US artificial intelligence company Anthropic said on Thursday it would not bow to pressure from the Pentagon to allow unrestricted military use of its technology, fearing its technology could be used for mass surveillance or fully autonomous weapons.
What did Antropic say?
“Using these systems for mass domestic surveillance is inconsistent with democratic values,” CEO Dario Amodei said, adding that AI systems are not yet reliable enough to eventually power deadly weapons without human control.
“We do not knowingly provide products that endanger U.S. warfighters or civilians,” he said.
Anthropic, along with OpenAI, Google and Elon Musk’s Grok chatbot, was commissioned by the Department of Defense in 2025 in a contract worth $200 million (€170 million) to supply AI models for various military applications.
After speaking with Amodei earlier this week, Defense Secretary Pete Hegseth gave Anthropic an ultimatum: Like other companies, it must release its artificial intelligence technology for “classified” use by Friday or risk losing government contracts.
Military officials warned they could go further and designate the company as a supply chain risk or invoke a Cold War-era law known as the Defense Production Act, giving the military broader authority to use its products.
“These threats do not change our position. We cannot in good conscience comply with their demands,” Amodei said, calling the threats “inherently contradictory” because they label Anthropic’s systems as both a “national security risk” and “essential to national security.”
“Antropic understands that military decisions are made by the Department of the Army, not by private companies,” he said, referring to the Trump administration’s Pentagon title. “However, we believe that in limited cases, AI could undermine rather than protect democratic values.”
Department of Defense: “Not interested in using AI for mass surveillance”
Pentagon spokesman Sean Parnell said on social media Thursday that the military “has no interest in using AI for mass surveillance of American citizens, nor does it want to use AI to develop autonomous weapons that operate without human involvement.”
But officials also acknowledged communications between Anthropic and the Pentagon over intercontinental ballistic missiles, highlighting the sensitive nature of the topic at the heart of the conflict.
Parnell said opening up the use of the technology would prevent Anthropic from “jeopardizing critical military operations” and warned that “we will not let any company dictate the terms of how we make operational decisions.”
Meanwhile, Sen. Mark Warner (Virginia), the ranking Democrat on the Senate Intelligence Committee, said he was “very disturbed” by reports that the Pentagon was “trying to bully some of America’s leading companies.”
“This is further evidence that the Department of Defense is willing to completely ignore AI governance,” he said in a statement, “further emphasizing the need for Congress to enact strong and binding AI governance mechanisms from a national security perspective.”
Editor: Wesley Dockery
