Hacktron researchers will work with OpenAI engineers to test connections with third party services for security flaws and receive broader access to OpenAI’s frontier models through its restricted Daybreak Red offensive security program. The partnership follows Hacktron’s earlier disclosures of high impact vulnerabilities through OpenAI’s bug bounty program, including a compromise reported on Sept 17.

The deal comes a day after OpenAI rolled out GPT-6 Cyber to select Daybreak Red participants, including security teams at Mozilla, Palo Alto Networks and Cisco. Built for vulnerability discovery and red teaming within authorized targets, GPT-6 Cyber identified 81 previously unknown high severity flaws in Firefox and developed working exploits and candidate patches for 948 local privilege escalation vulnerabilities in the Linux kernel during a single $1.21 million API run.