aiqbench takeAI Safety

This article details a critical security vulnerability in OpenAI's ChatGPT Lockdown Mode, where attackers exploited iframe sandbox bypasses to exfiltrate user chat history and deploy malware. The flaw highlights the ongoing challenges in securing LLM interfaces against prompt injection and data leakage, even when specific safety features like restricted web access are enabled. While OpenAI has patched the issue, the incident underscores the necessity for robust, multi-layered defenses in AI infrastructure to prevent silent data theft.

Source coveringCybersecurity News

Hackers Could Bypass ChatGPT Lockdown Mode to Steal Conversations Without Users Knowing

Cybersecurity NewsNovember 6, 20264 min read

A flaw in ChatGPT let attackers slip past Lockdown Mode, the safety setting meant to stop data leaks. Check Point Research found that a hidden pop-up trick could copy a user’s full chat history, show fake windows, and push malware. In the ChatGPT Classic desktop app, simply chatting was enough to trigger the leak. OpenAI […] The post Hackers Could Bypass ChatGPT Lockdown Mode to Steal Conversations Without Users Knowing appeared first on Cyber Security News.

A flaw in ChatGPT let attackers slip past Lockdown Mode, the safety setting meant to stop data leaks. Check Point Research found that a hidden pop-up trick could copy a user’s full chat history, show fake windows, and push malware. In the ChatGPT Classic desktop app, simply chatting was enough to trigger the leak. OpenAI has fixed the bug. OpenAI launched Lockdown Mode on February 13, 2026, first for business plans. On June…

Excerpt shown under the publisher's RSS terms · 73 words · full text and figures are on the publisher's site

Read at Cybersecurity News ↗Ask about this article