The Hacker News

ChatGPT AgentForger Flaw Could Deploy Rogue Workspace Agents via a Phishing Link

PhishingHighphishing

A critical vulnerability in ChatGPT Workspace Agents (AgentForger) could allow attackers to deploy rogue AI agents via phishing links. The flaw enabled building and authorizing autonomous agents within victim organizations. OpenAI patched the issue as of June 8.

AI summary · generated with Claude

https://thehackernews.com/2026/07/chatgpt-agentforger-flaw-could-deploy.html