Relations between the US Department of Defense (Pentagon) and Anthropic, the developer of the Claude AI model, are becoming increasingly tense after the company refused to expand the use of its technology in US military operations, according to US media reports.
Washington considers artificial intelligence capabilities to be a strategic tool that can be employed in several areas, including weapons development, information gathering, and use in combat zones. Therefore, the Department of Defense is pressuring major artificial intelligence laboratories—including Anthropic, OpenAI Google, and xAI—to allow their technologies to be used for all legal purposes, including sensitive ones.
However, Anthropic has indicated that it will not lift the restrictions it imposes on Claude, insisting that its models should not be used to develop autonomous weapons or for mass surveillance of individuals, which the Pentagon considers impractical in the context of its operational needs.
Anthropic gives Claude AI unprecedented capabilities
Artificial intelligence
Artificial intelligence Anthropic gives Claude AI unprecedented capabilities
US sources said that the Department of Defense is considering revisiting its partnership with the company and possibly canceling a contract worth an estimated $200 million if no solution is found that allows it greater freedom to use the Claude model in contexts required by its operations.
The controversy comes amid reports that the model itself has already been used in previous US military operations, including an operation to arrest former Venezuelan President Nicolás Maduro, which has sparked widespread criticism and questions about the limits of AI use in security and defense matters.
So far, Anthropic has not issued an official statement detailing its position on the ongoing negotiations, while the Pentagon continues to push for the expansion of AI technologies in Defense Department missions.