Anthropic's hiring a chemical weapons expert — to stop Claude

Anthropic is recruiting a Policy Manager for Chemical Weapons & High-Yield Explosives — not to build anything, but to shape how Claude refuses to help with it. The New York-based role pays up to $285k and requires 5+ years in chemical weapons or explosives defense, plus knowledge of radiological dispersal devices.

The hire will design and monitor the guardrails that govern how AI models respond to weapons-adjacent prompts, and run rapid-response when escalations are detected. OpenAI is running a nearly identical search, with a bio/chem risk role offering up to $455k.

Timing is notable: Anthropic is currently in a standoff with the US Department of War after being designated a "supply chain risk" — a label that lets the government block contracts. The company has mounted a legal challenge. These hires signal that both labs are trying to get ahead of the hardest misuse scenarios before regulators force their hand.

Why it matters: When AI labs start competing to hire weapons experts, the "catastrophic misuse" threat model has moved from theoretical to operational.