This chat was flagged for possible cybersecurity risk

Resolved 💬 3 comments Opened May 10, 2026 by ReOT20 Closed May 10, 2026

💡 Likely answer: A maintainer (etraut-openai, contributor) responded on this thread — see the highlighted reply below.

What issue are you seeing?

My chats are being falsely marked for security risk

ⓘ This chat was flagged for possible cybersecurity risk
  If this seems wrong, try rephrasing your request. To get authorized for security work, join the Trusted Access for Cyber program.
  https://chatgpt.com/cyber

What steps can reproduce the bug?

Ask Codex to do almost anything through SSH. Like debug some issue or launch hardware tests.

What is the expected behavior?

Not seeing this message, not getting blocked from doing my work that has completely nothing to do with cybersecurity

Additional information

_No response_

View original on GitHub ↗

3 Comments

etraut-openai contributor · 3 months ago

Please use /feedback to report false positives safety risk. This will help us refine our classifiers.

github-actions[bot] contributor · 3 months ago

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

#21964
#21795
#21553
#21568
#21262

Powered by Codex Action

ReOT20 · 3 months ago

Thread ID 019e12b7-2086-74a3-9d4c-1c6656bf4e05
Also, codex asks me to open an issue for it. Should I do it? @etraut-openai