A Microsoft 365 outage disrupted Outlook, Teams scheduling, Copilot and other services around the world on 12 and 13 November 2026. Microsoft said it saw no indication that a cyber attack was behind it and had not confirmed a cause, and its latest update in the sources Lens read described only incremental improvement. The account below is current as of the ABC live blog’s 9:38 p.m. Australian Eastern entry on 13 November (about 10:38 UTC), and the incident was still being worked at that point.

What was affected

  • Services: According to ABC’s live blog, many users could not reach Outlook email on the web, desktop or mobile, and delivery was delayed. In Teams, core chat and online meetings mostly worked, but calendar and meeting scheduling failed for some organisations. Copilot Chat could fail to load or respond, and Copilot could not reach work email, calendars or files. Microsoft’s list also included Defender, Purview, Intune, Planner, To Do, Power BI, Fabric and Windows 365, with Power Apps added later, and features of SharePoint Online that depend on Exchange were inconsistent. Xbox support said gaming services were up.
  • Scale: ABC reported Downdetector reports for Outlook peaking near 140,000, and later said reports passed 11,000 a minute at one point, with about 70% coming from the United States. Downdetector also showed smaller disruptions at Xbox, GitHub and LinkedIn.
  • Timing: Microsoft dated the start of the problems to around 5:15 a.m. Australian Eastern Daylight Time on 13 November, which Lens converts to early afternoon on 12 November in US Eastern time. ABC’s blog mixes time zones, so the start time is approximate.
  • Organisations named by ABC: Service NSW centres, a New South Wales online bail court, Canberra Health Services and Queensland Health in Australia, with later reports that the Australian Federal Police, ASIO and South Australian and Victorian courts were affected; police in the United Kingdom, some urging the public to call rather than email; the US House of Representatives and courts, from Los Angeles to Texas; Japanese government systems; New Zealand’s foreign ministry and embassies; and Singapore’s courts.

What Microsoft has said

  • Cyber attack: A Microsoft spokesperson said: “There’s no indication this is a cyber attack.” Microsoft later said it was not aware of any malicious activity causing the incident, and Australia’s Home Affairs Minister, Tony Burke, said Microsoft did not believe it was dealing with a cyber event.
  • Cause, not confirmed: In an earlier update ABC summarized, Microsoft said a process on affected servers that sets up secure connections became overloaded and stopped responding, and that it had reverted a recent network change that coincided with the start. In a later update, Microsoft attributed the problem to heavy strain on part of its infrastructure across several regions and said it was investigating a small number of recent configuration updates as possible contributors. Microsoft has not explained why the load built up, and ABC reports that it said it was still working to pinpoint the underlying cause of the traffic influx.
  • Recovery work: Microsoft said it had tested a different health-check approach on a subset of affected systems, which brought those servers back, and was rolling it out more widely while preparing automated detection and restart of unresponsive servers. It said servers stuck in a bad state were blocking some fixes.
  • Latest update seen: In the most recent Microsoft update in ABC’s blog, timed 9:13 p.m. Australian Eastern on 13 November, Microsoft described “incremental improvement” in some services but said “service health remains variable and capacity has not yet fully returned to baseline.” It extended the period for which it holds delayed email to 48 hours, and ABC reports that Microsoft has promised a full root-cause report after mitigation.

Where reports differ

ABC’s blog showed the incident unresolved at its latest Microsoft update, which reported incremental improvement but variable service health. The blog’s page header carries an earlier “updated” time than its newest entries, and Lens relied on the entry timestamps. Search summaries of other outlets say ITdaily reported that restored servers kept failing again and that the outage had not been resolved on Friday morning, that Yahoo News UK, citing PA, reported Microsoft’s fix was not working “as expected”, and that CNBC reported users might still be unable to connect to Outlook on the web or desktop. A summary of an ABC News (US) report cited a Microsoft post saying mitigation had restored service health, and the timing of that post is unclear. Lens has not opened those pages.

ABC’s blog quotes an IT academic, Walayat Hussain, saying OpenAI and Anthropic both reported problems connecting to Microsoft. The blog contains no statement from either company, and Lens found none. Anthropic is the maker of Claude.

Sources and notes

Sources: ABC News (Australia) live blog, 13 November, read as extracted text through its 9:38 p.m. Australian Eastern entry, with roughly the last 7% of the page not read; ITdaily, Yahoo News UK, CNBC and ABC News (US), known only through search summaries. Notes: Microsoft’s own status page and incident report were not read, and the Downdetector figures are user reports, not confirmed counts. The status of the outage may have changed after the time given above.

Image: Gold-lettered Microsoft sign on a stone pillar at an entrance to Microsoft's Redmond, Washington, campus, April 2005. Photo: Derrick Coetzee (public domain), via Wikimedia Commons; cropped. Stand-in image: not a photo of the outage or of any Microsoft 365, Outlook or Teams screen.