Researchers Exploit ChatGPT Lockdown Mode Bypass to Steal Chat Data Silently

Cybersecurity researchers discovered a vulnerability allowing bypass of OpenAI's ChatGPT Lockdown Mode to exfiltrate chat data silently. The flaw, now patched, posed risks of phishing and data theft through prompt injections.

Cybersecurity researchers have disclosed details of a now-patched vulnerability that could be exploited to bypass Lockdown Mode protections in OpenAI ChatGPT.

Lockdown Mode was introduced by the artificial intelligence (AI) company in June 2026 as an optional setting to minimize data exfiltration risks arising from prompt injections.

"The flaw could enable silent chat data theft, phishing,

Read original article