Global outage takes down Outlook and Microsoft 365 for nearly two days

- Reading
- 4 min
A global outage affected Outlook, Microsoft 365, Teams, SharePoint Online and Copilot over nearly 48 hours. Microsoft says there was no attack and no AI involved: the cause was a network capacity change in Azure.
A global outage in Outlook and Microsoft 365 left millions of users without access to email and other services for nearly two days, according to a report by The Verge. Broad recovery only began to appear on Friday morning, November 13, 2026, when Microsoft confirmed the issue had been addressed and that message-sending flow could still suffer delays while systems stabilized.
The company published an official statement acknowledging the impact on customers and ruling out two hypotheses that had been circulating among users and the press: a cyberattack and a failure linked to AI or autonomous agents.
We are deeply sorry for the disruption this incident has caused our customers and partners. We know the trust they place in our services and take that responsibility extremely seriously. We are seeing broad recovery and stability across our services. This was a technical issue not related to cyber or AI / agents. The incident was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change. We have addressed the underlying issue and continue to restore service to remaining customers and regions. Given the scale of demand, we restored services gradually, validating stability at each stage. While this took longer, it helped ensure a stable restoration. We will provide a full post-incident report once our review is complete.
Microsoft, official statement
What broke down, beyond email
The scope of the incident was not limited to the inbox. According to the Verge, SharePoint Online, Teams, Copilot AI and other 365 suite apps were also affected. For those working at companies that concentrate communication, file storage and automation within the same Microsoft ecosystem, this meant practically every front stopped at the same time: email, corporate chat, shared documents and the AI features built into Office.
This concentration is the point that matters most to those who design systems. A network failure inside Azure, the platform that underpins all of 365, simultaneously took down services that many organizations treat as independent of one another simply because they have different names in the interface.
The technical cause, according to sources
Microsoft attributed the problem to a network capacity upgrade made earlier, whose side effect only appeared after a more recent network change. The company's statement avoids detailing the exact layer where the error occurred, promising a full post-incident report after the internal review.
Tom Warren, senior correspondent at the Verge, reported based on his own sources that the outage is linked to configuration changes in Azure VNet, the virtual network layer that isolates and connects resources within Microsoft's cloud. Sky News technology correspondent Rowland Manthorpe received from Microsoft the same characterization passed on to the Verge: a technical issue, unrelated to a cyberattack, AI/agents or DDoS.
In short: a network infrastructure change, designed to increase capacity, interacted poorly with a later change and generated a cascading failure that Microsoft itself took nearly two days to safely reverse.
Why recovery was slow
The company justified the delay by citing the scale of demand: instead of restoring everything at once, it chose to re-enable service region by region, validating stability at each stage. This is the definition of a cautious rollback for a platform the size of Azure: reverting too quickly, under the production load of millions of accounts, risks generating a second wave of instability worse than the first.
It is also the least visible part of any incident at this scale. The end user only sees how long the service was down; whoever operates the infrastructure must decide, in real time, between restoring quickly with the risk of relapse or restoring slowly with guaranteed stability.
What this means for those who depend on the cloud
For engineering and architecture teams running critical operations on top of Microsoft 365, the episode reinforces a point that incidents at major cloud providers repeat with some frequency: dependence on a single vendor for email, chat, storage and AI automation means that an internal network failure takes everything down at once, with no automatic fallback between these pillars.
In Brazil, where a significant share of mid-size and large companies run corporate email, Teams and SharePoint under the same Microsoft 365 tenant, the outage is a concrete reminder to review continuity plans:
- Having an alternative communication channel that does not depend on the same provider (a group on another platform for urgent notices, for example);
- Documenting, before an incident happens, which business flows break if Outlook, Teams or SharePoint go down simultaneously;
- Treating cloud SLAs as part of the disaster recovery plan, not as an absolute guarantee of availability;
- Following the post-incident report promised by Microsoft to understand the root cause at the VNet layer, in case the company's own architecture depends on similar network patterns in Azure.
What remains open
Microsoft had not released, as of the Verge's update on November 13, the full post-incident report nor a detailed timeline of when each region and service was restored. There is also, based on available sources, no official number of affected users or organizations, nor any estimate of financial impact. This data tends to appear in the post-incident report that the company promised to publish after completing its internal review, and it is that document that will confirm, in technical detail, how a network capacity change meant to improve the service ended up taking down much of it for nearly two days.
Translated from the Brazilian Portuguese original · Read the original
Fake job offer hides malware in a Git repository inside a ZIP
A Russian developer documented how a supposed recruiter used Git hooks and filters to run malicious code with nothing more than a git status, without requiring the execution of any suspicious file.






