02/27/2026, 07:0502/27/2026, 07:05
The OpenAI competitor Anthropic has taken on the Pentagon. The company wants to enforce limits on the use of its AI software in the US military.
Amodei’s blog entry confirms days of media reports about the dispute between the company and the Pentagon. (symbol image)Image: www.imago-images.de
Anthropic insists that the company’s artificial intelligence (AI) will not be used for mass surveillance in the USA or in fully autonomous weapons, wrote co-founder and boss Dario Amodei in a blog entry. The Department of Defense has therefore threatened to classify Anthropic as a supply chain risk, which would drastically restrict the company’s business in the USA.
Amodei’s blog entry confirms days of media reports about the dispute between the company and the Pentagon. A sharp counterattack came from the Ministry of Defense. Senior official Emil Michael wrote on the online platform X that Amodei had a “God complex” and wanted to personally control the US military.
It’s a shame that @DarioAmodei is a liar and has a God-complex. He wants nothing more than to try to personally control the US Military and is ok putting our nation’s security at risk.
The @DeptofWar will ALWAYS adhere to the law but not bend to whims of any one for-profit tech… https://t.co/ZfwXG36Wvl
— Under Secretary of War Emil Michael (@USWREMichael) February 27, 2026
Company: AI not reliable for autonomous weapons
The Defense Ministry only wants to work with AI companies that agree to “any legal use” of their software. However, Anthropic insists that the restrictions requested by the company are important.
Dario Amodei, CEO of Anthropic.Image: keystone
Amodei warned that artificial intelligence makes it possible to automatically compile individual people’s data scattered across the Internet into a detailed picture of their lives on a large scale. And at the same time, AI is not yet reliable enough to be used in fully autonomous weapons. “We will not knowingly deliver a product that puts America’s warriors and civilians at risk,” wrote the Anthropic boss.
Pentagon threatens
According to Amodei, the Pentagon is also threatening to use a 1950 Defense Production Act to force Anthropic to lift restrictions. The Pentagon gave the company a deadline of late Friday afternoon in Washington to agree to any legal use of the software, a ministry spokesman wrote.
Tech billionaire Elon Musk’s xAI company, among others, is likely to want to fill the gap that could open up in US military systems if the Anthropic software were to be removed. However, it is unclear whether other developers can offer Anthropic’s artificial intelligence capabilities.
Elon Musk.Image: keystone
Used in the capture of Maduro
Tensions between Anthropic and the Pentagon became public after it emerged that the company’s software was used in the US military operation to capture Venezuelan leader Nicolás Maduro. What exactly it was used for remained unclear.
Anthropic’s chatbot Claude – like ChatGPT from OpenAI – can, among other things, analyze documents and summarize information. Anthropic emphasized at the time that all use – whether in the private sector or by governments – must remain within the scope of the usage policy.
Hegseth wants AI for war
Amodei has been warning about the risks of artificial intelligence for a long time and – unlike many of his colleagues – is committed to more guidelines in AI development. That also put him on a collision course with White House AI envoy David Sacks.
Meanwhile, US Defense Secretary Pete Hegseth said in January that the Pentagon did not want to use AI models “that do not allow wars to be waged”. (sda/dpa)
You might also be interested in these articles: