Pentagon Pressures Anthropic for Military AI Access

Defense Secretary Pete Hegseth has issued a deadline to Anthropic CEO Dario Amodei, demanding that the company allow unrestricted military access to its artificial intelligence technology by Friday or risk losing its government contract. This ultimatum follows a meeting where Hegseth emphasized the importance of integrating AI into military operations, particularly as Anthropic remains the only AI company not supplying its technology to the U.S. military's internal network. The Pentagon has warned that failure to comply could lead to designating Anthropic as a supply chain risk or invoking the Defense Production Act to facilitate military use of its products without the company's consent.

WRAL

During the meeting, Amodei reiterated his ethical concerns regarding the use of AI in military contexts, specifically opposing fully autonomous military targeting operations and domestic surveillance of U.S. citizens. He has previously expressed apprehension about the potential for AI to enable mass surveillance and the dangers posed by autonomous armed drones. Hegseth's vision for military AI emphasizes operating without ideological constraints, stating that the Pentagon's AI will not be "woke." This reflects a broader debate over the role of AI in national security and the ethical implications of its deployment in high-stakes situations.

WRAL

Anthropic, which positions itself as a safety-minded AI company, has faced pressure to align with the Pentagon's policies, unlike its peers such as Google and Elon Musk's xAI, which have already begun integrating their technologies into military applications. Owen Daniels, an analyst at Georgetown University, noted that Anthropic's bargaining power is limited, and its reluctance to comply with military demands could jeopardize its influence in the Department of Defense's AI initiatives. The company has previously aligned with the Biden administration, advocating for third-party scrutiny of its AI systems to mitigate national security risks.

WRAL

Anthropic's advocacy for stricter AI regulations has historically put it at odds with the Trump administration, which has accused the company of fear-mongering regarding AI risks. David Sacks, a top AI adviser under Trump, criticized Anthropic's regulatory strategies, suggesting they were based on exaggerated fears. Despite these tensions, Anthropic has attempted to maintain a bipartisan approach by appointing former Trump administration official Chris Liddell to its board.

WRAL

The ongoing discussions between the Pentagon and Anthropic highlight the urgent need for greater oversight of AI technologies, particularly as their military applications expand. Experts, including Amos Toh from the Brennan Center for Liberty and National Security, argue that the law is lagging behind technological advancements, emphasizing that the Department of Defense should not have unrestricted authority in deploying AI systems, especially for surveillance purposes.

WRAL

• “A powerful AI looking across billions of conversations from millions of people could gauge public sentiment, detect pockets of disloyalty forming, and stamp them out before they grow,” — Dario Amodei, CEO of Anthropic

WRAL

• “the use of AI in military contexts is already a reality and it is not going away,” — Owen Daniels, Georgetown University

WRAL

• “The law is not keeping up with how quickly the technology is evolving,” — Amos Toh, Brennan Center for Liberty and National Security

WRAL