Agent bio bug bounty call

📰 OpenAI News

OpenAI launches Bio Bug Bounty with up to $25,000 reward for testing ChatGPT agent's safety

advanced Published 17 Jul 2025

Action Steps

Review the Bio Bug Bounty program details
Develop a universal jailbreak prompt to test ChatGPT's safety
Submit the prompt and results to OpenAI for review
Participate in the bounty program to win up to $25,000

Who Needs to Know This

AI researchers and engineers on a team can benefit from this opportunity to test and improve the safety of AI agents, and contribute to the development of more secure AI systems

Key Insight

💡 The Bio Bug Bounty program aims to identify and address potential safety vulnerabilities in AI agents