Government departments, courts and police agencies across the United States lost access to essential communications as a major Microsoft failure spread through international business networks. The disruption affected email, workplace applications, cybersecurity systems, cloud services and artificial intelligence platforms, raising serious questions about the ability of public institutions to function when commercial technology fails.
A major Microsoft outage brought parts of American government administration to a standstill on Thursday, 12 November, disrupting federal courts, police departments, local authorities and businesses across the world.
In Maine, the Judicial Branch warned that applications for protection orders submitted by email might never have reached the courts. The US Court of Appeals for the Tenth Circuit lost email communications for much of the day, while Los Angeles County reported that government departments could not access email and essential communications platforms.
Police agencies in California, local authorities across several American states and municipal governments in Canada experienced similar problems. In Australia, the country’s National Cyber Security Coordinator began examining the disruption as businesses and public agencies struggled to access Microsoft’s systems.
The failure extended across Outlook, Exchange Online, Teams, SharePoint and numerous Microsoft 365 services. It also interrupted connections used by artificial intelligence platforms operated by OpenAI and Anthropic.
At its peak, outage-monitoring service Downdetector registered approximately 138,000 reports concerning Outlook alone. The figures do not represent the total number of people affected, which Microsoft has not established publicly.
The incident exposed the dependence of public institutions on commercial cloud infrastructure, particularly where government communications, legal applications and administrative services rely on systems operated outside their direct control.
US Courts Warn That Legal Applications May Not Have Been Received
Among the most serious consequences was an interruption affecting the American justice system.
The Maine Judicial Branch issued an urgent notice after its email services failed at approximately 1 p.m. local time on Thursday.
By 4:30 p.m., the problem had not been resolved, and officials had received no confirmed estimate of when Microsoft would restore normal access.
The court administration warned that applications for protection orders submitted by email after 12:30 p.m. might not have been received.
Applicants were advised to consider filing paper copies with the courts the following morning.
Protection orders can be sought by people facing domestic abuse, threats, harassment or other circumstances requiring legal intervention.
The notice raised the possibility that individuals who believed they had submitted a request for protection might discover that the court had never received it.
There is no evidence that Thursday’s interruption resulted in an individual being harmed or denied protection. Nevertheless, the warning identified a serious risk associated with the failure of an electronic submission system.
The United States Court of Appeals for the Tenth Circuit also reported losing email services.
The federal court, which covers Colorado, Kansas, New Mexico, Oklahoma, Utah and Wyoming, confirmed that its Clerk’s Office had been unable to send or receive emails for much of Thursday.
The interruption threatened delays in correspondence concerning legal proceedings and notices of activity in pending cases.
The court’s electronic filing systems, CM/ECF and PACER, remained operational. Staff continued providing assistance through telephone and live-chat services.
The contrasting experiences of the two courts demonstrated that the impact depended on which systems an institution used and whether alternative communications arrangements were available.
Neither court reported that its entire judicial operation had ceased.
Los Angeles County and Police Agencies Lose Access
In California, the failure reached one of the largest local government administrations in the United States.
Los Angeles County posted an emergency service notice confirming that departments were unable to access email and important communications platforms.
Residents were directed to contact county offices and agencies by telephone while Microsoft worked to restore service.
The disruption also reached law enforcement.
The Los Angeles Police Department and Los Angeles County Sheriff’s Department confirmed difficulties accessing Microsoft email.
A Sheriff’s Department representative reported that departmental email had stopped updating at approximately 9:40 a.m. local time.
Although the outage affected ordinary communications, officials did not report a general failure of emergency policing.
Nearby Culver City warned residents that certain municipal services were disrupted and that responses from City Hall could be delayed.
The city advised postponing non-essential visits where possible and using alternative contact arrangements.
Emergency services remained available.
Across Nevada, Clark County, Nye County, North Las Vegas and Boulder City warned of interruptions involving government email communications.
In Arizona, Cochise County issued a public advisory stating that messages to county employees and departments might be delayed or might not have been delivered.
Residents dealing with urgent matters were advised to telephone the relevant department.
The county reported that Microsoft had identified problems involving automated server health checks and was attempting to restore affected infrastructure.
In Florida, Jacksonville confirmed that the worldwide outage was affecting city applications, including email.
The city maintained that public safety systems were unaffected and directed residents to existing telephone and online arrangements for non-emergency enquiries.
North Carolina’s Buncombe County reported problems involving automated notices, application confirmations, appointment messages and other email-dependent services.
Departments activated backup communications procedures, monitored affected workflows manually and provided additional staff at public service locations.
The county confirmed that its 911 emergency service remained available.
In Oregon, the State Interoperability Executive Council postponed its scheduled meeting because of the outage.
The council is responsible for coordinating policies and standards intended to improve communication between emergency services and public safety organisations.
