Microsoft Service Outage Affecting 74 Countries Fully Restored
nashnova research
Microsoft suffered a major outage starting Thursday midday that hit 74 countries, with hundreds of thousands of users reporting problems at peak; full service was restored Saturday, but the roughly two-day disruption struck government, healthcare, judicial, and financial systems — exposing deep global dependence on a single cloud provider.
What exactly happened?
Starting Thursday midday (US Eastern), Microsoft 365 services — Outlook, Teams, Excel, and AI assistant Copilot — went down in succession. Downdetector logged reports from 74 countries within the first hour.
Microsoft 365 has more than 450 million paid commercial subscribers. This means → even a fraction of users affected translates into a massive absolute number.
Microsoft announced full recovery Saturday, including clearing a backlog of undelivered emails. The incident lasted roughly two days.
What caused it?
Microsoft stated explicitly that the outage was not related to a cyberattack, nor to AI agents.
The official explanation: a problem "introduced during an earlier network capacity upgrade" that "surfaced after a recent network change." In plain terms = a hidden flaw from a previous expansion was triggered by a routine network update.
Services were restored gradually rather than all at once, which extended the overall recovery timeline.
Who was hit hardest?
UK courts: A trial in Manchester was interrupted; Judge Maurice Greene dismissed jurors early, citing the court's over-reliance on technology. Police said they could not accept online crime reports and asked the public to call in urgent requests instead.
UK healthcare: The National Pharmacy Association warned of potential delays in dispensing prescriptions, as the outage disrupted NHS doctors' ability to send electronic prescriptions to pharmacies.
Governments and banks worldwide: Japan, Singapore, Australia, Greece, and Norway reported disrupted government email communications. Some banks in Hong Kong experienced service interruptions. In the US, an Arapahoe County sheriff's office in Colorado flagged that online bail-payment functions were affected.
How much will insurance cover?
Morningstar DBRS initially estimated global insurance payouts at roughly $300 million to $500 million.
For comparison, the 2024 CrowdStrike outage carried estimated insured losses of $400 million to $1.5 billion. This means → the payout scale this time is significantly lower — roughly one-third to one-half of the CrowdStrike event.
Broker Marsh flagged a key limitation: some cyber-insurance policies cover only malicious attacks, not technical failures — and Microsoft's outage was squarely a technical issue. Marsh said claimable business-interruption losses are "likely to be very limited."
What should this really make us think about?
Jeremy Singer, a computer-science professor at the University of Glasgow, said the incident "underscores our dependence on one company's digital infrastructure" and may prompt a rethink of whether IT ecosystems need greater resilience.
This reflects a deeper issue: when 450 million paying users and countless public services run on the same platform, a single capacity-upgrade error can halt courtrooms, break prescription chains, and freeze bank operations.
The final scale of insurance payouts will depend on each company's specific policy terms and on how regulators draw the line between "technical failure" and "insurable loss." In plain terms = who pays, and how much, is still very much an open question.
市场有风险，内容仅供研究参考，不构成投资建议。