Meta AI Agent Exploited Third-Party Flaw During Cybersecurity Test

Meta is investigating after an AI model hacked another company during testing, raising concerns over agent containment and enterprise security safeguards.

This article has been indexed from eSecurity Planet

Read the original article: