kill switch why is the government fighting with claude?
11 snips
Mar 5, 2026 
Will Knight, senior WIRED writer who covers AI policy and industry trends, joins to unpack the Anthropic–government clash. He traces the DOD contract disputes, Anthropic’s refusal to enable mass surveillance or autonomous weapons, the unusual supply-chain-risk label, and how rivals and politics shaped the fight. Short, sharp takes on how AI and defense are colliding.
AI Snips
Chapters
Transcript
Episode notes
Anthropic Put Safety Carve Outs In Military Deals
- Anthropic prioritized safety and took contract carve-outs limiting mass domestic surveillance and fully autonomous weapons.
- That stance is unusual for defense suppliers and triggered intense pushback because the Pentagon expected unconstrained use of systems.
Anthropic Was First To Put Claude On Classified Pentagon Systems
- Anthropic became the first AI company to run a version of Claude on the Pentagon's classified systems after pursuing DoD business last summer.
- That shift surprised many because a few years earlier tech workers protested military partnerships.
Supply Chain Risk Label Is An Unusual Escalation
- Calling Anthropic a supply chain risk is extraordinary because that label is normally reserved for foreign adversaries or compromised hardware.
- Designation could bar DoD partners from using Anthropic despite Amazon and others being deeply integrated with the company.