By Hadas Gold, Haley Britzky, NCS
(NCS) — Defense Secretary Pete Hegseth gave Anthropic CEO Dario Amodei a Friday deadline to adjust to calls for to peel again safeguards on its AI mannequin or danger dropping a Pentagon contract.
He additionally threatened to put the AI firm on what might quantity to a authorities blacklist.
At difficulty is the guardrails Anthropic positioned on its AI mannequin Claude. The Pentagon, which has a $200 million contract with Anthropic, desires the corporate to raise its restrictions for the army to have the opportunity to use the mannequin for “all lawful use,” in accordance to two sources conversant in the discussions.
But Anthropic has considerations over two points that it isn’t keen to drop, the supply stated: AI-controlled weapons and mass home surveillance of American residents. According to one supply acquainted, Anthropic believes AI is just not dependable sufficient to function weapons, and there are not any legal guidelines or rules but that cowl how AI might be utilized in mass surveillance.
A supply conversant in the Tuesday assembly says the Pentagon plans to terminate Anthropic’s contract by Friday if the corporate doesn’t agree to its phrases. A Pentagon official instructed NCS the corporate has till 5:01pm on Friday to “get on board or not.” And if it doesn’t, Hegseth will guarantee “the Defense Production Act is invoked on Anthropic, compelling them to be used by the Pentagon regardless of if they want to or not.” Hegseth may even label Anthropic a provide chain danger, the official stated.
The DPA is a law that offers the federal government the power to affect companies within the curiosity of nationwide protection, lately invoked by the Trump administration in the course of the pandemic. The provide chain danger designation would prohibit corporations with army contracts from utilizing Anthropic’s merchandise in any of their army work. It might deal a main blow to the AI agency at a time when it’s attempting to increase its attain within the enterprise area, contemplating many massive firms have army contracts. The designation is often reserved for corporations seen as extensions of overseas adversaries like Russia or China.
NCS has requested the Pentagon for clarification on how Anthropic might be each compelled beneath the DPA whereas additionally being labeled a provide chain danger.
During the assembly, the tone was cordial and respectful and there have been no raised voices, the supply acquainted stated, including that Hegseth praised Anthropic’s merchandise and stated they need to work with them.
But Amodei reiterated Anthropic’s redlines on autonomous weapons and mass surveillance, and a supply acquainted stated Anthropic has no plans to budge and cling to the Pentagon’s calls for.
The Pentagon official instructed NCS the difficulty has “nothing to do with mass surveillance and autonomous weapons being used” and that the “Pentagon has always followed the law.”
“You can’t lead tactical ops by exception,” the official stated. “Legality is the Pentagon’s responsibility as the end user.”
An Anthropic spokesperson described the assembly to NCS as a “good-faith” dialog about utilization of the corporate’s know-how.
“Anthropic CEO Dario Amodei met with Secretary Hegseth at the Pentagon this morning. During the conversation, Dario expressed appreciation for the Department’s work and thanked the Secretary for his service,” Anthropic stated in a assertion to NCS. “We continued good-faith conversations about our usage policy to ensure Anthropic can continue to support the government’s national security mission in line with what our models can reliably and responsibly do.”
The negotiations have been ongoing for a couple months, the supply stated, however in current weeks reports started surfacing in regards to the tensions between the 2 sides.
Then final week, Axios reported Hegseth was shut to slicing the Pentagon’s contract with Anthropic and designating the corporate a “supply chain risk.”
“Anthropic is committed to using frontier AI in support of US national security,” the Anthropic spokesperson stated on Tuesday. “That’s why we were the first frontier AI company to put our models on classified networks and the first to provide customized models for national security customers.”
Anthropic’s combat with the Pentagon might create a gap for its rivals. The Pentagon official confirmed Elon Musk’s xAI firm is “on board with being in a classified setting,” whereas others are “close.”
Anthropic has lengthy positioned itself because the AI firm most involved with AI security. Its founders have been all former OpenAI workers who left the corporate over disagreements in regards to the ChatGPT maker’s path, strategy to security and tempo of AI growth. Anthropic additionally lately introduced it is giving $20 million to a political group campaigning for extra regulation of AI.
NCS’s Kaanita Iyer contributed to this report.
The-NCS-Wire
™ & © 2026 Cable News Network, Inc., a Warner Bros. Discovery Company. All rights reserved.