`stream disconnected before completion: This user's access to gpt-5.3-codex-premium-1p-codexswic-ev3 has been temporarily limited for potentially suspicious activity related to cybersecurity` false positive error

Resolved 💬 6 comments Opened Mar 3, 2026 by fullofcaffeine Closed Mar 3, 2026

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

All of the sudden I got a `stream disconnected before completion: This user's access to gpt-5.3-codex-premium-1p-codexswic-ev3 has been temporarily limited for potentially suspicious activity related to cybersecurity. Learn more about our
safety mitigations: https://platform.openai.com/docs/guides/safety-checks/cybersecurity" error message. Typing continue seems to solve the issue (see below).

I'm doing nothing directly related to cybersecurity nor "suspicious" in the project. Check the thread ID 019cb0d4-be62-71d3-a8a2-ab22dbdd983d (submitted using /feedback).

Codex seems to be back to normal and it still shows GPT5.3-codex-xhigh as the model. I'm assuming this was a false positive / fluke of that safety guard system, but thought I'd share and file a report because this does look concerning from a end-user's perspective, so some clarification from OpenAI developers would be nice.

View original on GitHub ↗

6 Comments

acSpock · 4 months ago

you an everyone else man

Howchie · 4 months ago

Yep can confirm I just received this error too, are we actually blocked from the right model??

019cb12c-a51d-7800-a59c-c0e6a45c2dab thread id

JD3Lasers · 4 months ago

Yep can confirm I just received this error too, are we actually blocked from the right model?? 019cb12c-a51d-7800-a59c-c0e6a45c2dab thread id

RUST_LOG='codex_api::sse::responses=trace' codex exec --skip-git-repo-check --sandbox read-only --model gpt-5.3-codex 'ping' 2>&1 | grep -m1 'SSE event: {"type":"response.created"' | sed 's/.*SSE event: //
' | jq -r '.response.model'

run that in the chat or cli and it will return the model you're on

waml · 4 months ago

> Yep can confirm I just received this error too, are we actually blocked from the right model?? > 019cb12c-a51d-7800-a59c-c0e6a45c2dab thread id RUST_LOG='codex_api::sse::responses=trace' codex exec --skip-git-repo-check --sandbox read-only --model gpt-5.3-codex 'ping' 2>&1 | grep -m1 'SSE event: {"type":"response.created"' | sed 's/.*SSE event: // ' | jq -r '.response.model' run that in the chat or cli and it will return the model you're on

it shows 5.3-codex

i think this is a different issue

etraut-openai contributor · 4 months ago

Thanks, we're looking into the cause and will report back when it's mitigated. Apologies for the disruption.

etraut-openai contributor · 4 months ago

The problem has been mitigated. We're resetting rate limits to compensate everyone for the inconvenience.