OpenAI launches GPT-5.6 Bio bug bounty offering up to 47L
Technology
OpenAI is rolling out a Bio Bug Bounty Program for its new AI model, GPT-5.6, offering up to $50,000 (about ₹47 lakh) to researchers who can break through its biosafety features.
The goal is to spot and fix any vulnerabilities before the program becomes the company's primary biosafety testing after July 27, 2026.
OpenAI seeks cross interaction jailbreaks
"Jailbreaking" here means finding ways around the AI's built-in safety controls, basically exposing loopholes that could let users bypass restrictions.
OpenAI wants researchers to look for methods that work across different interactions, not just one-off tricks.
If you applied for the GPT-5.5 bounty last time, you're automatically eligible for this round too, making it easier to join in and help strengthen the system.