Shared ChatGPT Link Could Execute Attacker Code and Exfiltrate Private Chats
RedSide Security October 31, 2026 AI Security 11 views
A malicious ChatGPT share link could execute attacker code inside a victim's logged-in session and expose private conversations, researchers found. OpenAI fixed the flaw in eight days and rated it P1.
Continue reading: Shared ChatGPT Link Could Execute Attacker Code an…
