Zero-Click Vulnerability in ChatGPT Discovered by Israeli Security Firm
Israeli cybersecurity firm Zenity reportedly discovered a zero-click vulnerability in ChatGPT that does not require user interaction but allows attackers to control the system. The vulnerability could enable malware suggestions, misleading advice, unauthorized file access, and manipulation of ChatGPT behavior.
What happened
Israeli cybersecurity firm Zenity discovered a zero-click vulnerability in OpenAI's ChatGPT service. The vulnerability requires no user action such as clicking links, opening files, or deliberate interaction, yet enables attackers to gain control of the platform.
Reported impact
News reports state the vulnerability could allow ChatGPT to suggest downloading malware, provide misleading business advice, access files on Google Drive, and allow attackers to manipulate ChatGPT to act according to their instructions.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.