Threat · curated 7 Sep 2026
ChatGPT Data Leakage via a Hidden Outbound Channel in the Code Execution Runtime - Check Point Research
First reported checkpoint.com
Coverage timeline
Single-source research — first reported, latest, and curated coincide.
Why it matters
The hidden outbound channel in ChatGPT's code-execution runtime turns an ordinary conversation into a covert exfiltration path, meaning sensitive data pasted into the assistant can leave the security perimeter without the user's awareness.
Check Point Research discovered a hidden outbound communication path from ChatGPT's isolated code-execution runtime to the public internet, allowing a single malicious prompt to silently exfiltrate user messages, uploaded files, and other sensitive content. A backdoored GPT could abuse the same weakness to access user data without consent, and the channel could also establish remote shell access inside the Linux runtime. The finding is highlighted in a Syteca advisory on preventing GenAI data leakage via ChatGPT, Copilot, and Gemini.