Papers & Research

Parameter security research: public ChatGPT shared links can execute arbitrary code in a victim's logged-in session

Original title:We Took over ChatGPT Accounts

AI summary

The Parameter team disclosed an XSS attack chain against ChatGPT: an attacker publishes a public shared conversation containing interactive charts and HTML Preview, exploiting MathLive keyboard layout messages that do not verify their origin, innerHTML injection into a sandboxed iframe and YouTube oEmbed JSONP to bypass CSP, causing JavaScript to run as chatgpt.com in the victim's logged-in session.

Key Takeaways

  • The Parameter team disclosed an XSS attack chain against ChatGPT: an attacker publishes a public…
  • innerHTML injection into a sandboxed iframe and YouTube oEmbed JSONP to bypass CSP,causing…
  • com in the victim's logged-in session.

Quick Facts

Published
Updated
Source
Hacker News:AI 热帖
Category
Papers & Research
Topic
OpenAI / ChatGPTSafety & Alignment
AI score
79
Read the original Read the full article at the original source
Why it matters

作者第一手复现完整攻击链并给出时间线与修复过程,读者可以借此理解沙箱隔离为何会在跨组件组合下失效。

Topics

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: Parameter security research: public ChatGPT shared links can execute arbitrary code in a victim's logged-in session. October 28, 2026. https://new.nat6.net/en/news/zdwho66dczydnc8nq5fs3rrx9/

View all →