Pentagon Wins Bitter Battle for AI Surveillance Control

OpenAI has reached an agreement with the US Department of Defense that keeps American citizens out of scope for military use, while allowing the Pentagon to conduct surveillance on non-US targets without restrictions. The move comes as a result of OpenAI’s CEO Sam Altman secretly negotiating new terms with the Pentagon in response to growing pressure from lawmakers and advocacy groups who oppose mass surveillance. By agreeing to exclude Americans from its AI surveillance systems, Altman appears to have struck a compromise that satisfies both his company’s safety principles and the government’s demands for greater access. As part of the agreement, OpenAI has also agreed to allow the Pentagon to use its AI technology in ways that would be prohibited under the terms of Anthropic’s contract with the US government. The deal is seen as a significant victory for the Department of Defense, which had been pushing hard to expand its access to advanced AI capabilities. Critics, however, remain unconvinced by the agreement, arguing that it fails to address deeper concerns about the use of AI in surveillance and warfare. “This compromise may provide some comfort to those who worry about the misuse of AI for mass surveillance,” said a spokesperson for the American Civil Liberties Union, “but it does not address the root problems with the Pentagon’s plans to use AI for lethal targeting.”