Anthropic's Hiring a Chemical Weapons Expert — For Safety

Anthropic posted a role for a Policy Manager for Chemical Weapons and High-Yield Explosives — someone to shape how Claude handles prompts about bombs, dirty nukes, and chemical agents. Salary: $245K–$285K.

The hire will design guardrails and run rapid-response protocols when dangerous prompt patterns spike. OpenAI is running an almost identical search, offering up to $455K for a biosecurity and chemical risk specialist — a sign this is becoming standard practice at frontier labs.

The timing is charged: Anthropic is simultaneously fighting the US Department of War, which labeled the company a "supply chain risk" after demanding unrestricted Claude access for autonomous weapons systems. CEO Dario Amodei has drawn a hard line against deployments involving mass domestic surveillance or fully autonomous lethal systems.

Why it matters: AI labs are now treating weapons misuse as an engineering problem — and they're paying defense-grade salaries to solve it before regulators force their hand.