Microsoft’s Global Outage Shows How Much of the World’s Business Runs on One Company

A prolonged disruption to Microsoft’s cloud-based productivity services has disrupted businesses, government agencies and technology companies across several continents, exposing the operational risks of organisations increasingly dependent on a handful of technology providers for their daily operations.

The outage, which began on Thursday, November 12, affected Microsoft 365, Outlook, Teams, SharePoint and Copilot, leaving users unable to access email, coordinate meetings, retrieve workplace information and use some cloud-based business applications. The disruption extended into Friday’s business hours in Asia-Pacific, as Microsoft struggled to restore affected infrastructure and acknowledged that its initial recovery measures were not delivering the expected results.

The incident has raised questions about the resilience of enterprise technology infrastructure at a time when companies are moving more of their operations into cloud environments and integrating artificial intelligence into essential business processes. It also highlights the potential economic consequences of a technical failure at a provider whose software is deeply embedded in corporate communications, financial administration, cybersecurity and government services.

Microsoft initially attributed the disruption to servers that had become unexpectedly unavailable, creating capacity constraints across affected services. The company said the problem was not caused by an increase in customer traffic. As engineers attempted to restore the infrastructure, however, some servers became unresponsive again after receiving traffic, complicating efforts to return services to normal.

Microsoft’s investigation subsequently identified an overwhelmed core process on affected servers as a factor contributing to their failure to respond to requests. The company began isolating parts of its infrastructure, restarting affected components, restricting certain types of traffic and testing servers using an earlier software configuration. It also examined whether separating traffic across regions could allow affected areas to recover independently.

In its latest service updates, Microsoft said adjustments to the affected servers had not produced sustained recovery, although some regions had experienced partial improvement. The company said it was investigating additional measures to improve infrastructure stability and restore capacity.

The disruption was not limited to Microsoft’s email service. Its effects extended to applications and business tools that depend on Exchange Online, the cloud-based platform responsible for email, calendars, contacts and other messaging functions.

Users experienced difficulties accessing Outlook through web browsers and desktop applications, while email delivery was delayed. Teams remained largely functional for basic messaging and online meetings in some environments, but activities dependent on Exchange Online, including calendar access, meeting scheduling and the creation of certain chats, were affected.

SharePoint users could continue accessing files stored directly on the platform in some cases, although features connected to Exchange Online experienced disruptions. Copilot users encountered problems loading the service, receiving responses and accessing workplace information such as emails, calendars and documents.

The incident also affected parts of Microsoft’s cybersecurity and enterprise management infrastructure, including Defender XDR, Defender for Office 365, Purview, Planner, To Do and Intune. Some Microsoft Graph requests involving email, calendar and contact information also experienced failures, potentially affecting applications that rely on Microsoft’s programming interfaces to exchange business data.

The breadth of the disruption illustrates how interconnected enterprise software has become. A company may use Outlook for communications, Teams for collaboration, SharePoint for document management and Copilot for information retrieval, but these applications are not necessarily independent systems. They frequently rely on shared identity, messaging, data and infrastructure services.

Consequently, a failure in one underlying component can affect several products simultaneously, even when their individual applications appear operational.

The commercial consequences were already visible on Thursday. Electric truck manufacturer Workhorse Group postponed the release of its third-quarter financial results, which had been scheduled for November 12, until November 16, citing business disruptions associated with the Microsoft outage.

The incident also affected artificial intelligence companies that rely on Microsoft’s infrastructure to connect their services to enterprise applications. OpenAI said it was investigating problems with Microsoft connectors that could prevent new connections from working correctly. Anthropic also identified an issue with an upstream provider affecting its Microsoft 365 connector.

These disruptions illustrate the expanding dependence of the AI industry on enterprise cloud infrastructure. As organisations integrate AI assistants into email, document management, customer support and internal knowledge systems, the reliability of those tools increasingly depends on the availability of the underlying platforms and application programming interfaces.

A failure in an enterprise software provider can therefore affect not only its direct customers but also businesses operating through interconnected technology ecosystems.

The outage attracted more than 130,000 reports of Outlook problems on Downdetector, a website that tracks service disruptions through user reports, at its peak on Thursday. The number of reports subsequently declined, although problems continued into Friday, including in parts of Asia-Pacific. Downdetector figures represent reported incidents rather than a verified count of affected customers, meaning the actual number of users experiencing problems could differ.

Government institutions were also affected. Agencies in Australia, Japan and the United States reported disruptions to email and other information technology services. Australia’s federal police and security agencies were among organisations affected, while the U.S. House of Representatives said some of its users were experiencing problems and that engineers were working with Microsoft.

In Australia, government departments began assessing affected systems and activating business continuity arrangements. Service NSW, which provides government transactions to residents and businesses, reported that some services at customer centres and call centres were unavailable and advised people to postpone non-urgent visits.

The disruption also affected administrative systems used by educational institutions and public services. In Japan, government officials said they were monitoring the impact on government IT infrastructure.

The effects on airlines appeared more limited. Air India and IndiGo reported that their flight operations remained unaffected, although Air India rerouted some email communications and IndiGo relied on alternative customer support channels. This distinction highlights the importance of operational redundancy: organisations with independent systems for essential functions can maintain services even when supporting communications infrastructure fails.

Microsoft said it was not aware of malicious activity causing the disruption. Australian Home Affairs and Cyber Security Minister Tony Burke said the company believed the incident was not a cyberattack and had drawn comparisons with the 2024 CrowdStrike disruption, although Microsoft had not confirmed that the two incidents shared the same technical cause.

The distinction is important because outages caused by infrastructure or software failures require different responses from those involving malicious activity. A cyberattack may require organisations to isolate compromised systems, investigate unauthorised access and restore data integrity. An infrastructure failure instead requires identifying the malfunctioning components, restoring capacity and ensuring that recovered systems can handle normal traffic without failing again.

Stay ahead of the Stories shaping our world. Subscribe to Impact Newswire and join our 
WhatsApp Channel for updates on global tech, business, and innovation—all in one place.

Dive deeper into the future with the Cause Effect 4.0 Podcast, where we explore the ideas, trends, and technologies driving the global AI conversation.

Got a story to share? Contact Us to reach a global audience with Impact Newswire.