OpenAI's Sam Altman admits ‘rushed’ deal with Defense Department after backlash

Sam Altman, the CEO of OpenAI, acknowledged on Monday that the company acted too quickly in finalizing its recent agreement with the U.S. Department of Defense. This statement follows the announcement of the deal, which came just hours after the White House instructed federal agencies to halt the use of competing AI firm Anthropic's technologies, coinciding with imminent military actions against Iran. In a message posted on social media platform X, Altman indicated that adjustments would be made to the contract, specifically to ensure that the AI system would not be employed for domestic surveillance of U.S. individuals and nationals. He also mentioned that the Defense Department confirmed that OpenAI's technologies would not be utilized by intelligence entities like the NSA. Altman expressed concerns regarding the readiness of the technology, stating, "There are many things the technology just isn't ready for, and many areas we don't yet understand the tradeoffs required for safety." He emphasized that OpenAI would collaborate with the Pentagon to establish necessary technical safeguards. Admitting to a misstep, Altman remarked, "We shouldn't have rushed" to finalize the deal. He explained that the aim was to reduce tensions and prevent a more severe outcome, but acknowledged that the haste may have appeared "opportunistic and sloppy." This admission follows a heated conflict between Anthropic and the government over the safety protocols surrounding its Claude AI systems. Defense Secretary Pete Hegseth identified the company as a potential supply-chain threat. Anthropic had requested assurances that its technologies would not be used for domestic surveillance or the development of autonomous weapons without human oversight. The controversy intensified after it was revealed that Anthropic's Claude had been deployed by the U.S. military during a raid to apprehend Venezuelan President Nicolás Maduro earlier this year, a situation that the company did not publicly contest. OpenAI's agreement with the Pentagon was reached shortly after talks between Anthropic and the Defense Department fell apart, sparking a public outcry that led many users to shift from ChatGPT to Claude in app stores. In his social media post, Altman reiterated his belief that Anthropic should not face designation as a supply chain risk, hoping that the Department of Defense would extend similar terms to them as those agreed upon with OpenAI.