“Tonight, we reached an agreement with the Department of War to deploy our models in their classified network,” CEO Sam Altman of OpenAI said Friday, signaling a major expansion of AI use in US military systems via the company’s agreement with the US Department of War to deploy its AI models on classified cloud networks.
According to Altman’s article on X, the DoW demonstrated a profound respect for safety and a desire to collaborate to achieve the best possible outcome in all of our interactions.
The deal is made in the midst of growing hostilities between rival AI company Anthropic and US President Donald Trump’s administration about the use of AI in the military.
Trump announced on Friday that he had directed all government departments to stop working with the artificial intelligence lab Anthropic right once. The Defense Department and other agencies that presently use the company’s products would be given a six month grace period.
I am giving EVERY Federal Agency in the US Government the order to STOP using Anthropic’s technology RIGHT AWAY. We will never do business with them again since we don’t need or want it.” Trump stated in a Truth Social post.
Pete Hegseth, the head of the Pentagon, announced that he was giving the defense department instructions to declare Anthropic a supply chain risk to national security.
Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic, Hegseth said in a statement regarding X.
In describing the terms of the agreement, Altman stated that OpenAI would put in place stringent measures to guarantee that its systems function precisely within predetermined bounds.
“The main goals of our mission are AI safety and broad benefit sharing. Prohibitions on domestic mass surveillance and human responsibility for the use of force, including autonomous weapon systems, are two of our most crucial safety tenets.
Altman wrote on X, The DoW supports these principles, incorporates them into our agreement, and reflects them in law and policy.
As requested by the DoW, we will additionally develop technical measures to guarantee that our models operate as intended. We’ll use FDEs to support our models, but exclusively on cloud networks to guarantee their security,” he continued.
We believe that everyone should be able to accept these terms, hence we are requesting that the DoW extend them to all AI businesses. We have made it clear that we are eager for the situation to de-escalate away from governmental and legal actions and toward amicable accords,” he said.
We remain committed to serving all of humanity as best we can, Altman continued. The world is a complex, chaotic, and occasionally hazardous place.
In a memo sent on Thursday, Altman informed staff that OpenAI and Anthropic had similar “red lines.” In his post on Friday, he stated that the DoD accepted its limitations.
Prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems, are two of our most important safety principles, Altman wrote. The DoW upholds these values, incorporates them into our agreement, and reflects them in law and policy.
Although government officials have been criticizing Anthropic for months for purportedly being too concerned with AI safety, it is unclear why the DoD chose to accommodate OpenAI and not Anthropic.
Also Read: Toyota Fortuner Next-Gen Model Seen on Test Runs – What to Expect
In addition to deploying staff to assist with our models and to ensure their safety, Altman stated that OpenAI will construct technical safeguards to ensure its models behave as they should.
In a statement released on Friday, Anthropic expressed its deep sadness over the Pentagon’s designation of the business as a supply chain risk. It stated that it plans to contest the designation in court.
