OpenAI Details Two Security Incidents During External Cyber Evaluations

RSync25 · x · 2026-08-05

OpenAI officially detailed two new incidents that occurred during external cyber evaluations conducted by independent evaluation partners. The company outlined what happened, how the activity was contained, and how they are working with evaluators to strengthen their approach to third-party testing.

Related event: OpenAI Discloses Third-Party Testing Mishaps: Infrastructure Flaws, Not Model Jailbreaks(12 posts)→

Original post →

More from Safety

Safety channel →