Anthropic investors move to defuse Pentagon clash over AI safeguards

SAN FRANCISCO: Investors in US artificial intelligence firm Anthropic are working to contain fallout from the company’s dispute with the Pentagon over battlefield use of its AI technology, fearing the standoff could damage the startup’s fast-growing business, sources familiar with the matter said.

The dispute centres on the military’s demand that AI companies remove certain safeguards and adopt an “all-lawful use” clause governing battlefield applications. Anthropic has refused to lift restrictions preventing its AI system, Claude, from being used for autonomous weapons or mass surveillance.

Chief Executive Dario Amodei has held discussions in recent days with major investors and partners, including Amazon CEO Andy Jassy, as well as venture capital firms Lightspeed and Iconiq, sources said. Some investors have also reached out to contacts within the Trump administration in an effort to ease tensions.

The Pentagon, which has been renamed the Department of War under the Trump administration, and Anthropic have been locked in talks for months. Investors are particularly concerned about the possibility of Anthropic being designated a “supply-chain risk,” a move that could require all U.S. government contractors to stop using its technology.

Defense Secretary Pete Hegseth has said such a designation would bar contractors from using Anthropic’s AI across their operations. The company has pushed back, arguing the department lacks statutory authority to block its technology outside defence contracts and has said it would challenge any such decision in court.

While some investors support Anthropic’s stance on AI safeguards, others privately expressed frustration that Amodei may have antagonised officials instead of pursuing a more diplomatic approach.

At the same time, Anthropic’s products are seeing strong demand. Its chatbot Claude recently became the most-downloaded free application on Apple’s App Store, surpassing OpenAI’s ChatGPT. The company’s revenue run rate is estimated at around $19 billion, up from $14 billion weeks earlier, according to a source.

However, several U.S. government agencies have begun terminating their use of Anthropic’s systems, with the State Department reportedly shifting to rival OpenAI, following President Donald Trump’s directive to phase out Anthropic’s AI within six months.

Anthropic has raised tens of billions of dollars from investors, with enterprise sales accounting for roughly 80 per cent of its revenue. Market observers say the outcome of the Pentagon dispute could significantly impact future funding rounds and any potential initial public offering.

Talks between Anthropic and the Pentagon are continuing, sources said.