Attackers can craft a malicious prompt that coerces the language model into executing arbitrary JavaScript in the context of the web page.
- http://seclists.org/fulldisclosure/2024/Aug/3
- http://www.openwall.com/lists/oss-security/2024/08/08/6
- https://korelogic.com/Resources/Advisories/KL-001-2024-005.txt
- https://github.com/fkie-cad/nvd-json-data-feeds