A Shared ChatGPT Link Could Run Attacker Code in Your Session and Read Private Chats

A vulnerability in ChatGPT allowed a malicious share link to execute attacker code in the session of anyone who opened it, potentially exposing private conversations. Researchers at Parameter discovered the flaw and reported it to OpenAI through their bug bounty program. The issue was reported on October 18 and fixed by OpenAI eight days later. OpenAI categorized the flaw as P1, its most severe rating.