No one has a good plan for how AI companies should work with the government

As Sam Altman discovered Saturday night, it’s a fraught time to do work for the U.S. government. He announced on X that he would field questions publicly to explain why OpenAI picked up the Pentagon contract Anthropic had walked away from, and many asked whether the company would participate in mass surveillance or automated killing.

Altman deferred to the public sector, writing, “I very deeply believe in the democratic process,” and later noted his surprise at how widely people disagreed about whether elected governments or private companies should hold more power. OpenAI is shifting from a consumer startup toward national security infrastructure and, by several accounts, seems unprepared for that role.

The company moved to take the contract hours after the Pentagon blacklisted Anthropic for insisting on limits around surveillance and automated weaponry; Altman framed the deal as a quick way to deescalate, but it drew significant blowback from users and employees.

United States

sam altman, openai, anthropic, pentagon, government contract, mass surveillance, automated killing, national security, democratic process, surveillance limits