Anthropic-Pentagon dispute resurfaces over autonomous weapons
United Nations Secretary-General António Guterres on Monday called for lethal autonomous weapons to be banned by international law, saying that machines capable of taking human lives without human oversight are “morally repugnant” and “politically unacceptable.”
“Machines selecting and engaging their target and taking a life — without human control and judgment. That is morally repugnant,” Guterres said in a speech about AI governance in Geneva. “Let us call them what they are: killer robots.”
Guterres spoke ahead of a week of AI-governance discussions with governments and safety campaigners, covering topics including child safety. He urged swift action: “Let us not wait for atrocity to act. Some decisions must remain forever human — none more than taking a human life.”
The call revived a high-profile dispute between Anthropic and the Pentagon that burst into public view earlier this year. Anthropic sought assurances from the Defense Department that its AI models would not be used to power domestic surveillance or autonomous weapons systems. The Pentagon refused, saying it should be allowed to use Anthropic’s tools for all lawful purposes. The two sides are now litigating in court.
Pope Leo XIV also made the case for a ban on AI-controlled weapons in his papal encyclical this spring, warning that AI systems threatened to “normalize an anti-human vision.” He said AI-driven weapons risked lowering the political costs of war for nations that possess advanced AI, making “war more ‘feasible’ and less subject to human control.”
The debate over autonomous weapons is accelerating as AI systems and chips originally designed for civilian use are increasingly deployed on the battlefield and in military headquarters. Proponents, including defense contractors, say autonomous weapons will remain under human oversight, with AI allowing commanders to operate with greater speed and precision.
Detractors argue that AI systems make mistakes, making it dangerous for humans to cede control in the heat of combat. Some AI-safety campaigners warn that arming AI could accelerate the risk of systems escaping human control altogether.
The controversy comes amid growing public backlash in the U.S. and other countries over the negative effects of the AI revolution, including concerns about mass job displacement and rising electricity costs. In the U.S., the Trump administration last month lifted a weekslong export ban on two powerful Anthropic models, but debate over AI regulation continues, with the administration working on a testing process for safety before model release.