Microsoft said on Friday it was seeing signs of stabilisation after Outlook, Teams and Copilot were disrupted around the world for more than 12 hours.
More than 12 hours after the failure began, the company said it had made only limited progress restoring its most widely used workplace services.
Users began reporting they couldn't access Outlook at about 16:00 GMT on Thursday 12 November. Downdetector recorded reports across 74 countries and territories within about an hour.
- 12 NovOutlook goes down
Users report they cannot reach Outlook from about 16:00 GMT.
- 12 Nov74 countries
Downdetector logs reports from 74 countries and territories within about an hour.
- 13 Nov43,000 reports
By 07:00 GMT, at least 43,000 Outlook reports from users in 72 countries.
- 13 NovRollback
Microsoft rolls back a network change but does not say whether it was to blame.
- 13 NovFixes widened
Microsoft applies fixes more broadly across affected regions as it seeks persistent recovery.
By 07:00 GMT on Friday, users in 72 countries had filed at least 43,000 reports of Outlook problems with Downdetector.
Microsoft said features of Microsoft 365 that depend on Exchange Online and email were affected, while the elements that do not rely on Exchange Online kept working. Outlook, Teams, SharePoint Online, the Copilot chatbot, OneDrive and OneNote were among the products hit.
Microsoft said its investigation found that a core process on affected servers was becoming overwhelmed, causing servers to stop responding and further reducing capacity. Several servers unexpectedly became unavailable, leaving the others to take on their workload; the higher load wasn't caused by increased traffic.
For now I can confirm that there is no indication this is the result of malicious or cyber activity
David Arroyo Guardeño, a senior researcher at Spain's National Research Council (CSIC), said the abnormal rise in network traffic initially seemed consistent with a technical fault, probably involving BGP and DNS management, rather than an attack.
Microsoft has not published a root cause and has not given an estimated time for full resolution. On Friday it was rolling back a network change, but it did not say whether that change was to blame.
We’re seeing positive results for service stability and increased capacity as a result of our latest traffic optimization actions. We’re implementing these actions more broadly across affected regions as we seek to attain persistent recovery.
The failure reached Congress and US AI developers
The outage affected users in the US House of Representatives, which said its engineers were working with Microsoft. Microsoft connectors used by OpenAI and Anthropic also experienced problems, putting two of the country's leading AI developers among those hit.
Microsoft's own figures put its Microsoft 365 customer base at 450 million corporate users and 95 million individual customers or more. Teams has more than 320 million users.
Airlines, exchanges and earnings calls
Email support was disrupted at Indian airlines IndiGo, Akasa Air and Air India.
Euronext said its listing teams were having email communication issues. Aviva in Britain prepared backup plans for earnings calls, and Denmark's Maersk warned customers that emails and updates could be delayed.
According to two sources, the disruption hit fund management at Taiwanese securities firms, life insurers and investment trust companies. One company cancelled a bond repurchase transaction; another used fax to confirm trades.
Workhorse Group, a US electric truck maker, postponed its results, while Richemont in Switzerland briefly pushed back its first-half results announcement on Friday.
PVH Corp., which owns Calvin Klein and Tommy Hilfiger, planned to cut about 4% of its global workforce. According to a person familiar with the matter, some conversations due to start in Asia on Friday morning didn't happen because of the outages.
Governments from Canberra to Tokyo
The organisations affected included Australia's spy agency and federal police.
Email was unavailable to Japan's finance ministry for most of the day, it said, and ministries put up paper notices to tell journalists about daily schedules. Masanao Ozaki, the Deputy Chief Cabinet Secretary, said the outage was affecting the government's information system infrastructure; communications at the Supreme Court of the Philippines were also disrupted.
Australia invoked a clause in its digital resilience agreement with Microsoft, signed in June, requesting that the company put Australia's interests first while restoring services.
They believe they are dealing in some ways similar to what we saw with respect to CrowdStrike where there is an update that has gone awry
In New Zealand, immigration officials paused the processing of all visa applications. In Singapore, Downdetector began recording reports at about 2am; by 9.15am it listed 1,130 Outlook reports, up from about 730 an hour earlier. Microsoft 365 reports climbed from 434 at 8.30am to 566 at 9.20am.
The Government Technology Agency of Singapore said some government systems in the country had been affected by the disruption.
Repeat failures put pressure on IT departments
Microsoft's 365 services have now suffered a second disruption in 2026. The company also experienced a 365 outage in January 2026 and one in October 2025, and reported an Azure cloud outage last month.
The bigger worry is that this isn’t the first 365 outage this year, and if reliability starts to look like a recurring problem, it hands every IT department a reason to think twice before putting even more of their business in one provider’s hands
I don’t think it’s unprecedented. I mean, the scale is impressive
Microsoft shares were down 0.4% in extended trading.
Accounts of the scale differ. Downdetector counts put the number of countries and territories with reports at 74 within about an hour of the start, and at 72 by 07:00 GMT on Friday; complaints have been described as about 200,000 and as hundreds of thousands. One account put the disruption to Outlook and Teams at about 10 hours, while others described an outage of more than 12.
About 430 million people use Microsoft services, said Walter Pasquarelli, a researcher affiliated with the University of Cambridge's Bennett School of Public Policy, adding that the problem is first and foremost a communication risk.
What we know
- The outage began at about 16:00 GMT on Thursday 12 November 2026.
- Microsoft said it was not aware of any malicious activity causing the incident.
- The US House of Representatives said its users were affected and its engineers were working with Microsoft.
Still unclear
- The root cause, which Microsoft has not published.
- Whether the network change being rolled back on Friday was to blame.
- The total number of users affected, which Microsoft has not confirmed.
- When full service will be restored, for which Microsoft has given no estimated time.
Workhorse Group's rescheduled results are due on 16 November.
