Skip to content
EventUpdating

ChatGPT shared link vulnerability could read private chats, now fixed

3 reports3 reporting sourcesplus 2 outlets covering itLatest report

Start here

AI overview

Researchers at security firm Parameter say ChatGPT shared links had a cross-frame messaging vulnerability: a malicious link could execute attacker-supplied JavaScript in the session of a logged-in opener, enumerate recent conversations, read private chats that were never shared, and exfiltrate data to an external server. Initially a single click on Preview was required; later attack versions triggered automatically when the link was opened. The researchers reported it through OpenAI's bug bounty program on October 18; OpenAI reproduced it, raised the severity from P2 to P1, and completed the fix in the frontends of two versions on October 26 (eight days after the report), with no user action required. Parameter said it found no real-world exploitation.

Generated by AI from reports · Updated 15 hr ago

Progress timeline

Nov 3
Nov 2
  1. Profesional Review
    ChatGPT previously had a vulnerability that allowed reading users' private conversations, which has now been fixed

    Security company Parameter released a report stating that between October 18 and 26 there was an attack that read users' private conversations through ChatGPT shared links, and OpenAI has fixed the vulnerability. The attack link only needed to contain an HTML preview and a chart: when ChatGPT loaded the chart using the MathLive library, its math formula keyboard accepted messages from any other window, inserting unconfirmed HTML, forming an unsandboxed frame that the browser treated as part of ChatGPT and that could even execute JavaScript. Attackers could thereby obtain the full chat list, start new conversations, and send content to an external server, and the report states they could also act in the victim's name. OpenAI fixed the issue directly after setting a $5000 bounty; users do not need to make changes themselves, and can check shared links in the data controls in settings.

Nov 1
Oct 31
  1. CSA Labs ResearchSelected
    One click into someone else's ChatGPT: a cross-frame messaging vulnerability

    Researcher Parameter disclosed on October 28, 2026 that a publicly shared ChatGPT conversation link could execute attacker-supplied JavaScript in the session of whoever opened it while logged in, enumerate their recent conversations, read private chats that were never shared, and exfiltrate data to an external server. After reproducing it, OpenAI raised the severity from P2 to P1 and completed the fix on the frontend of two versions on October 26 (eight days after the report), with no user action required; Parameter said no real-world exploitation was found. Initially a single click on Preview was required, after which it was changed to open automatically, so that merely opening the link triggers it, and the payload executes after about 4.5 seconds. The researcher said the same script context could perform any action on the account on behalf of the victim, but this is an inference based on the permissions obtained, and the PoC deliberately limited the exfiltrated content.

  2. The Hacker NewsSelected
    A malicious ChatGPT shared link can run attacker code in a logged-in session and read private chats

    Researchers at the security company Parameter demonstrated that a malicious ChatGPT shared link can run attacker code in a victim's logged-in session and read private chats; the researchers reported it on October 18 through OpenAI's bug bounty program, and OpenAI fixed it eight days later and rated it the highest severity, P1. The victim only needs to open the link and click Preview once (in a later attack version, opening the link triggers it automatically), and the attack code runs with the same permissions as the ChatGPT site, can call logged-in APIs to list recent conversations, open private chats that were never shared, and send the contents to a server controlled by the researchers. The vulnerability stems from three overlapping design issues, among which a listener left on the main page by the MathLive math library does not verify the message origin and allows raw HTML to be carried, enabling code inside the Preview sandbox to create a frame that inherits the chatgpt.com identity, and then execute code via a CSP-allowed YouTube script.

Related discussion

Total 60 items

Attention trend

New sources per day

52 in 4 days·most 36 (Oct 31)·today 7

  • Oct 31: 36 new sources (3 outlets, 33 accounts)
  • Nov 1: 5 new sources (0 outlets, 5 accounts)
  • Nov 2: 4 new sources (1 outlets, 3 accounts)
  • Nov 3: 7 new sources (1 outlets, 6 accounts)

Hourly attention

Current attention 53·Peak in the comparable period 75 (Oct 31 23:00)·Change over the comparable past 24 hours +2%

020406080Oct 3115:00Nov 116:00Nov 216:00Nov 317:00