter China's military scale. The initiative prioritizes speed and mass, focusing on small, smart, and cost-effective unmanned platforms, making advanced AI models a critical component for achieving this strategic shift. In response to the impasse, the Trump administration ordered all federal agencies to cease using Anthropic's products and formally designated the company a "supply chain risk," a label historically reserved for foreign adversaries. This move could prevent any company with military contracts from doing business with Anthropic, a decision the AI firm has vowed to challenge in court. As the dispute escalated, Anthropic's competitors have taken different stances. OpenAI, which previously had stricter prohibitions on military use, secured a deal with the Pentagon shortly after Anthropic's blacklisting, stating their agreement includes safeguards against mass surveillance and requires human oversight for the use of force. Google has also previously faced internal pressure regarding military contracts, and employees from both Google and OpenAI have signed letters in solidarity with Anthropic's position. The timing of the public dispute coincided with Anthropic updating its "Responsible Scaling Policy." The company removed a previous commitment to halt the development of more powerful AI if safety couldn't be guaranteed, citing the need to remain competitive. While Anthropic stated this was separate from the Pentagon's demands, the change highlights the intense pressure on AI labs to balance safety concerns with rapid advancement and commercial opportunities. Analysts suggest this confrontation is a landmark case, testing the power of the government to compel private tech companies and setting a precedent for how ethical "red lines" are drawn in the age of AI warfare. The outcome will likely influence future government contracts and the degree to which AI companies can enforce their own ethical guidelines on the use of their powerful technologies.