The postponement did not indicate that Oregon’s emergency radio networks had failed, but demonstrated the interruption reaching an organisation directly involved in government communications planning.
The incidents establish that the outage affected federal, state, county and municipal government operations.
They do not establish that the entire United States Government stopped functioning. Essential services continued in numerous locations, and some authorities successfully switched to alternative arrangements.
The scale of the disruption was nevertheless significant, particularly because multiple government institutions experienced failures originating from the same commercial technology provider.
How the Microsoft Failure Developed
Microsoft’s incident records show that the disruption began with a Teams messaging failure in the Americas at 2:45 p.m. Scottish time on Thursday.
Users encountered problems sending messages, creating conversations and retrieving existing chats.
Microsoft attributed the initial incident to excessive resource consumption. Automated recovery measures restored the affected Teams functionality by 3:53 p.m.
A separate and more extensive Exchange Online failure began at 4:17 p.m.
Exchange Online provides cloud-based email and calendar infrastructure for organisations using Microsoft 365.
Users began encountering errors opening Outlook, retrieving messages and connecting to mailboxes through web, desktop and mobile applications.
By 6:26 p.m., Microsoft had opened a wider incident covering multiple services and warned that customers outside the Americas could also experience connectivity failures.
Engineers found that substantial numbers of servers had unexpectedly become unavailable.
Automated health checks were removing affected servers from service, forcing more traffic onto the remaining infrastructure and increasing pressure on systems that were still functioning.
Microsoft tested alternative health checks, restarted affected servers, redirected network traffic and temporarily halted software deployments while examining the cause.
Some machines returned to operation, but others became unresponsive again when traffic resumed.
The company subsequently investigated whether particular requests or recent configuration changes were overwhelming critical server processes.
In an update shortly after midnight on Friday, 13 November, Microsoft said it was rebuilding a subset of affected servers using an earlier software configuration and introducing additional traffic restrictions to determine what was triggering the failures.
The underlying cause had not been conclusively established in the available incident records.
There was no verified evidence that the outage resulted from a cyberattack.
Complaints Reached Nearly 140,000
The United States recorded the largest publicly reported volume of complaints.
Downdetector showed a sharp increase in problems accessing Outlook during Thursday’s working day, with customers reporting that messages could not be sent or received and that inboxes would not open.
Reports peaked at approximately 138,000 before gradually declining.
At 2:07 p.m. Eastern Time, the monitoring service recorded more than 97,000 reports involving Outlook, over 38,000 concerning Microsoft 365 and approximately 4,500 involving Teams.
These figures cannot be added together to calculate the number of affected users.
Downdetector collects reports submitted by individuals and other monitoring sources. Some users report several problems, while many people experiencing disruption do not submit any report.
A single organisational failure can also affect thousands of employees without generating an equivalent number of complaints.
By early evening in the United States, the number of reports had fallen considerably.
Microsoft had restored some server capacity, but intermittent failures continued as engineers attempted to return additional infrastructure to operation.
The Disruption Reaches Other Countries
Although the first major reports concentrated on North America, Microsoft confirmed that customers in other regions could also experience problems.
In Britain, outage-monitoring services recorded complaints involving Outlook, Microsoft 365 and related applications.
Microsoft did not publish figures identifying how many Scottish businesses, public authorities or individual users were affected.
Scottish organisations using the affected cloud services were potentially exposed to the same failures, but the available evidence does not establish a complete interruption of Scottish government or NHS operations.
Canada experienced disruption to municipal services.
The City of Cambridge in Ontario reported interruptions affecting telephone systems, email, web services and online applications.
The neighbouring Region of Waterloo also warned that certain services could be delayed.
In Australia, the outage continued into Friday morning.
Downdetector Australia recorded more than 4,400 reports concerning Outlook by 8:20 a.m. local time.
The country’s National Cyber Security Coordinator, Lieutenant General Michelle McGuinness, was informed of the incident and began examining the situation.
Australian businesses, government agencies and other organisations reported difficulties accessing Microsoft services.
Some users also reported problems signing into other accounts because verification codes sent through affected email services could not be retrieved.
Similar complaints emerged in New Zealand, where the interruption coincided with the beginning of Friday’s working day.
Microsoft did not release a complete country-by-country assessment or confirm that every customer in every affected region had lost service.
Microsoft Products and Services Affected
The incident extended well beyond Outlook, although individual products experienced different levels of disruption.
Exchange Online and Outlook were central to the failure. Customers encountered difficulties connecting to email through Outlook on the web, desktop clients, mobile applications and systems using MAPI, Exchange ActiveSync or REST connections.
Microsoft Teams experienced an earlier messaging incident and was included among the products potentially affected by the wider infrastructure problems.
SharePoint Online, widely used by organisations for document storage, internal websites and collaborative working, was also listed.
