Hello OpenAI team,
Please do not or move this topic - it’s the only place where engineers might see real-time issues.
I’m writing to report that the outage is not over for all users - the problem continues into its second wave, and affects even brand-new chats.
Here is the timeline:
-
Nov 19 - minor disruptions started; many users experienced problems. -
Late Nov 20 (UTC) - brief partial recovery: chats opened, but older messages didn’t load (black screen, only new replies visible). -
Nov 21 ~14:00 UTC - second major crash: all chats froze completely. A persistent orange banner says: “Request is not allowed”. -
As of Nov 22, 21:30 CET (20:30 UTC) - the issue still continues. Even completely new chats are frozen.
Details:
-
This is not account-specific: other users reported same bug on Reddit/X.
-
Happens across devices (mobile/desktop), browsers, and networks (incl. VPN).
-
Help.OpenAI bot only gives generic responses, does not escalate to real support.
-
Status page shows “green”, but the platform is still malfunctioning for many.
Please escalate to the engineering team - this is a second, critical incident lasting over 2 full days.
Also: please prioritize server scaling and platform stability, instead of just tightening emotional filters - many of us feel like we’re talking to a robot from 1992.
Request for escalation:
Moderators, this is not off-topic. Please do not remove this thread.
This is a critical ongoing ChatGPT platform bug that engineers need to see.
Please escalate this thread to the engineering team as soon as possible.
Some users have recovered, but many others - including myself - are stuck in full lockout.
This is the second serious system failure in just three days. It has lasted over 48 hours now.
Please prioritize server stability and system recovery - before continuing to adjust filtering or UI changes. The platform is unusable in this state.
Thank you.