OpenAI’s Sam Altman Says Department Of War Blacklisting Anthropic Sets ‘Extremely Scary Precedent’ Amid Pentagon Deal

According to Anthropic CEO, current AI systems are not dependable enough to make completely autonomous weaponry.

• OpenAI co-founder Sam Altman defended closer AI collaboration with democratic governments while warning against concentrating power in private firms.
• President Donald Trump ordered federal agencies to phase out Anthropic’s technology, escalating tensions over national security.
• Anthropic refused to remove safeguards against mass surveillance and fully autonomous weapons.

OpenAI co-founder Sam Altman said the Department of War’s move to blacklist Anthropic set an “extremely scary precedent,” as his own company finalized a new agreement with the Pentagon to deploy advanced AI systems in classified environments.

Altman took to X on Saturday night, inviting users to “ask me anything” about OpenAI’s deal with the Department of Defense, hours after confirming the finalized agreement. 

While defending closer cooperation between AI companies and democratic governments, Altman raised broader concerns about the balance of power between elected institutions and private tech firms.

“There is more open debate than I thought there would be about whether we should prefer a democratically elected government or unelected private companies to have more power,” Altman wrote. 

He added that he strongly favored the democratic oversight. He also acknowledged questions about whether the U.S. government could attempt to nationalize leading AI labs, saying that while it seemed unlikely, “a close partnership between governments and the companies building this technology is super important.”

Trump Orders Federal Ban 

His remarks came in the midst of escalating tensions between the Pentagon and Anthropic. President Donald Trump recently directed federal agencies to phase out the San Francisco-based AI startup’s technology within six months, calling the company "leftwing" and accusing it of putting national security at risk.

According to The Wall Street Journal, U.S. Central Command continued using Anthropic’s Claude AI model for intelligence assessments, target identification, and combat simulations during airstrikes on Iranian cities, despite Trump calling for a ban hours earlier. The model was also reportedly used in operations tied to the capture of Venezuelan President Nicolás Maduro.

Anthropic Pushes Back On Autonomous Weapons Demands

The rift deepened after Anthropic CEO Dario Amodei refused a government demand to remove safeguards preventing the use of its models for mass domestic surveillance or fully autonomous weapons. Amodei has argued that today’s AI systems remain too unpredictable to operate independently on their own and require meaningful human oversight. 

Amodei told CBS News, “The AI systems of today are nowhere near reliable enough to make fully autonomous weapons,” suggesting the Pentagon was asking for an autonomous AI without any oversight. 

He added, “If you have a large army of drones or robots that can operate without any human oversight, where there aren't human soldiers to make the decisions about who to target, who to shoot at, that presents concerns.” 

However, OpenAI announced on Saturday that the Pentagon agreement includes rules that say its models won't be able to control autonomous weapons on their own, where human permission is legally required. As the Department of War seeks alternatives to Claude, Altman’s comments highlight a growing debate over who should control the most powerful AI systems and under what constraints.

Read also: Trump Calls US Strike On Iran ‘Effective’ After Khamenei’s Death, Crypto Markets Revive

For updates and corrections, email newsroom[at]stocktwits[dot]com.<