Science & Technology · 4 Beginner · October 6, 2026
Pentagon stops using AI from Anthropic
Photo: Emmanuel Codden via Pexels
Listen
Vocabulary
militaryˈmɪlɪterinoun
the armed forces of a country, including the army, navy, and air force
The US military used AI tools made by a company called Anthropic.
Pentagonˈpɛntəɡɑnnoun
the main building and office of the US Department of Defense
In February, the Pentagon said Anthropic was a problem.
safety featuresˈseɪfti ˈfɪtʃərznoun phrase
special protections built into a product to prevent harm or danger
Safety features protect people from harm.
weaponsˈwɛpənznoun
tools or objects made to hurt or kill people in war or fighting
The company worried about weapons and spying.
removerɪˈmuvverb
to take something away or get rid of something
The Pentagon asked Anthropic to remove safety features from its AI.
deep insideˈdip ɪnˈsaɪdadverbial phrase
far within or completely embedded in something
Anthropic's AI was deep inside Pentagon computer systems.
Article
The US military used AI tools made by a company called Anthropic. These tools helped the military sort through information and do research. In February, the Pentagon said Anthropic was a problem. Now the Pentagon does not use Anthropic anymore.
Why did the Pentagon stop?
The Pentagon asked Anthropic to remove safety features from its AI. Safety features protect people from harm. Anthropic said no. The company worried about weapons and spying. So the Pentagon chose other AI companies instead.
The change was slow
Anthropic's AI was deep inside Pentagon computer systems. It was hard to remove quickly. The Pentagon used it for months after deciding to stop. Now the military uses different AI tools from other companies.
Discussion Questions
Why did the Pentagon first use Anthropic's AI tools?
What safety feature did the Pentagon want to remove?
Why did Anthropic say no to the Pentagon?
Why was it hard to remove Anthropic's AI quickly?
Which other AI companies does the Pentagon use now?