The Pentagon is developing alternatives to Anthropic, report says

Read the full articleThe Pentagon is developing alternatives to Anthropic, report says on TechCrunch

↗

What Happened

After their dramatic falling-out, it doesn't seem as though Anthropic and the Pentagon are getting back together.

Our Take

This is just sad. DoD relationship cratered because Anthropic said "we won't help with autonomous weapons." Now the Pentagon's funding alternatives because a vendor won't play ball. Classic government move: if someone won't compromise, fund competitors until someone does. Anthropic's probably right on the ethics. But they've just become less relevant to the biggest buyer in America. This accelerates the US/EU LLM split and probably helps open-source models that have no ethics to enforce.

What To Do

Watch what the Pentagon's "alternative" actually is—it'll probably be fine-tuned Llama or a defense contractor deal.

Builder's Brief

Who

AI companies pursuing US government and defense contracts

What changes

responsible use policies are now a material procurement risk factor; companies must explicitly address override and disable scenarios in defense contracts

When

months

Watch for

which AI vendor publicly wins the procurement contracts Anthropic lost

What Skeptics Say

The Pentagon pivoting away from Anthropic doesn't solve its underlying problem — it just selects for AI vendors without safety constraints, which is a worse long-term outcome for defense AI reliability.