Home / Mar 31, 2026 / Story
0
#3 The Hacker News general March 30, 2026 at 18:05 UTC

OpenAI Patches ChatGPT Data Exfiltration Flaw and Codex GitHub Token Vulnerability

By [email protected] (The Hacker News)

AI Summary

Check Point researchers disclosed a vulnerability in OpenAI ChatGPT that allowed attackers to exfiltrate sensitive conversation data, uploaded files, and user messages through malicious prompts without user knowledge. OpenAI has patched the flaw along with a separate Codex vulnerability that exposed GitHub tokens, preventing covert data extraction channels.

Relevance score: 90.0/100

# More from March 31