OpenAI has introduced a Safety Bug Bounty program to proactively identify and address potential AI abuse and safety vulnerabilities. The program specifically targets issues like agentic vulnerabilities, prompt injection, and data exfiltration risks.
Source: OpenAI