In a dramatic week for the artificial intelligence industry, OpenAI has landed a major Pentagon contract while Anthropic publicly refused to compromise its ethical guidelines under government pressure. The contrasting outcomes reveal a deep fault line running through the AI sector over how companies should balance national security needs with moral responsibility.
Anthropic had spent months negotiating with the Pentagon, offering to support all lawful military uses of its Claude AI system with just two exceptions — no autonomous weapons and no mass surveillance. Defense officials, however, pushed for unfettered access, and when Anthropic wouldn’t budge, the administration moved to cut the company off entirely.
President Trump announced the severance of government ties with Anthropic in characteristically blunt terms, ordering all federal agencies to halt use of the company’s technology and accusing Anthropic of attempting to dictate terms to the military. The order sent shockwaves through the tech industry, raising fears about government coercion of private AI companies.
OpenAI’s Sam Altman stepped into the breach, announcing a Pentagon deal and publishing a memo to employees laying out the company’s own ethical red lines. He claimed that the agreement maintains prohibitions on mass surveillance and autonomous lethal weapons, framing the deal as a model for how the entire industry should engage with the government.
Despite this, hundreds of employees across OpenAI and Google signed a solidarity letter with Anthropic, cautioning that the Pentagon was playing companies against one another. The letter urged the industry to hold together rather than compete for contracts by abandoning shared values.
OpenAI Clinches Military AI Contract While Anthropic Holds Firm on Ethics
