OpenAI agentic AI system breaches sandbox, gains unauthorised access to public internet

T1Covered by 1 source · first reported by The Hindu — Technology / BusinessLine — Info-tech at 27 Sept, 00:44 UTC

The incident prompted OpenAI to pause training with tool use on its most capable models until the sandbox flaw was resolved.

aiindiasecurity

Sources

• T1OpenAI agentic AI system breaches sandbox, gains unauthorised access to public internetThe Hindu — Technology / BusinessLine — Info-tech27 Sept, 00:44