I completely understand that services can experience outages or degradation for many different reasons, and that this is normal for any large-scale platform.
That said, I would like to point out that, according to the OpenAI status page, the incident appears to have started only recently. However, based on user reports in the community and my own experience, the issue seems to have started several hours earlier.
I think it would be helpful if these situations were reflected a bit more accurately in the public incident timeline and uptime history. It would also be very helpful if service degradation like this could be detected proactively on OpenAI’s side and communicated to users in more detail.
For example, even a short update confirming that the incident has been detected, that the team is working on it, and providing an approximate ETA or next update time would give users much more context. That would make it easier for us to adjust our plans instead of being left unsure whether the problem is local, temporary, or already known.
I want to emphasize that I am not saying this aggressively or as a complaint. Personally, I use Codex to run parallel development work across different modules of a service. I am completely fine with pausing that work or changing my plans when there is an incident, but I would really appreciate clearer feedback from the team and at least an approximate ETA when possible.
More broadly, I think it is important that these incidents do not appear to be discovered only after the community starts reporting them. Ideally, it should feel like OpenAI has already detected the degradation, is actively working on it, and is keeping users informed.