Key Takeaways
- Sam Altman acknowledged the Pentagon partnership announcement appeared rushed and poorly executed
- Contract revisions explicitly prohibit using OpenAI’s technology for domestic surveillance of American citizens
- The Defense Department verified intelligence agencies like the NSA cannot access OpenAI’s systems
- The partnership emerged shortly after Trump prohibited federal entities from deploying Anthropic’s AI solutions
- Altman advocated publicly for extending identical contract opportunities to Anthropic
OpenAI Restructures Defense Agreement Following Public Response, CEO Acknowledges Rushed Timeline
Sam Altman, OpenAI’s chief executive, has taken responsibility for the company’s handling of its Defense Department partnership announcement. Through what he characterized as an internal memo shared on X, he acknowledged the company should have taken more time before making the agreement public.
“We were genuinely trying to de-escalate things and avoid a much worse outcome, but I think it just looked opportunistic and sloppy,” Altman wrote.
The partnership went public last Friday, arriving within hours of President Donald Trump’s directive halting federal use of Anthropic’s AI platforms. The announcement also preceded U.S. military operations against Iran by mere hours.
The timing generated substantial public criticism. Numerous users reportedly removed ChatGPT from their devices and migrated to Anthropic’s Claude application in response.
OpenAI has begun collaborating with Pentagon officials to restructure the agreement. These modifications will embed the company’s ethical principles directly into the binding contract language.
A critical provision now specifies that “the AI system shall not be intentionally used for domestic surveillance of U.S. persons and nationals.” Defense officials confirmed that intelligence organizations including the NSA will have no access to OpenAI’s platforms.
Providing services to such agencies would necessitate a completely separate contractual arrangement, Altman explained.
Background: The Anthropic Negotiations Breakdown
This development follows unsuccessful negotiations between Anthropic and military officials. Anthropic had requested assurances preventing its technology from enabling domestic surveillance operations or autonomous weapons systems lacking human control.
Defense Secretary Pete Hegseth announced Friday that Anthropic would receive a supply-chain threat classification following the failed discussions. Government representatives had spent months expressing concerns about Anthropic’s emphasis on AI safety protocols.
The disagreement became widely known after reports revealed that Anthropic’s Claude AI had supported U.S. military operations during a January mission to apprehend Venezuelan president Nicolás Maduro. Anthropic remained silent about this deployment at the time.
Anthropic had actually pioneered AI deployment on the Defense Department’s secure classified infrastructure through an agreement established the previous year.
CEO Advocates for Competitor’s Fair Treatment
Altman’s statement also addressed the consequences facing Anthropic. He revealed weekend conversations with government officials where he challenged the supply-chain designation.
“I reiterated that Anthropic should not be designated as a supply chain risk, and that we hope the Department of Defense offers them the same terms we’ve agreed to,” he wrote.
Anthropic emerged in 2021 when several OpenAI team members departed due to philosophical differences regarding the organization’s strategic direction.
The company has built its identity around prioritizing AI safety. Pentagon representatives have yet to publicly address Altman’s proposal for equivalent contractual terms.
