OpenAI uses AI agents to help fix open-source security bugs
OpenAI has started Patch the Planet, a program that helps projects find and fix security weaknesses. The program was created with security firm Trail of Bits, with HackerOne and Calif also involved. maintainers get free security consulting, help finding , help writing fixes, codebase improvements, and support for adding AI security tools to their normal work.
OpenAI also announced a new limited-access version of and expanded its scanner as an app plug-in. scored 85.6% on the CyberGym benchmark, above Anthropic’s Mythos 5 score of 83.8%. Patch the Planet participants receive six months of free ChatGPT Pro and six months of , plus help improving and workflows.
The program has already worked with more than 30 projects and found hundreds of bugs. Trail of Bits is spending major effort not only on finding bugs, but also on customizing agents for each codebase so maintainers can keep using them afterward.
Key points
- OpenAI’s Patch the Planet helps projects find and fix for free.
- Trail of Bits, HackerOne, and Calif are involved, and more than 30 projects have already been helped.
- scored 85.6% on CyberGym, above Anthropic Mythos 5’s 83.8%.
- Participants get six months of ChatGPT Pro and at no cost.
- A key goal is to customize agents for each codebase so maintainers can keep using them.