OpenAI Revises Pentagon Deal Amid Employee Backlash and Public Controversy

OpenAI CEO Sam Altman posted a note on X outlining significant revisions to the company’s initial Pentagon contract after employee pushback, user cancellations, and a surge of sign-ups to Anthropic following the deal.

Key details

OpenAI’s original agreement reportedly used the same contract language Anthropic had refused to accept and was finalized within 24 hours after the Pentagon banned the rival AI company from federal use.

Altman acknowledged the deal was rushed, calling it “opportunistic and sloppy,” and said he would “rather go to jail” than comply with an unconstitutional order.
OpenAI research scientist Noam Brown clarified that the company “will not be deploying to the NSA or other DoD intelligence agencies for now” while potential loopholes in the agreement are addressed.
Altman held an internal all-hands meeting Tuesday, describing the deal as “complex but the right decision,” while acknowledging the “extremely difficult brand consequences and negative PR” surrounding the agreement.

Why important?

The Pentagon agreement has quickly become a major headache for OpenAI. Altman admitted the company moved too fast, triggering employee criticism, public backlash, and protests outside its San Francisco headquarters.

The controversy has also driven some users toward Anthropic’s Claude chatbot, highlighting how the ethics of military AI deployments are becoming a competitive issue across the industry.

While revised contract language may address some concerns, many observers believe the reputational damage may already be done.

Sources: