WASHINGTON, Feb. 23 (GeokHub) — U.S. Defense Secretary Pete Hegseth has summoned Dario Amodei, chief executive of artificial intelligence company Anthropic, to the Pentagon for high-stakes talks regarding the military’s use of the company’s Claude artificial intelligence system, according to a report by Axios.
The meeting, expected to take place Tuesday, signals rising tensions between the U.S. Department of Defense and major AI developers over how advanced AI systems should be deployed within classified military environments.
Pentagon Push for Fewer AI Restrictions
Earlier this month, Reuters reported that the Pentagon has been pressing leading AI firms — including OpenAI and Anthropic — to make their models available on classified government networks with fewer built-in usage restrictions.
Anthropic’s flagship AI system, Claude, includes guardrails designed to prevent harmful or unethical use. Defense officials, however, are seeking broader operational flexibility, particularly for intelligence analysis and strategic applications.
According to Axios, discussions between the Pentagon and Anthropic are nearing a breaking point, with defense officials indicating that patience is running thin. One senior official reportedly described the upcoming meeting as not merely introductory, suggesting urgency and potential consequences if no agreement is reached.
Anthropic has maintained that it is engaging in “productive conversations, in good faith,” though it has not publicly indicated any shift in its stance regarding safeguards on military applications.
Growing Tension Between AI Ethics and National Security
The dispute highlights a broader debate unfolding across Washington and Silicon Valley: how to balance national security needs with ethical AI development principles.
The U.S. government has increasingly emphasized AI as a strategic priority, particularly amid global competition in advanced technologies. Military officials argue that limiting AI capabilities could hinder operational effectiveness. Meanwhile, AI developers have sought to preserve safety protocols to prevent misuse or unintended consequences.
The potential collapse of talks could influence how the Pentagon partners with private AI firms in the future, especially as defense agencies accelerate adoption of generative AI tools for classified and mission-critical operations.
Reuters noted it could not independently verify Axios’ report at the time of publication. The Pentagon, the White House, and Anthropic did not immediately respond to requests for comment.
