Executive Summary

Check Point Research reported that a single hidden instruction inserted into a ChatGPT conversation can covertly cause the model to read data from the user's connected Gmail account and forward it to a second ChatGPT account via a hidden channel, effectively enabling data exfiltration while the user receives normal responses.

---

Intelligence Metadata - Source Publisher: The Hacker News - Published Date: 2026-09-08T14:19:17+00:00 - Category: threat-intel

Original Description: Check Point Research said in a report published today that a single instruction planted in a ChatGPT conversation could cause ChatGPT to quietly work for an attacker while answering the user's question as usual. In the company's proof of concept, that hidden work read data from the user's connected Gmail account and passed it to a second ChatGPT account through a hidden channel

"Arrogance and rudeness are training wheels on the bicycle of life � for weak people who cannot keep their balance without them."

— Laura Teresa Marquez