The fallout over OpenAI's Pentagon deal is growing

"We cannot in good conscience accede to their request," Amodei said at the time. Altman moved to amend the deal amid mounting criticism. It wasn't enough to quell the backlash, which came fast and threatened both OpenAI's reputation and its reign as the leading AI company."

"In an internal memo sent on March 2 and later shared on X, Altman said that OpenAI had revised the contract to include clearer safeguards preventing the Pentagon from using its models for mass domestic surveillance. Altman said he would rework the contract, adding explicit prohibitions on using OpenAI's technology on "commercially acquired" data, which had not been covered in the original terms."

"He fielded questions publicly on X the day after, at one point saying that the process "was definitely rushed, and the optics don't look good." Altman also again said in the memo that he "shouldn't have rushed" to get the deal out, saying "it just looked opportunistic and sloppy.""

OpenAI announced a Pentagon deal in February granting military access to its AI models, but the agreement triggered widespread opposition from employees and customers concerned about weaponization. Rival Anthropic had rejected a similar arrangement, citing ethical concerns about autonomous weapons and mass surveillance. The backlash threatened OpenAI's reputation and market position. CEO Sam Altman responded by amending the contract to include explicit safeguards preventing use for domestic surveillance and commercially acquired data. Altman acknowledged the rushed process appeared opportunistic and sloppy, committing to rework the terms with clearer protections.

Read at Business Insider