OpenAI has recently shed more light on the specifics of its collaboration with the Pentagon. CEO Sam Altman acknowledged that the progression of this agreement seemed "rushed" and "did not project an optimal image to the external audience." Earlier, talks between Anthropic and the Pentagon fell through, subsequently leading to Trump directing federal agencies to cease utilizing its technology, and the Secretary of Defense categorizing it as a "supply chain risk." Amidst this context, OpenAI declared a pact with the Pentagon to implement its models within classified settings. Both Anthropic and OpenAI have imposed restrictions on the utilization of their technologies. OpenAI detailed its approach in a blog post, explicitly pinpointing three prohibited zones and affirming that it would safeguard these boundaries through a "comprehensive, multi-tiered strategy." Nevertheless, Mike Masnick from Techdirt contends that the agreement "fully accommodates domestic surveillance." In reaction, Katrina Mulligan, OpenAI's leader in national security collaboration, underscored the significance of the deployment architecture. When queried about the deal, Altman conceded that the progress was too hasty and sparked criticism from the outside world. He mentioned that the deal was expedited with the intention of easing tensions, and if the assessment was accurate, it would aid in resolving the issue; otherwise, OpenAI might face a negative reputation.