Microsoft 365 Copilot, the company’s AI assistant integrated with workplace applications, was among the affected services.
Microsoft Planner and Microsoft To Do, used for work allocation, scheduling and task management, could experience failures involving functions connected to Exchange Online.
Microsoft Defender XDR and Microsoft Defender for Office 365 were included in the service-health notices, raising concerns about interruptions to security-related functions dependent on affected Microsoft services.
Microsoft Intune, used by organisations to manage computers, mobile devices and security policies, was also listed.
Microsoft Purview, the company’s information governance and compliance system, experienced problems involving functions such as authorisation, automated labelling, data loss prevention, communication compliance and data classification.
Power BI customers could encounter intermittent errors when exporting reports containing Microsoft Information Protection sensitivity labels to PDF.
Microsoft 365 Apps Admin Center functions, including Cloud Update, Cloud Policy and Inventory, were also identified.
Universal Print, Microsoft’s cloud printing platform, could generate errors when submitting print jobs or performing administrative operations.
Windows 365 customers using particular Microsoft Purview Customer Key configurations could encounter problems connecting to cloud computers.
Microsoft’s incident notice made clear that the affected functions were principally those requiring interaction with Exchange Online or email services. Other functions within the same applications could continue operating normally.
Users also reported difficulties involving Word, Excel, PowerPoint and OneDrive.
Not every reported problem was necessarily caused by the same failure. Microsoft was separately investigating a Word for the web incident involving documents that failed to open or remained on loading screens.
That problem was linked to a browser compatibility issue following a configuration change, and had not been established as part of the principal Exchange disruption.
OpenAI and Anthropic Services Also Affected
The disruption spread into artificial intelligence platforms whose products connect to Microsoft services.
OpenAI confirmed that its Microsoft 365 connectors in ChatGPT were experiencing problems.
The connectors allow authorised users to retrieve information from supported Microsoft accounts and use it within ChatGPT.
New connections could fail, while tools relying on existing Microsoft integrations were not functioning as expected.
The incident affected connected features rather than establishing that ChatGPT itself had experienced a general outage.
Anthropic reported similar difficulties with Claude.
Its first public notice appeared at 5:52 p.m. Scottish time.
The company subsequently identified problems involving an upstream provider, affecting access to mail, calendars, search and other Microsoft-connected tools.
Claude for Outlook was also affected, with users reporting failures opening the add-in or carrying out mail and calendar operations.
Anthropic later confirmed that the problem was caused by Microsoft’s service incident.
The failures showed how disruption within one cloud provider could reach businesses operating entirely separate products.
An organisation using an AI assistant to retrieve emails, search corporate documents or manage appointments could lose those functions even while the assistant’s underlying language model remained operational.
Businesses Delay Operations
The commercial consequences extended beyond the technology industry.
American industrial technology and electric vehicle manufacturer Workhorse Group postponed its scheduled third-quarter financial results announcement and conference call.
The company confirmed that business disruption caused by the Microsoft outage was responsible for moving the event from 12 to 16 November.
Other organisations encountered interruptions involving correspondence, document access, internal communications and administrative systems.
Some businesses were unable to send invoices, retrieve records or access applications connected to Microsoft’s cloud infrastructure.
The full financial cost has not been calculated, and Microsoft has not published a comprehensive list of affected organisations.
The available information nevertheless demonstrates how an infrastructure failure can interrupt commercial activity without damaging customers’ own computers or local networks.
When a Technology Failure Becomes a Public-Service Failure
The most serious consequences of Thursday’s outage involved the interruption of communications between government institutions and the people who depend on them.
Courts process legal applications and orders affecting individual rights. Local authorities provide social services, housing assistance, public records and other essential functions. Police departments require reliable communications to coordinate their work.
Not all these activities were interrupted during the Microsoft incident. Several authorities confirmed that emergency services remained available, and some departments continued operating through telephone systems, unaffected applications and manual procedures.
Nevertheless, the failure exposed the vulnerability of public administration when essential services depend on infrastructure outside government control.
Maine’s warning about protection-order applications was particularly significant.
A person seeking urgent legal protection could submit an application electronically and reasonably expect it to reach the court. If the submission failed without adequate notification, the applicant might remain unaware that the court had received nothing.
The same risk applies to time-sensitive government correspondence, legal notices, applications for assistance and other matters where proof of receipt can be important.
An unanswered email may ordinarily be an inconvenience. A court application that never arrives may carry much greater consequences.
Public authorities retain their legal responsibilities even when a commercial technology provider experiences an outage.
That raises questions about whether their contingency arrangements are adequate.
A second broadband connection cannot restore access to a cloud system whose servers are unavailable. Having electricity, functioning computers and an operational internet connection does not guarantee that an organisation can reach its records.
