Introducing the OpenAI Safety Bug Bounty program

In three linesOpenAI launches a Safety Bug Bounty program to identify AI abuse and security risks including agentic vulnerabilities, prompt injection, and data exfiltration. Researchers receive rewards for reporting discovered flaws.

Summary generated by Claude — human-verified