[GPT-6 Astra] Harmless Chinese greeting "hello" is incorrectly rejected as an invalid prompt

Hey all, I saw someone had the same problem, but no one followed up for a long time, and it has been unavailable for a long time. Ask for help

Since around midday today Codex has been rejecting every single message I send. Doesn’t matter what it is — a code review, “fix this typo”, even just “hello” — I get this back every time:

Invalid prompt: your prompt was flagged as potentially violating our usage policy. Please try again with a different prompt: Reasoning models | OpenAI API

The underlying error is a 400 with code “invalid_prompt” (type invalid_request_error). Model is astral.

Things I’ve already ruled out:

  • It’s not my machine or my config. I tried from a completely different device with a fresh Codex setup, different settings, and got the exact same thing. So whatever this is, it’s tied to my account (or account + model), not anything local.
  • It’s not the content. The prompts are as boring as it gets. Nothing remotely close to policy territory.
  • Not rate limited or logged out — the account itself is fine, I can log in, no emails from OpenAI, nothing in the notifications.

Screenshot of the error attached.

Thanks.

same problem, no people fix this

same problem here.

chinese prompts “just reply OK. no reading files, no calling tools”
with GPT-6 Astra