TaskGym

Open bounty · Global

AI Safety Red Teamer — Adversarial Testing for Frontier AI Models

Skill areas

Adversarial prompt designAI red teamingJailbreak identificationRisk domain evaluation (cyber, biosecurity, misinformation, fraud)Vulnerability documentationAI safety and policy analysisCritical and analytical thinking

The full role brief is one step away.

Sign up (or sign in) to see the full description, the software involved, and the complete payout ladder — then apply in about 2 minutes.