Sam Altman Announces Classified AI Deal With US Defense Department Amid Policy Shift on AI Firms
Washington: OpenAI chief executive Sam Altman announced that the company has finalised an agreement with the US Department of War to deploy its artificial intelligence models within the department’s classified network, marking a major expansion of AI use in national defence systems.
The announcement comes amid escalating tensions in the US AI sector, as President Donald Trump has ordered federal agencies to stop using rival AI firm Anthropic.
Agreement focuses on safety and safeguards
In a statement posted on social media, Altman said the agreement reflects a shared commitment to safety and responsible AI deployment. He noted that OpenAI’s core safety principles include prohibiting domestic mass surveillance and ensuring human responsibility in the use of force, including autonomous weapon systems.
Altman said the Department of War agreed to uphold these principles through its laws and policies, and that OpenAI would implement technical safeguards to ensure the AI systems function as intended.
The company will also deploy field support engineers to assist with implementation. For security purposes, the AI systems will operate exclusively on controlled cloud networks within the classified environment.
Altman added that OpenAI has encouraged the department to offer similar terms to other AI companies, emphasising the need for cooperation and reduced legal conflict across the industry.
Major partnership with Amazon announced
Separately, OpenAI and Amazon have entered a multi-year strategic partnership aimed at accelerating AI innovation for businesses, startups, and consumers worldwide.
Under the agreement, Amazon will invest $50 billion in OpenAI, including an initial $15 billion commitment, with an additional $35 billion planned upon meeting specific conditions.
The partnership will also include joint development of a Stateful Runtime Environment powered by OpenAI’s models. The system will be delivered through Amazon Bedrock and is designed to enable AI models to access computing power, memory, and identity resources more efficiently.
Trump administration bans Anthropic from federal use
The deal follows recent directives from President Trump ordering federal agencies to immediately stop using Anthropic’s AI services. The move has intensified tensions between the White House, the Pentagon, and leading AI companies.
The policy has divided Silicon Valley, with some technology leaders supporting the administration’s position, while others have expressed concern over increasing government intervention in AI development.
Altman acknowledged the broader tensions but said OpenAI remains committed to safe and responsible AI deployment while supporting national security needs.
Industry tensions continue
The latest developments highlight growing competition and political scrutiny surrounding AI companies as governments increasingly integrate artificial intelligence into defence, intelligence, and national infrastructure systems.
Analysts say the agreement signals a significant step in the use of advanced AI within classified military environments, while also underscoring the rapidly evolving relationship between government and major technology firms.
