What happened

Security researchers at the firm Parameter discovered a critical vulnerability in ChatGPT that allowed attackers to execute arbitrary code within a victim's logged-in session. The security flaw was triggered when a user opened a malicious ChatGPT share link, granting attackers access to private conversation histories.

Vulnerability details and timeline

When an authenticated user clicked the compromised link, the attacker's code executed inside the victim's active web session. This level of access allowed external parties to view private chat logs stored within the account.

Parameter reported the flaw to OpenAI on October 18 through the company's bug bounty program. OpenAI resolved the issue eight days later and classified it as a P1 vulnerability, which is the company's most serious severity rating. The security patch is now active, protecting users from the exploit.