For essential government services, resilience may require locally accessible information, independent communications systems, alternative submission methods and tested procedures that allow operations to continue without the principal cloud provider.
It also requires clear public instructions.
When electronic communications fail, citizens need to know whether their requests were received, how deadlines are being handled and where they can obtain assistance.
The American courts and local authorities that issued public notices demonstrated the importance of those arrangements.
The incident also raises questions about the concentration of public-service infrastructure among a small number of commercial suppliers.
Microsoft provides services used by governments, hospitals, schools, businesses and other organisations throughout the world. Many apparently separate applications depend on common underlying systems.
When those systems fail, disruption can spread far beyond the original technical problem.
Governments may purchase technology through commercial contracts, but those contracts do not remove their responsibility to maintain public access to essential services.
Recovery and the Questions Still Unanswered
Microsoft’s incident reports showed that engineers had restored some affected infrastructure, but problems continued when certain servers were returned to normal traffic.
As of the company’s update shortly after midnight GMT on 13 November, the wider Microsoft 365 incident remained unresolved.
Microsoft had not established the final cause, the number of affected customers or the complete financial consequences.
There was also no comprehensive assessment of the impact on public authorities across different countries.
The confirmed evidence was already substantial: federal courts, police agencies, municipal governments and businesses had lost access to essential communications, while international cloud services and third-party applications experienced related failures.
For individual users, the immediate consequences included lost access to email and workplace applications.
For businesses, the disruption meant delays, interrupted transactions and reduced productivity.
For government institutions, the implications were more serious.
An infrastructure failure at a private technology company had interfered with court communications, legal applications and the ability of public authorities to receive information from citizens.
The incident demonstrated that the continuity of public administration now depends, in part, on infrastructure that governments neither own nor directly operate.
The central question is whether those institutions can continue meeting their obligations when that infrastructure becomes unavailable.
Sources
Microsoft 365 Service Health — Incidents EX1550543, MO1550821 and TM1550488. Microsoft, 12–13 November 2026.
https://status.cloud.microsoft/
Microsoft Incident Reports for PennO365. University of Pennsylvania Information Systems and Computing, reproducing Microsoft service-health reports, 12–13 November 2026.
https://mailservices.isc.upenn.edu/computing/email/penno365/alerts/ms-incidents.html
Clerk’s Office Email Communications Affected by Microsoft Outage. United States Court of Appeals for the Tenth Circuit, 12 November 2026.
https://www.ca10.uscourts.gov/clerk/news/clerks-office-email-communications-affected-microsoft-outage
Urgent Notice: Outlook Outage Nov. 12, 2026. State of Maine Judicial Branch, 12 November 2026.
https://www.courts.maine.gov/news/article.html?id=13367403
Microsoft Service Outage — Active Incident. County of Los Angeles, 12 November 2026.
https://lacounty.gov/emergency/response/active-incident/
Microsoft 365 Service Disruption. Cochise County, Arizona, 12 November 2026.
https://www.cochise.az.gov/AlertCenter.aspx?AID=Microsoft-365-Service-Disruption-34
Microsoft Issues Impacting County Operations. Buncombe County, North Carolina, 12 November 2026.
https://www.buncombenc.gov/AlertCenter.aspx?AID=Microsoft-Issues-Impacting-County-Operat-140
Global Microsoft Outage. City of Jacksonville, Florida, 12 November 2026.
https://www.jacksonville.gov/welcome/featured-news/microsoft-outage
State Interoperability Executive Council. State of Oregon, 12 November 2026.
https://www.oregon.gov/siec/Pages/welcome.aspx
Elevated Errors Affecting the Microsoft 365 Connector. OpenAI Status, 12 November 2026.
https://status.openai.com/incidents/9gn0ywf2
Microsoft 365 Connector Service Incident. Anthropic, Claude Status, 12 November 2026.
https://status.claude.com/
Workhorse Group Postpones Third Quarter Earnings Release and Conference Call. Workhorse Group, 12 November 2026.
https://ir.workhorse.com/news-events/press-releases/detail/343/workhorse-group-postpones-third-quarter-earnings-release
Australia’s Cybersecurity Chief Investigating Microsoft Outage. ABC Australia, 13 November 2026.
https://www.abc.net.au/news/2026-11-13/microsoft-outage-teams-outlook-copilot/107381196
Microsoft Outage Hits Tens of Thousands of Users. Al Jazeera Staff and Reuters, Al Jazeera, 12 November 2026.
https://www.aljazeera.com/news/2026/11/12/microsoft-outage-hits-tens-of-thousands-of-users
Microsoft Services Down in Global Outage. Nine News Australia, 13 November 2026.
https://www.nine.com.au/australia-news/microsoft-services-down-in-global-outage-20261113-p61dup.html
