ChatGPT goes dark worldwide as OpenAI scrambles to fix login failures
A global outage starting late Wednesday locked users out of ChatGPT, Codex and a dozen OpenAI API endpoints.
Key points
- ChatGPT began failing worldwide at roughly 8 PM ET on Wednesday, 19 August, blocking logins, signups and chat loading.
- OpenAI's coding assistant Codex and about 12 endpoints on the OpenAI API were hit in the same incident.
- OpenAI acknowledged the problem on its status page within about 14 minutes and said it was working on a fix.
- Affected users saw endless loading spinners and "too many concurrent requests" errors when trying to send messages.
- This is a service outage, not a security breach: no data theft has been reported.
ChatGPT stopped working for people around the world on Wednesday night, and OpenAI has confirmed the problem.
The trouble started at about 8 PM Eastern Time on 19 August. Users in the US and Europe found they couldn't sign in, couldn't create new accounts, and couldn't open any old conversations. When people tried to load the site, the sidebar spun forever. Sending a message threw back a "too many concurrent requests" error, meaning the system is swamped and can't take more work.
The outage was first reported by BleepingComputer.
What actually broke?
OpenAI's login system appears to be the main casualty, and that failure cascades into almost everything else the company offers. Without a working sign-in, nobody can reach their chats, and developers can't call the tools that sit behind ChatGPT.
On its public status page, OpenAI said it had "identified" the issue and was "working on implementing a mitigation." That's company shorthand for: we know what's wrong and we're trying a fix.
The knock-on damage is wider than the chatbot. Codex, OpenAI's coding assistant, went down at the same time. So did around 12 endpoints on the OpenAI API, the plumbing that other companies use to build ChatGPT-style features into their own apps. It's not the first time: we covered a separate ChatGPT outage on 25 July that similarly left users in the US, Europe and elsewhere unable to reach the service.
| Detail | What we know |
|---|---|
| Start time | ~8 PM ET, Wednesday 19 August |
| Acknowledged by OpenAI | ~8:15 PM ET (about 14 minutes in) |
| Services affected | ChatGPT, Codex, OpenAI API (~12 endpoints) |
| Regions affected | Global, including US and Europe |
| Cause | Login and authentication failures |
Is this a hack?
No. Nothing in OpenAI's status updates or the early reporting points to an attack. This looks like a classic availability failure, most likely inside the authentication service that checks who you are when you log in. Outages of this shape happen to every large cloud service eventually.
That matters because a lot of people, understandably, assume the worst when a login page stops working. There's no indication that passwords or payment details have been exposed.
What should users do?
Wait it out. There's nothing a normal user can do from their side to make ChatGPT load faster. Refreshing repeatedly or trying to reset a password won't help, and hammering the login button contributes to those "too many concurrent requests" errors in the first place.
A couple of sensible habits for anyone who relies on ChatGPT for work: keep a local copy of anything important you've generated, because your chat history isn't a backup. Also be wary of any email claiming to be from OpenAI during an outage. Criminals often push fake "account recovery" messages when a big service is down, hoping panicked users hand over their passwords.
For developers whose products depend on the OpenAI API, this is a reminder that a single upstream provider is a single point of failure. Building in a fallback, even a graceful error message, tends to age well.
OpenAI hadn't published a root cause at the time of writing. We'll update this story as the company shares more.
