Agents
155d agoOpenAI Launches Safety Bug Bounty
OpenAI’s new Safety Bug Bounty invites researchers worldwide to spot hidden flaws in its systems, from agentic glitches to prompt‑injection tricks. By exposing these risks early, the program aims to strengthen global AI security and trust, safeguarding tools like ChatGPT.
The Signal
The initiative signals a shift toward proactive safety governance in the rapidly evolving AI landscape.