Microsoft 365 suffered a widespread outage on Thursday, November 12, leaving some users unable to open their Outlook inboxes, delaying email and disrupting features in Teams and Microsoft’s AI assistant, Copilot. Downdetector logged about 138,000 Outlook problem reports at the US peak. The disruption lasted more than five hours for some users, and two other AI services, ChatGPT and Claude, reported problems connecting to Microsoft 365.
Is Outlook working again? #
Microsoft had restored some capacity by early Friday in India, but full recovery was still unconfirmed in the update reported at 7:18 a.m. IST. The company said some servers worked after a restart, then stopped responding when they began taking requests again. Microsoft was testing a fix on a small group of machines and had begun reversing a recent change to its network settings. Some customers could reconnect while others still had trouble reaching their mail.
The Wealth Issue 2026 #
06 Nov 2026 - Vol 05 | Issue 45
Portraits of Dreamers and Doers under 35
Read Now The Wealth Issue 2026
How big was the outage, and did it stop at the US border? #
Outlook problem reports peaked at 138,058 around 1 p.m. US Eastern time on Thursday, or 11:30 p.m. IST, and fell to about 34,490 by the US evening. Those are reports to an outage tracker, which measures how many problems people flag. Microsoft initially identified mailbox problems in the Americas and later said users in any region could be affected. Australia offered a concrete example of the spillover. Service NSW, which handles transactions for residents and businesses, asked customers to delay visits to its centres because some transactions were unavailable.
What is Microsoft 365, and which part broke? #
Microsoft 365 is the company’s subscription bundle for work apps and online services such as Outlook, Teams and SharePoint. The main trouble was with Exchange Online, the service that stores and delivers work email and runs calendars for Microsoft 365 customers. Outlook is the app people use to reach that service. If the service cannot answer, the app can open while the inbox fails to load or a message takes longer to arrive. Microsoft logged mailbox trouble from 9:47 p.m. IST on Thursday and recorded the broader Microsoft 365 incident separately from 11:56 p.m. IST.
What made the servers stop answering? #
Microsoft said software that establishes secure connections on the affected servers became overwhelmed. As servers dropped out, fewer machines were left to handle incoming requests. The company said the strain came from lost capacity. It was checking whether a network settings change that coincided with the start of the incident contributed to the failure. A Microsoft spokesperson told The Guardian the company had seen no indication of a cyberattack.
Why could Teams take a call but struggle to book one? #
Microsoft said Teams’ main chat and online meeting functions remained largely usable during the broader incident. Scheduling a new meeting or opening a calendar could fail because those features use Exchange Online. Microsoft had also fixed a separate Teams messaging incident earlier on Thursday. In the later outage, the Outlook inbox and the Teams calendar could fail together because both relied on the same underlying service.
Why did ChatGPT and Claude get caught up in it? #
Their Microsoft 365 connectors, which let the AI assistants access a user’s work data with permission, ran into errors. OpenAI said new connections and some functions might fail. Anthropic said most requests involving mail, calendar and search through its connector were failing at the time of its update. It also flagged problems with Claude inside Outlook. Microsoft’s own Copilot could struggle to access work email, calendars or files.
Is this the first time Outlook and Teams have gone down together? #
Both had trouble on January 25, 2023. Microsoft’s investigation traced that disruption to a change on its global network that interfered with connections to Microsoft 365 and its Azure cloud service. Indian users reported problems that day. In October 2025, a separate failure in Azure Front Door, a service that directs internet requests to the right servers, also disrupted Microsoft 365. In each case, customers saw problems in the apps they opened. The failure started in systems those apps relied on.
What do we know about India this time? #
Microsoft said the November outage could reach any region. A reliable India-specific count or list of affected organisations was unavailable. India has seen what dependence on a shared provider can mean. In July 2024, airlines at Delhi airport’s Terminal 3 handed out handwritten boarding passes after a faulty CrowdStrike security update disrupted Windows computers. That was a different failure, on the computers themselves. In the November 2026 incident, the confirmed problem was on Microsoft’s servers.
How will Microsoft know the fix has held? #
The test is whether repaired servers keep answering as Microsoft restores normal traffic. For a workplace, that means checking whether new messages arrive and calendar invitations go through once Outlook opens again. Microsoft 365 administrators can follow the wider incident under MO1550821 and the mailbox incident under EX1550543. Microsoft’s account of the failure still has to establish what set it off and whether the network rollback helped.