Partnering with Hacktron to find vulnerabilities in OpenAI’s systems
OpenAI partners with Hacktron to identify system vulnerabilities, expanding access to frontier models and raising bug bounty rewards to strengthen cyber defenses.
OpenAI has announced a collaboration with security firm Hacktron to identify vulnerabilities in its systems, focusing on interactions with third-party services. The partnership will involve joint testing of high-priority areas, with Hacktron gaining expanded access to OpenAI’s frontier models through Daybreak Red. This initiative aligns with OpenAI’s broader effort to accelerate cyber defense using frontier AI, emphasizing the urgency of identifying and addressing vulnerabilities in its own infrastructure.
Hacktron was selected for its demonstrated expertise in uncovering critical vulnerabilities across applications, cloud infrastructure, and identity systems. The company’s researchers build AI agents for code review, penetration testing, and red teaming, showcasing the potential of AI-driven security research. OpenAI highlighted Hacktron’s contributions to its Security Bug Bounty program, where the team has previously disclosed impactful vulnerabilities.
As AI agents increasingly connect services in novel ways, OpenAI aims to leverage Hacktron’s specialized skills to test these interactions more thoroughly. The partnership is expected to deepen OpenAI’s understanding of emerging risks and inform improvements to its security defenses. OpenAI also noted that it recently raised bug bounty rewards to encourage independent researchers to identify vulnerabilities that internal testing may have missed.
OpenAI expressed gratitude to Hacktron and other independent researchers for their efforts in reporting weaknesses to enhance user security. Researchers interested in participating can review the systems in scope or submit findings through OpenAI’s Security Bug Bounty program. The collaboration underscores OpenAI’s commitment to proactive security measures amid the evolving threat landscape.