Nvidia CEO: Pentagon-Anthropic clash is ‘not the end of the world’

Nvidia CEO Jensen Huang downplayed the dispute between the US Department of Defense and AI firm Anthropic, saying it is “not the end of the world.”

The disagreement centers on Anthropic’s request for assurances that its Claude AI models will not be used for autonomous weapons or mass surveillance of Americans.

The Department of Defense wants access for all lawful use cases.

Huang told CNBC that the Pentagon has the right to use AI tools to serve its interests, and Anthropic can decide how its products are marketed and used.

Negotiations have stalled over these specific use cases.

Founded in 2021 by former OpenAI researchers, Anthropic won a US$200 million contract with the DoD in 2025.

Huang noted Anthropic is not the only AI company, and the government has other options if no agreement is reached.

🔗 Source: CNBC

🧠 Food for thought

Implications, context, and why it matters.

Jensen Huang downplays a high-stakes government ultimatum

The Pentagon set Anthropic a Friday deadline to accept its terms, not a minor dispute 1.
If Anthropic refuses, the Defense Department could cancel its US$200 million contract. It could also invoke the Defense Production Act (a US law that can compel companies to prioritize and support government needs in emergencies) to force access to its technology 1.
The Pentagon also warned it may label Anthropic a “supply chain risk,” a tag usually aimed at foreign adversaries 2.
That label would block firms seeking military work from using Anthropic products, and experts call the tactic unusual when aimed at a US company 3.

Anthropic as a holdout amid shifting military AI terms

Nvidia CEO Jensen Huang said the government has “other options,” since several competitors are moving toward the Pentagon’s “all lawful use” terms.
Reports say OpenAI and xAI accepted “any/all lawful use” language. Google has also been reported as renegotiating to match the Pentagon memo 3.
Anthropic now stands more alone, as rivals accept broader Department of Defense conditions. That puts its ethical limits up against commercial pressure 4.
Anthropic still has near-term leverage because Claude is the only frontier AI model running on the Pentagon’s classified networks. Other models are not yet cleared for that setting 3.