The Microsoft outage spread worldwide on Thursday, beginning as a failure affecting Outlook email and Teams video conferencing before disrupting cloud tools used by consumers, businesses and public institutions. The BBC reported that Teams, Outlook and Copilot were inaccessible in at least 70 countries. Microsoft said the disruption was a technical issue, not the result of cyber activity or AI agents, and that it was restoring services gradually because of the scale of demand.
Microsoft outage: From email failure to global cloud disruption
TechCrunch reported that engineers initially struggled to recover downed servers as the problem spread from the U.S. and the Americas around the world. The failure developed into what it described as a full-blown meltdown across Microsoft cloud services for consumer and business email, files and documents. Copilot AI tools and Defender security screening software were also knocked out, while administrators lost access to internal Intune dashboards and features.
The sequence outlined by Microsoft shows how quickly the disruption developed. At 18:07, the company warned that Microsoft 365 connectivity could be affected globally rather than only in the Americas. By 19:31, Microsoft said an investigation had found that a number of services had become unexpectedly unavailable, placing further strain on the remaining infrastructure. Engineers concentrated on an issue that caused servers to fail automated health checks and become unavailable.
Microsoft said at 23:04 that affected servers were being returned to service and had remained stable. At 04:36, it reported partial recovery in some regions. At 09:24, Microsoft described early stabilization and capacity improvements. At 13:32, it reported broader improvements in North America, sustained improvements across much of Europe and encouraging recovery signals in Asia Pacific.
What triggered the disruption
Microsoft said an issue during an earlier network capacity upgrade surfaced after a recent network change. The company stated that it had addressed the underlying problem, but continued to investigate the wider incident. In a statement sent to the BBC, a Microsoft spokesperson said it was deeply sorry for the disruption, adding that it takes its responsibilities extremely seriously.
SiliconANGLE reported that the bug appeared at about 12:35 p.m. EST on Thursday and took Exchange Online offline. That cloud platform stores employee inboxes and calendars. Its on-premises Exchange Server deployments were not affected, although Teams and Copilot 365 features that depend on Exchange inbox data were disrupted. Microsoft also reported positive results at 2:30 a.m. after patching an upstream authentication and routing configuration module.
Microsoft said at 15:39 on November 13 that there was no indication the incident resulted from malicious or cyber activity. It also said it was not related to AI agents. The company said teams were reviewing what happened and how similar issues could be prevented. It promised a full post-incident report once the review was complete.
Scale of the disruption
The Guardian reported nearly 130,000 reports of problems from 74 countries. The affected services included Word, Excel, PowerPoint, Outlook and Teams, as well as enterprise platforms such as SharePoint and OneDrive. Copilot was also unavailable in some cases. The Guardian separately said the disruption began on Thursday evening and lasted roughly 18 hours before services began recovering just before midday GMT on Friday.
Microsoft says more than 430 million people use its Microsoft 365 applications. The company also says more than 90% of Fortune 500 companies use Copilot. Those figures explain why an outage limited to email would still have had broad commercial consequences. Productivity software can become the operating layer for communication, document access, customer contact and internal administration once employees are unable to connect.
The impact extended to government agencies worldwide. The Guardian reported disruption among municipalities in France and the US House of Representatives, whose chief administrative officer sent a campus-wide alert to staffers. SiliconANGLE said court cases had to be delayed and several US municipalities reported technical problems. In Australia, hospitals and health clinics were affected, although the Australian Medical Association said there had been no impact on patient care.
Public services and operational disruption
In the UK, healthcare systems were disrupted, and several British police forces urged people not to report crimes online but to telephone them instead. The National Pharmacy Association warned that email issues could delay some prescriptions. Some schools asked parents to report attendance or absence issues by telephone. National Rail warned that people might not be able to access tickets sent by email.
UK cyber minister Liz Lloyd said the outage mainly affected back-office functions such as messaging and email. She said 999 calls, NHS 111 services, HMRC and DWP phone lines and the transport network were unaffected. The BBC reported that NHS England staff were disrupted, while hospital and rail passengers were affected in Canada. Court services and government offices in the US were also disrupted, although operations were slowly returning to normal.
The disruption reached several business services. SiliconANGLE reported that Workhorse Group postponed its earnings call, while Maersk and Euronext notified customers of communications problems. The BBC also reported that a South African winemaker was unable to process orders for Europe. For organizations that route orders, alerts or operational instructions through Microsoft products, the loss of email can interrupt workflows well beyond individual users.
AI services and external platforms
The outage also affected connectors used by OpenAI and Anthropic. The Guardian reported that OpenAI was investigating problems with Microsoft connectors, including broken new connections and tools that did not work as expected. Anthropic identified an issue with an upstream provider affecting its Microsoft 365 connector. These disruptions show how failures in a large cloud platform can propagate into services built on top of it.
Microsoft 365 was not the only cloud service involved in the wider disruption. The Guardian reported that Microsoft last month disclosed an outage of its Azure cloud platform that affected productivity software and industries worldwide. TechCrunch said the previous major incident affecting Microsoft services occurred in September, when a dayslong cloud outage led the company to blame a core authentication configuration issue.
Market reaction was comparatively restrained. The Guardian reported that Microsoft shares fell 1.1% to $516.86 when Wall Street trading opened. The BBC said the shares dipped by only 0.2% in early US trading, suggesting investors were not overly worried. That muted response came while services were still recovering and the company had not yet completed its post-incident review.
Resilience questions remain
The Microsoft outage has renewed questions about concentration risk in cloud infrastructure. The BBC quoted Chi Onwurah, chair of the House of Commons science and technology committee, who said the incident highlighted a lack of resilience in UK digital infrastructure. The committee previously described reliance on a small number of US-based digital infrastructure providers as a strategic and economic vulnerability.
Mark Gregory, an associate professor in the School of Engineering at RMIT University in Melbourne, said the outage appeared to affect the network layer in Microsoft systems and was potentially similar to the configuration issue that brought down Optus’s network in 2023. The incident therefore carries implications beyond a short service interruption. It exposes how one technical change can affect email, workplace software, security tools, administration, public services and connected AI platforms at the same time.
Microsoft said it remained focused on addressing residual impact and warned that some users could continue to experience connection problems or delayed email delivery. The company was still draining queued messages and rebalancing traffic at 15:39 on November 13. With more than 430 million Microsoft 365 users and extensive enterprise adoption, completing the recovery, explaining the failure and preventing a repeat will be central to how customers judge the Microsoft outage.