Introducing the OpenAI Safety Bug Bounty program

OpenAI launches a Safety Bug Bounty program to identify AI security vulnerabilities and misuse risks. The initiative targets agentic vulnerabilities, prompt injection attacks, and data exfiltration threats. By crowdsourcing security research, OpenAI aims to strengthen AI safety defenses and enable responsible disclosure of critical issues before public exploitation.