Microsoft Down: Why Are Outlook, Teams & Copilot Facing Such A Massive Outage Globally?

A widespread Microsoft 365 outage on Thursday left users around the world unable to open Outlook inboxes and disrupted Teams and the Copilot AI assistant. Some users were affected for more than five hours, and the problems carried into Friday morning in India.

How big is the disruption?

Downdetector logged about 138,000 Outlook problem reports at the US peak, around 1 pm Eastern time (about 11:30 pm IST). Reports fell to roughly 34,000 by the US evening. These figures count user complaints, not confirmed affected accounts.

Microsoft first flagged mailbox trouble in the Americas, then said users in any region could be hit. In Australia, Service NSW asked residents to postpone visits to its centres because some transactions were unavailable. ARN said global media reports put the number of affected users above 100,000. Microsoft's service health page listed Exchange Online, Teams, SharePoint Online, Copilot, Defender XDR and several Purview services as potentially affected.

What went wrong?

Microsoft said a large number of servers became unexpectedly unavailable, pushing their load onto the remaining infrastructure. It also said the software that sets up secure connections on the affected servers became overwhelmed.

Microsoft is looking into whether a change to its network settings, which coincided with the start of the incident, played a part. It also said a small number of recent configuration updates were being investigated as possible contributors, and it paused all deployments and configuration changes while it reviewed them.

Why is the fix taking so long?

The recovery attempts ran into a recurring problem. Microsoft said it validated an alternative health check on a subset of affected servers and that those servers returned to service. But servers brought back online could become unresponsive again once traffic reached them.

The company then began restarting unresponsive servers, rerouting traffic where possible, and temporarily limiting which types of traffic could reach affected infrastructure in some locations. It also began rebuilding some servers on a previous configuration to test whether that stopped the failures.

Why could Teams work when its calendar did not?

Open Magazine reported that Teams' core chat and online meeting functions stayed largely usable. Scheduling a meeting or opening the calendar could fail because those features depend on Exchange Online, the service that stores and delivers work email and runs calendars. This is why a user could join a Teams call but be unable to book one.

Why were ChatGPT and Claude affected?

The outage also hit the Microsoft 365 connectors that let AI assistants access a user's work email, calendar and files with permission. OpenAI said new connections and some functions might fail. Anthropic said most mail, calendar and search requests through its connector were failing at the time of its update, and it flagged problems with Claude inside Outlook. Microsoft's own Copilot also struggled to reach work email, calendars and files.