Nearly 100 OpenAI employees have signed an open letter supporting the same red lines on mass domestic surveillance and autonomous weapons that Anthropic defended in its failed negotiations with the Pentagon, The Atlantic has reported. The letter emerged just hours after OpenAI CEO Sam Altman announced his own deal with the Defense Department.
As reported by "Hvylya", The Atlantic notes that the timing creates an uncomfortable question for Altman. Earlier in the week, the OpenAI CEO had publicly stated that, like Anthropic, his company would also refuse to allow its models to be used in autonomous weapon systems. But as he made those statements, he was already negotiating a new deal with the Pentagon.
That deal was announced just hours after Anthropic's agreement fell apart. OpenAI released a statement describing the broad contours of the arrangement and highlighting that its AI would be deployed "only in the cloud" - the same compromise that Anthropic had quickly dismissed as insufficient. Anthropic's team had concluded that in modern military AI architectures, the line between cloud and edge systems is blurred beyond meaningful distinction.
Altman did not respond to The Atlantic's request for comment. The gap between his public statements of solidarity with Anthropic and the deal he ultimately signed raises a pointed question: what, if anything, changed between those two moments?
As The Atlantic's report puts it, if Altman finds himself face-to-face with the letter's signatories in the office on Monday, "he may have to explain why this idea that Anthropic quickly dismissed out of hand proved so compelling to him." The open letter suggests that a significant portion of OpenAI's own workforce shares Anthropic's assessment of where the ethical boundaries should lie - even if their CEO does not.
Also read: Anthropic Ban May Reshape AI Industry More Than Iran War, MIT Economist Warns