“We apologize for the outage causing disruptions for our customers, and we are working to address it. We will continue to share information as soon as possible, and for the time being I can confirm there is no indication of malicious cyber activity,” the company spokesperson said.
Microsoft 365 and core communication infrastructure experienced a widespread global outage affecting millions of users across 74 countries and territories, beginning on November 12, 2026. The technical failure crippled Outlook, Teams, and the AI assistant Copilot, driving a massive surge in user disruption reports on Downdetector as businesses, hospitals, and government agencies scrambled to adapt.
“We're observing partial recovery across regions in which we've applied these adjustments; however, these actions aren't resulting in sustained recovery, so we're reviewing what additional actions we can take to improve overall infrastructure stability,” Microsoft said in a service health notice.
The disruption peaked when Downdetector logged roughly 138,058 reports for Outlook and tens of thousands more for Microsoft 365, with approximately 130,000 complaints originating in the United States alone. Services relying on Exchange Online—including SharePoint, Planner, To Do, and the Defender security portal—also suffered degraded performance or outright disconnects, as Bizportal detailed.
At peak moments globally, around 19,600 reports per minute were recorded regarding Outlook, OneDrive, Teams, and Copilot, spreading rapidly across the United States, Europe, Asia, and Africa within roughly an hour.
Microsoft said that the company launched Teams back in 2017 after passing on an acquisition of Slack, expanding usage drastically through the coronavirus pandemic until reporting roughly millions of monthly active users in 2023.
Global Government and Enterprise Fallout
Public sector operations and essential infrastructure felt the impact worldwide. In Japan, the deputy chief cabinet secretary reported that government information systems suffered disruptions, while the Supreme Court in the Philippines, police and hospitals in Britain, Euronext, Indian airlines IndiGo and Air India, and shipping giant Maersk faced operational hiccups.
Swiss luxury group Richemont, owner of Cartier, briefly delayed publishing its half-year results, and both OpenAI and Anthropic reported connection failures for their products linked to Microsoft services. In Taiwan, personal finance management across investment trusts and life insurance companies was disrupted, and British insurer Aviva switched to backup plans during investor calls.
Australia faced widespread public service friction, prompting the Australian Computer Society (ACS) to criticize overreliance on single IT vendors.
“Australians who could not receive a payment, send an invoice or access a government service are entitled to be concerned,” said the Australian Computer Society (ACS), as Channel 14 | C14 reported.
Medical institutions in Britain and Toronto encountered administrative hurdles, forcing staff to revert to paper records for prescriptions while maintaining direct patient care. In India, millions returning from Diwali holidays found their email systems dead. Financial institutions in Taiwan and shipping giant Maersk also reported operational hiccups, while Workhorse Group delayed its third-quarter earnings report.
Technical Root Cause and Recovery Bottlenecks
“The global failure in Microsoft services (like Outlook, Teams and Copilot) was caused by a failure in an automatic component that checks server health, which caused servers to appear unhealthy, disconnect and create a chain load on the rest of the infrastructure,” explained EON development vice president Eyal Ron, according to Globes.
Ron added that recovery efforts dragged on for hours because restoring servers triggered an immediate wave of accumulated traffic. This traffic spike overwhelmed CPU usage and knocked servers offline repeatedly, forcing engineers to step up system recovery gradually.
“The recovery from the outage is taking many long hours because restoring the servers to operation immediately creates a wave of accumulated traffic, which spikes CPU usage and knocks them down again, forcing engineers to return the systems to operation gradually and in controlled stages,” Ron told Globes.
An Australian government minister compared the incident to the CrowdStrike software glitch of summer 2024, which previously knocked out roughly millions of Windows computers and disrupted global flights, banks, and hospitals. Microsoft extended the timeframe for servers to attempt retrying stuck emails to 48 hours to ensure messages would not be lost.
As Channel 7 noted, domestic Israeli media outlets like Globes also faced operational hurdles, confirming that their websites and apps temporarily stopped updating due to the Microsoft infrastructure breakdown.