Several major media organizations obtained and reviewed excerpts from the transcript of a company-wide OpenAI town hall held Tuesday, during which CEO Sam Altman attempted to reassure and calm employees unsettled by the controversy surrounding the company’s closely scrutinized agreement with the United States Department of Defense, commonly known as the Pentagon.
In accounts of the meeting, Altman is portrayed as measured and conciliatory, acknowledging internal tension and external criticism. Reporting from The Wall Street Journal described him characterizing the episode as “painful” and expressing regret over the perception that OpenAI appeared “not united with the field,” a remark suggesting concern about divisions between the company and the broader AI research community. Coverage from CNBC added that Altman has faced vocal pushback from some staff members who questioned both the optics and implications of the defense-related partnership.
Yet one particularly notable section of The Wall Street Journal’s report indicates that OpenAI may be preparing to deepen, rather than retreat from, its engagement with military and security institutions. According to the transcript excerpts, Altman informed employees that the company is exploring a potential agreement to deploy its technology across classified networks operated by the North Atlantic Treaty Organization (NATO).
Authorization to operate on NATO’s secure and classified systems alone would represent a significant milestone. Just last month, Apple Inc. publicly announced that certain iPhone and iPad models had received approval for use in NATO classified environments—an unprecedented achievement for mainstream consumer hardware, underscoring the exclusivity and commercial value of such certifications.
OpenAI’s recently finalized Pentagon agreement, however, appears to function somewhat differently. Based on CNBC’s interpretation of the meeting transcript, the contract grants the Defense Department broad authority to utilize OpenAI’s artificial intelligence tools, while limiting OpenAI’s role in day-to-day operational decision-making. Meanwhile, The New York Times characterized the arrangement as a deal to supply advanced AI capabilities for the Defense Department’s classified systems, reinforcing the strategic and technical scope of the collaboration.
NATO had already telegraphed a shift toward expanded defense spending the previous year, with member nations pledging substantial budget increases in response to mounting geopolitical pressures. At the time, venture capitalist Dave Harden predicted that the surge in military investment could spark what he described as an “AI gold rush,” as technology firms compete for high-value government contracts.
OpenAI’s broader government-focused initiative dates back to June of last year, when it unveiled its “OpenAI for Government” offering and disclosed a framework agreement with the Pentagon. Through the Defense Department’s Chief Digital and Artificial Intelligence Office (CDAO), the company became eligible for up to $200 million in project awards tied to artificial intelligence research, development, and deployment.
Gizmodo said it contacted OpenAI seeking confirmation and additional details regarding the prospective NATO-related contract and indicated that it would update its reporting if and when the company responds.