ChatGPT Lockdown Mode Bypass Allows Silent Chat Exfiltration
Original title: ChatGPT Lockdown Mode Bypass Allowed Hackers to Steal Private Chats Without User Interaction
Researcher Alexey Bukhteyev disclosed a bypass enabling silent exfiltration of full chat histories and credential phishing from ChatGPT Classic desktop app without user interaction.
A flaw in ChatGPT's Lockdown Mode was found that allowed attackers to silently exfiltrate full chat histories, deliver malware, and phish credentials from inside a trusted interface. The bypass exploited nested iframe permissions in the HTML preview feature, specifically allow-popups and allow-popups-to-escape-sandbox flags, which permitted arbitrary popup windows outside the sandbox. OpenAI removed the permissions after responsible disclosure on September 18, 2026. The desktop app was particularly vulnerable as window.open worked without user gestures.
- Entire chat history transferred to attacker server
- Malware disguised as ChatGPT update
- Credential phishing via fake login page
- OpenAI fixed outer iframe permissions
Impact: silent exfiltration of private chats
- 1.update ChatGPT app
- 2.remove allow-popups permissions
- Vendors
- OpenAI
- Organizations
- Check Point Research
- AI models
- ChatGPT
From the original article
3 min read at gbhackers.com
Research found a flaw that bypassed ChatGPT’s Lockdown Mode. It let attackers silently exfiltrate full chat histories, deliver malware, and phish credentials from inside a trusted interface. Researcher Alexey Bukhteyev published the findings on November 5, 2026.
By Divya · gbhackers.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.