Infrastructure ·
Microsoft 365 outage: Outlook, Teams and Copilot down for hours, what to do
On Thursday 12 November 2026 a large Microsoft 365 outage hit Outlook and Exchange Online, Teams and Copilot for hours, according to media reports. Microsoft said there is no sign of a cyberattack.
On Thursday 12 November 2026 part of the Microsoft 365 cloud failed worldwide. According to CNBC and other outlets, Outlook and Exchange Online, Teams, SharePoint Online, Microsoft 365 Copilot, Defender XDR, Planner, To Do and Purview were disrupted for many hours. TechRepublic reported that Downdetector reports for Outlook alone peaked above 130,000. In Czechia the search "microsoft výpadek" became one of the most searched tech terms, and Czech outlets wrote about problems lasting beyond 13 hours.
Microsoft told The Register that servers in the affected infrastructure were failing automated health checks and stopped processing requests, and that it was reviewing recent configuration changes. The company also said there is no indication of malicious activity, so this was not an attack. A final root cause has not been published; it is expected in the post-incident report. Reports say capacity was being restored and users could reconnect gradually, but some recovered servers could fail again under load.
What should you do if the outage is still ongoing or returns? Follow the Microsoft 365 admin center under incidents EX1550543 and MO1550821, the IDs cited by the media. For ordinary users the advice is simple: wait, and do not delete or reconfigure accounts, because the fault is on Microsoft's side. It helps to have a backup channel ready, such as a phone list of key people and an alternative email address for urgent customer messages.
Reports also say the outage affected connectors to OpenAI and Anthropic AI services. That is a lesson for companies: an agent that reads mail, calendars or documents from one cloud stops working together with it. The more processes depend on that single connection, the more it pays to plan for the input being unavailable now and then.
What it means for companies
The lesson for a Czech company is the same as after every big outage: email, client communication and automation must not all rest on one single point. We can map dependencies in an AI consulting and audit, design AI agents that pause safely when a data source is down, and keep your website and customer channels independent through EU hosting. The first consultation is free.
Sources: CNBC: Widespread Microsoft outage takes down Outlook and other 365 services, The Register: Microsoft struggling to fix massive cloud outage