Skip to main content

Claude / Anthropic outage on September 22, 2026, lasting about 1 hour

Claude had a bigger problem starting just before 1 AM UTC on September 22, 2026. It lasted about an hour. Here is what happened.

By NoCrash Team Outage Severity Bigger problem Official source https://stspg.io/90s4l9yywzbk

Live status

No active incident for Claude / Anthropic right now.

See current Claude / Anthropic status →

Just before 1 AM UTC on September 22, 2026, Claude reported a bigger problem. It lasted about an hour and was resolved by 02:12 UTC the same morning. Anthropic has since marked it as resolved.

Who this kind of outage hits

If you use Claude inside an n8n workflow, a content pipeline, a support tool, or anything that runs quietly in the background, you probably did not know this was happening while it happened. That is the normal shape of this kind of trouble. No alarm goes off. No red screen appears. The work just stops moving. The first signal is usually a customer asking why their summary never arrived, or a colleague noticing that a queue has not moved since midnight. By the time you hear that, the outage may already be over, and you are left piecing together what broke and when.

Why this is especially rough without a technical background

A developer can pull logs and see exactly when requests started failing. A solo operator or a non-engineer founder has none of that. You see the output, or you do not. When Claude goes quiet, your workflow does not throw an error you can read. It either stalls silently or produces nothing, and there is no screen telling you why. The gap between “something broke” and “I understand what broke” can stretch for hours, and the whole time your customers are waiting.

What the timeline looked like

  • 00:58 UTC, September 22, 2026 - Claude’s public status page reported a bigger problem.
  • About 1 hour of disruption - Runs that depended on Claude would have stalled or failed during this window.
  • 02:12 UTC, September 22, 2026 - Claude’s status page marked the issue resolved.

How a watcher catches this before your users do

NoCrash reads Claude’s public status page every minute. The moment that page flips from working to having trouble, NoCrash sends you a plain-language message, in one place, alongside everything else you build on. You do not have to check anything. You do not have to wait for a customer to tell you.

It also watches the things you ship. If you have n8n workflows, NoCrash watches those through your API token. If you have an app or a page, you can give NoCrash a URL or drop in a small JS snippet and it will watch that too. So if Claude is fine but something on your own side has gone quiet, that surfaces as well.

The honest limit: NoCrash reads Claude’s own status page. It does not find trouble before Anthropic reports it. What it does is make sure you hear about it in plain English, within a minute of Anthropic’s own report, instead of hearing about it from a frustrated customer two hours later.

For the authoritative account of this outage, see Anthropic’s own status page: https://stspg.io/90s4l9yywzbk

Common questions

Frequently asked

What actually caused this?
Anthropic has not published a detailed cause for this outage. The status page says it has been resolved. For any further detail, check the official source at https://stspg.io/90s4l9yywzbk.
Could this happen again?
Yes. Any tool can have another outage. Claude is no different. The question is not whether it will happen again but how quickly you find out when it does.
How do I find out the next time Claude has a problem?
NoCrash reads Claude's public status page every minute. When it reports trouble, NoCrash sends you a plain-language heads-up right away, within a minute of Claude's own report. You get one calm message instead of a confused note from a customer.
My workflow was running during this window. What should I check?
Look at anything that was supposed to run between roughly 00:58 and 02:12 UTC on September 22. If a run was in that window and produced no output, it likely stalled. Re-run it manually and confirm the output looks right before treating it as done.

Catch the next one before your customers do.

NoCrash watches what you ship and sends a plain-language daily brief. Free forever on 3 things to watch.