Microsoft 365 Outage: Outlook, Teams and Artificial Intelligence Copilot Go Down, Explained

On Thursday, November 12, 2026, Microsoft 365 stopped working for people around the world. Outlook would not open, email went undelivered, and Microsoft's artificial intelligence assistant Copilot went down with it. More than 24 hours later, most services are back — but Microsoft has not said when the rest will fully recover, or exactly what broke. This page explains what happened, what it hit, and what is still unclear.

What happened

Microsoft first acknowledged problems on Thursday morning — a Teams disruption at 10:40 a.m. ET — and said the wider Microsoft 365 trouble first arose around 11:17 a.m. ET. Within an hour, outage reports were coming in from 74 countries and territories, and by 1 p.m. ET, Downdetector, a site that tracks user reports, had logged about 130,000 reports about Outlook in the US alone, and more than 47,000 about Microsoft 365. At the peak, reports across five Microsoft products were arriving at 19,650 a minute; on a normal Thursday, Downdetector sees fewer than 15 a minute for the same products.

The first fixes did not hold. Microsoft said on Thursday evening that the changes it had made to improve service were "not taking as expected," and that it was looking for other ways to add capacity. Early on Friday it applied a patch to an upstream authentication and routing component and saw "positive results." By Friday morning, Microsoft reported broader improvements in North America, sustained improvements across much of Europe, and encouraging recovery signals in Asia Pacific. By about 3 p.m. ET on Friday, its service health page listed all services, including Microsoft 365 and Outlook, as "operational" — but it also warned that some users would still see problems, and that email delivery would take longer to catch up than sign-in.

What it affects

The outage was concentrated in Microsoft 365, the subscription service that bundles Outlook, Teams, Office apps and Copilot. At 11:30 a.m. on Friday, Microsoft's health page still listed about 20 products with "service degradation" — and the 365 admin center, where Microsoft pointed people for updates, itself had loading problems.

Status as of Friday afternoon, November 13, from Microsoft's service health page and reporting of it.

OneDrive files were among the services affected — our Cloud Storage page compares OneDrive with the alternatives. Xbox was not part of the outage as far as Microsoft's status pages were concerned: its status page showed all services up, though some players reported problems, according to Vice. And browser users were warned they might still be affected across Outlook.com, OneDrive, OneNote, To Do and Teams.

Why it matters

Microsoft has not said how many people were affected. More than 430 million people use Microsoft 365 apps, according to Microsoft's own website — and Downdetector's reports only count the people who bothered to report, not everyone who was hit. The reports came from 74 countries. This was not a regional blip; it was a worldwide failure of the tools most offices run on.

It reached beyond offices. ABC News reported that major hospitals were affected, including Cleveland Clinic, Mount Sinai, Massachusetts General Hospital and Ohio State University Wexner Medical Center, and that the Department of Health and Human Services issued an urgent alert about government agencies hit by the outage. Email is not a side tool in a hospital; it is how referrals, results and schedules move.

It is also a reminder of how much now runs through one provider. Mail, calendar, files, meetings — and now artificial intelligence assistants that read your mail for you. Copilot went down with Outlook, and so did the ChatGPT and Claude connectors that read Outlook mail through Microsoft's servers. When the mail stops, the AI that sits on top of it stops too.

The bill is not known yet. One expert, RSM Australia risk advisory partner Ashwin Pal, told the Australian Broadcasting Corporation that losses from the downtime are likely in the billions. That is an estimate, not a count — and it is worth remembering that this was a technical failure, not an attack, which matters for insurance: some cyber policies only cover losses from malicious attacks.

What it doesn't prove

First, Microsoft has not published a root cause. What it has said, in a post on Friday afternoon, is that the incident "was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change," and that servers became "unexpectedly unavailable," leaving the rest overloaded. A recent configuration update coincided with the start of the disruption, and Microsoft paused deployments and looked at reverting some of them. A full Post-Incident Report is promised once the review is complete. Until then, the exact trigger is not confirmed.

Second, it was not, as far as Microsoft can tell, an attack. "At this time, we have no evidence of malicious activity associated with this incident," Microsoft said, and executive vice president Charles Lamanna said the same on Thursday night. No group has claimed responsibility. That matters because the first question after any big outage is whether it was a hack — and the answer here, from Microsoft, is no.

Third, it is not fully over. Microsoft's service health page listed all services as "operational" by Friday afternoon, but the company said it has no ETA for full recovery, that mail queues are still draining, and that "service access can recover before mail flow fully normalizes." In plain terms: you may be able to open Outlook while your email is still catching up. Microsoft extended the email timeout window to 48 hours, so delayed messages should arrive rather than bounce — and if you were affected, do not reinstall or reset Outlook: it is a server problem, and local fixes will not help.

Sources & further reading

← Back to DTN News