The Story: In a public note shared on X, OpenAI CEO Sam Altman explained updates to the company’s agreement with the U.S. Department of War, outlining new language intended to clarify civil liberties protections and operational boundaries.

Domestic Surveillance Restrictions
Altman stated that OpenAI amended its agreement to explicitly prohibit intentional domestic surveillance of U.S. persons and nationals. The updated language says the AI system cannot be used for deliberate tracking, monitoring, or surveillance of Americans, including through commercially acquired personal or identifiable information.

Limits on Intelligence Agency Use
According to the clarification, the Department of War also affirmed that OpenAI’s services will not currently be used by Department of War intelligence agencies such as the NSA. Any such usage would require a separate contract modification.

Democratic Oversight Emphasized
Altman wrote that decisions about how AI should be used in society must ultimately be made through democratic processes. He said OpenAI aims to participate in policy discussions while respecting constitutional limits and legal oversight.

Internal Communication and Criticism
Altman also acknowledged that the company may have communicated the agreement too quickly, describing the situation as complex and requiring clearer public explanation. Reporting by CNBC noted that during an internal all-hands meeting, Altman told employees that operational decisions about how the military ultimately uses AI systems would rest with government authorities.

Why It Matters: The episode highlights the growing tension between frontier AI development and national security collaboration. As AI systems become more capable, defining clear guardrails around surveillance, military use, and civil liberties is quickly becoming one of the most sensitive policy challenges facing the industry.