A massive, prolonged outage affecting Microsoft 365 applications has dragged on for more than 20 hours, crippling digital communications, enterprise operations, and public infrastructure across the globe. The prolonged failure, which has primarily targeted Microsoft’s Outlook email platform alongside connected cloud services, has sent shockwaves through government institutions, legal systems, healthcare networks, and major technology companies. With more than 430 million people using Microsoft 365 apps worldwide according to figures published on Microsoft’s website, the cascading disruption underscores the extreme global dependency on centralized digital infrastructure.
Although early signs indicated that email service via Outlook was beginning to resume for some users, the sheer length of the disruption has elevated it to an extraordinary event within the technology industry. In cloud computing and corporate IT, a major service disruption lasting over 20 hours is considered an exceptionally long time for systems to remain unstable. The full scope and true operational scale of the outage have not yet been formally confirmed by Microsoft, but real-time crowd-sourced technical monitoring data provided a stark picture of the immediate impact on Thursday evening.
According to data logged by the independent monitoring platform Downdetector, approximately 140,000 individuals across 74 countries submitted formal reports of severe issues with Outlook on Thursday evening alone. Industry analysts emphasize that the true number of affected individuals and organizations is undoubtedly much higher. Downdetector relies primarily on self-reporting by individual users who actively navigate to the platform to log a service disruption. Because Microsoft 365 is deeply embedded within large enterprise, commercial, and governmental networks worldwide, a single incident report logged by an IT administrator often represents an entire corporate tenant comprising thousands of impacted employees.
The widespread integration of Microsoft 365 across global enterprise infrastructure has triggered significant knock-on effects, disrupting third-party technology providers and high-profile artificial intelligence platforms. Leading AI developers OpenAI and Anthropic both confirmed experiencing operational problems directly resulting from the Microsoft infrastructure failure.
ChatGPT creator OpenAI confirmed that it was actively investigating an technical issue affecting its underlying connections to Microsoft’s services. The company acknowledged that these connection failures had directly caused several of its popular tools and features to stop working as expected. Similarly, rival AI developer Anthropic reported that its own systems were experiencing disruption as a direct consequence of the broader Microsoft outage, demonstrating how connected modern cloud ecosystems have become.
Beyond the corporate and technology sectors, the real-world operational fallout from the outage has been widespread, affecting public administration and emergency services internationally. From city municipal governments across the United States to intelligence agencies in Australia, severe disruption has been reported worldwide. In the United States, judicial proceedings and administrative operations faced immediate obstacles, with courts stretching from Los Angeles to Texas issuing official warnings to the public and legal practitioners that official communications, filings, and notices might suffer significant delays while email systems remained compromised.
In Canada, public broadcaster CBC reported that a major healthcare network operating several hospitals in Toronto was experiencing severe email communication issues linked directly to the Microsoft outage. Hospital officials confirmed that while internal administrative and external email communication channels had been disrupted, patient care services had not been affected by the system failure. Meanwhile, the national government of Japan reported that its centralized shared IT platform had suffered severe disruption, impacting digital administrative workflows across state departments.
Across the United Kingdom, the systemic failure created operational hurdles across public healthcare, law enforcement, education, and pharmaceutical services. Essential public services including General Practitioner (GP) medical surgeries, community pharmacies, local police forces, and primary and secondary schools reported being impacted by the ongoing outage.
The loss of primary digital communication channels forced public bodies across the UK to deploy alternative operational measures. Several GP surgeries publicly urged patients to refrain from submitting online messages or consultation requests via practice websites while the outage remained unresolved, directing them instead to urgent telephone channels or in-person visits. Simultaneously, several large police forces advised members of the public seeking to report non-emergency incidents to call switchboards directly rather than relying on digital reporting forms or web-based contact portals that depend on impacted cloud infrastructure.
Addressing the severity of the incident, software industry veteran Gergely Orosz—a former leading software engineer who previously held senior roles at major technology companies including Skype and Uber—suggested that the incident could represent an unprecedented landmark in corporate IT failures. Orosz observed that the ongoing event could ultimately prove to be the most disruptive business email outage of all time. He noted that while assessing the total precise number of impacted users remains challenging, he cannot recall an outage affecting enterprise business email ever taking this long to return to normal operation.
Technical experts have begun dissecting the root causes of the sustained disruption based on diagnostic disclosures released by Microsoft. Prof. James Ferryman, from the University of Reading’s department of computer science, explained that the technical details shared by the company offer critical insights into why the infrastructure crashed and why restoring full service is proving exceptionally complex.
Prof. Ferryman explained that the specific portion of Microsoft’s server infrastructure responsible for managing and establishing secure encrypted connections became completely overwhelmed by incoming traffic. What Microsoft has disclosed indicates that the part of its servers that sets up secure connections is getting swamped, and once that critical load threshold is breached, the server simply stops answering incoming requests altogether.
This initial point of failure sets off a severe domino effect across the network. Prof. Ferryman noted that the primary issue with this specific category of system failure is that it naturally snowballs. When one server falls over under intense traffic, its assigned operational workload is automatically passed along to the next adjacent server in the system architecture. However, because that secondary server is already operating near capacity, it quickly becomes overwhelmed and fails as well. Meanwhile, vast numbers of automated user devices—including millions of personal smartphones, laptops, and desktop clients globally—continuously and automatically attempt to re-establish connection requests, creating a constant loop of retry attempts that further floods the network.
Because of this cascading dynamic, bringing the infrastructure back online requires an intricately managed process. As Prof. Ferryman pointed out, systems engineers cannot simply turn the entire global infrastructure back on all at once. Doing so would instantly swamp the recovering systems with hundreds of millions of queued reconnection attempts, causing the servers to collapse immediately for a second time. Instead, technical teams are forced to bring systems back online incrementally, restoring access bit by bit to stagger incoming device traffic.
Consequently, the primary operational challenge facing engineering teams is not merely repairing the original technical fault that brought the servers down, but effectively managing the massive, unprecedented surge of devices attempting to reconnect simultaneously as services gradually resume.
The incident has reignited broad discussions regarding the reliance of modern global commerce on a small handful of cloud providers. Australian cyber-security expert Troy Hunt told the BBC that the outage carries far-reaching consequences for everyday life and business continuity. Hunt observed that Microsoft constitutes a massive part of the essential internet backbone, adding that a prolonged breakdown of this nature serves as a powerful reminder of how critical email remains to daily society.
Hunt noted that an outage of this magnitude impacts far more than an individual’s basic ability to perform workplace tasks. Beyond standard corporate productivity losses, a collapse of enterprise email infrastructure halts essential everyday administrative interactions. It directly impairs the ability of individuals and organizations to communicate about critical appointments, complete commercial transactions that require email-based identity verification, and create or access vital online services and personal accounts across the broader digital economy. As recovery efforts continue globally, the event stands as one of the most prolonged and disruptive cloud infrastructure breakdowns in recent history.