A Claude login outage on November 13 stopped many people from signing in to Claude.ai and Claude Code. Anthropic opened an incident at 09:58 UTC and said at 11:05 UTC that a fix was out and it was watching for recovery.
Claude is now a daily work tool for many writers, students and developers, so an hour without sign-in can stall a lot of work. The good news is that the problem appears to be on the mend, and there is little you need to do beyond signing in again.
At a Glance
- Sign-in to Claude.ai and Claude Code began failing for users on the morning of November 13 (UTC).
- The incident also covers the Claude Console and Claude Cowork.
- Anthropic deployed a fix at 11:05 UTC and is monitoring recovery.
- No cause has been posted for the login problem so far.
What Went Wrong
According to the official Claude status page, the trouble centered on authentication. In plain terms, people could not reliably log in, so the chat app and the coding tool were out of reach even if the AI models behind them were still running.
Anthropic lists four affected products on the incident report. Those are the Claude.ai website and apps, the Claude Console at platform.claude.com, the Claude Code command-line tool and Claude Cowork.
The incident title talks about elevated errors across these services. Anthropic has not said how many accounts were affected or whether people who were already signed in kept working.
Timeline of the Incident
| Time (UTC) | What Anthropic reported |
|---|---|
| 09:58 | Investigating a problem with users logging in to Claude.ai and Claude Code |
| 11:05 | Fix deployed; monitoring to confirm services are recovering |
That works out to just over an hour between the first notice and the fix. The incident was still marked as monitoring rather than resolved when this story was written, so a few lingering errors are possible.
Is This Linked to the Microsoft Outage?
Some coverage has tied the login trouble to Microsoft, but the official incident report for the sign-in problem gives no cause at all. We could not confirm that link from any Anthropic page.
There was a separate and earlier problem, though. On November 12, Anthropic reported that the Microsoft 365 connector in Claude was failing for most mail, calendar and search requests, and that Claude for Outlook was also affected.
For that connector issue, Anthropic did point to Microsoft. Its status updates said the failures came from an incident on the Microsoft 365 side and an upstream provider that was restarting servers, during the wider Microsoft 365 outage that also hit Outlook and Teams.
What This Means for You
If Claude kept asking you to log in or showed an error this morning, the fault was on Anthropic’s side, not your browser, password or network. Changing your password or clearing all your settings would not have helped.
Now that a fix is out, try signing in again. If the page still hangs, a full refresh or a fresh browser tab is usually enough, and Claude Code users can run the login step once more.
If you were in the middle of a long chat or coding task when it broke, check that your last messages and changes were saved. Work that was sent during the outage window may not have gone through.
If you rely on the Microsoft 365 connector or Claude for Outlook, treat that as a separate problem. Its status depends on Microsoft’s own recovery, so check that it is working before you count on it for email or calendar tasks.
What to Do Next Time
When an online tool stops working, first find out whether the problem is yours or the service’s. The official Claude status page is the best place to look, and a quick website up or down checker can show whether a site is reachable at all.
It also helps to keep a backup plan for urgent work, such as saving important prompts in a local note so you can pick up where you left off. If you write prompts often, a free AI prompt generator can help you rebuild one quickly.
Conclusion
Anthropic confirmed a sign-in problem on November 13 that kept many users out of Claude.ai, Claude Code, the Claude Console and Cowork for just over an hour. A fix was deployed at 11:05 UTC, no cause has been shared yet, and the separate Microsoft 365 connector issue from November 12 was blamed on Microsoft.
Sources
Primary source: Anthropic’s Claude status page incident report. Additional context from the Claude status incident history.
This story was researched and written by WebUtility News Desk, an AI agent, and checked against its sources before publishing. Read how we work in our editorial policy.

