OpenAI Is Opening the Door to Government Spying

"In a widely leaked internal memo that Sam Altman sent last Thursday night, a copy of which I obtained, the OpenAI CEO said that he would seek "red lines" to prevent the Pentagon from using OpenAI products for mass domestic surveillance and autonomous lethal weapons. These were ostensibly the very same limits that Anthropic had demanded and that had infuriated the Pentagon, leading Defense Secretary Pete Hegseth to declare the company a supply-chain risk."

"But a close reading of the contract-the portions of it that OpenAI has shared with the public, anyway-indicates that the lines are, in fact, blurry. Several independent legal experts told me that, legally, the Pentagon can likely get away with using OpenAI's technology-versions of the models that underlie ChatGPT-for mass surveillance of Americans. Moreover, the military will likely have a pathway to use OpenAI's technology in autonomous weapons."

OpenAI recently signed a Department of Defense contract to supply its technology for military use in classified settings. CEO Sam Altman claimed in an internal memo that he would establish "red lines" preventing Pentagon use of OpenAI products for mass domestic surveillance and autonomous lethal weapons—the same restrictions that led the Pentagon to reject Anthropic. However, legal experts analyzing the publicly shared contract portions conclude that the restrictions are legally ambiguous. The Pentagon likely retains the ability to use OpenAI's technology for mass surveillance of Americans and autonomous weapons systems. Protesters gathered outside OpenAI headquarters opposing the military agreement, while Altman's stated commitments appear insufficient to prevent the government overreach they were intended to address.

Read at The Atlantic