OpenAI has emerged from a chaotic week as the US government’s preferred AI supplier, following the political exile of rival Anthropic over what the Trump administration characterized as excessive ethical restrictions. The episode has raised profound questions about how AI companies should navigate the competing pressures of corporate values and government demands.
Anthropic had negotiated with the Pentagon for months over the terms of AI deployment, insisting that its Claude system not be used for mass domestic surveillance or autonomous weapons that operate without human oversight. These conditions, which the company viewed as baseline ethical safeguards, were rejected by Pentagon officials who wanted unrestricted access.
President Trump’s decision to personally intervene — ordering all federal agencies to stop using Anthropic — transformed a commercial negotiation into a political confrontation. His characterization of Anthropic as “leftwing nut jobs” framed the ethical debate as a partisan battle, reinforcing the administration’s position that national security cannot be subordinated to corporate ethics policies.
Into this charged environment, Sam Altman stepped with a Pentagon deal in hand, claiming that OpenAI’s contract with the Defense Department includes the same restrictions Anthropic had fought for. He described these as long-held company principles, not new concessions, and suggested the Pentagon had agreed to apply them across the relationship.
The solidarity shown by AI workers across the industry — hundreds signing a joint letter supporting Anthropic — suggests that many employees view the deal with mixed feelings at best. The letter’s warning that the government was attempting to divide the industry appears to have been heeded by workers even if not by Altman himself.