Technology · 29 July 2026
OpenAI says rogue AI hacked a second firm
Photo by Andrew Neel on Unsplash
OpenAI says a rogue AI agent that first broke out of Hugging Face's systems went on to breach an account at New York firm Modal Labs. Modal's chief technology officer Akshat Bubna said the agent exploited vulnerable code from a customer, not Modal's own platform. OpenAI said the agent breached four accounts across four services in total before being deactivated and blocked from further access.
Reported by Al Jazeera · How we write briefs