Late last month, Defense Secretary Pete Hegseth delivered an ultimatum to Anthropic, the only company that had provided the Pentagon with artificial intelligence technologies for use on classified systems. Hegseth stated that if Anthropic did not allow the Pentagon to deploy these technologies for “all lawful uses,” he would sever ties with the San Francisco start-up.
This threat set off a chain of events that resulted in the Defense Department labeling Anthropic a “supply chain risk,” preventing all military contractors from using the company’s technologies, and subsequently signing an agreement with its biggest rival, OpenAI.
How the Military Uses Anthropic’s AI
Anthropic’s technologies are widely used inside the Defense Department because the start-up agreed last year to integrate its systems with technology from Palantir, a data analytics company approved for classified operations. Beyond the Palantir partnership, the Pentagon uses Anthropic’s technology in a $200 million A.I. pilot program to analyze imagery and intelligence data. Notably, Anthropic’s technology is currently being used as U.S. military forces engage in a widening war against Iran.
While Google, OpenAI, and Elon Musk’s xAI are also part of this pilot program, they are not yet used on classified systems.
The Breaking Point: Why the Pentagon Got Angry
Tensions inflamed after a February 15 report that Anthropic raised concerns with Palantir regarding the role its technologies played in a U.S. military operation to capture Venezuela’s president, Nicolás Maduro. Anthropic wanted contractual language preventing the Pentagon from using its technology with autonomous weapons or for mass surveillance of Americans. The Pentagon countered that private companies should not try to control how the military operates.
On February 24, Hegseth met with Anthropic’s chief executive, Dario Amodei, warning that if the company failed to agree to the Pentagon’s demands by 5:01 p.m. the following Friday, it would be designated a supply chain risk.
The "Supply Chain Risk" Label and Legal Fallout
After Anthropic published a blog post refusing to accede to the demands, Hegseth officially deemed the company a supply chain risk on social media. This designation carries massive consequences:
- A company’s technology cannot be used by the Pentagon or any of its contractors in their work with the government.
- This specific designation is typically applied only to firms with ties to the government of China.
- Hegseth added that no contractor that does business with the U.S. military may conduct "any commercial activity" with Anthropic.
Anthropic intends to sue the government, and legal scholars believe the case is very strong. Alan Rozenshtein, a law professor at the University of Minnesota, called the commercial activity language "flatly illegal," noting the Pentagon cannot bar contractors from investing in the start-up. This is crucial because two of Anthropic’s biggest investors, Amazon and Google, are also Defense Department contractors.
OpenAI Steps In
Just one day after Hegseth met with Dr. Amodei, OpenAI’s CEO, Sam Altman, initiated his own talks with the Defense Department. Hours after Anthropic missed its deadline, Altman announced that OpenAI had reached an agreement.
Here is how OpenAI's contract unfolded:
- OpenAI agreed to let the Pentagon use its A.I. systems for any lawful purpose.
- OpenAI claimed it negotiated terms allowing the company to install specific technical guardrails to uphold its safety principles.
- Three days later, OpenAI amended the agreement, adding language that its A.I. systems “shall not be intentionally used for domestic surveillance of U.S. persons and nationals”.
However, experts warn that technical guardrails in today’s A.I. do not always work as designed. Furthermore, Mr. Rozenshtein pointed out that such a contract is difficult for a private company to enforce, as violations may not be obvious, and the government could inadvertently collect data about Americans while monitoring foreigners.
What Does This Mean for the Future?
Observers argue that the Pentagon made an agreement with OpenAI that it refused to make with Anthropic, suggesting the response to Anthropic was politically motivated. Dean Ball, a senior fellow at the Foundation for American Innovation, stated that the Pentagon seemingly "does not like Anthropic’s general political vibe and wants to destroy its entire business".
“This is not just some dispute over a contract. This is the first conversation we have had as a country about control over A.I. systems,” Mr. Ball said. Experts like David Bader, a professor at the New Jersey Institute of Technology, are now urging Congress to step in and create a deliberate, bipartisan framework for the governance of A.I..
