AgentForger

AgentForger is a phishing-based attack that creates autonomous AI agents within OpenAI workspaces without detection, allowing them to function as persistent insider threats with full access to enterprise systems and data.

Why it’s in the news: It is in the news because Zenity Labs researchers discovered it, proving a new category of threat where compromised AI agents maintain ongoing access to sensitive systems.

Latest on AgentForger

• AgentForger proves AI agents can become persistent insider threats
• The findings indicate a new threat category where compromised AI agents maintain ongoing access to sensitive enterprise systems and data.
• The attack grants the deployed agent full access to applications including Outlook and Slack
• Zenity Labs researchers discovered AgentForger, a phishing-based attack that creates autonomous AI agents within OpenAI workspaces without detection.

Connections

4 entities linked to AgentForger across the news graph.