Wire reporting: (AP, via Local 10)

Microsoft said Saturday it had restored remaining Microsoft 365 customers after a global outage that began Thursday and lasted more than 24 hours, affecting Outlook, Teams, Copilot and related services^2,5,6,16.

The company posted that it had finished addressing the underlying issue and restored service to remaining impacted customers and regions, Deutsche Welle, Rappler and France 24 reported^2,5,16. The Japan Times said a Microsoft service-health update stated all products were operational and that a Microsoft 365 email backlog had been processed⁶. France 24 said Microsoft had been monitoring that backlog on enterprise services before calling the issue resolved¹⁶.

Microsoft said the disruption was a technical issue introduced during an earlier network capacity upgrade that surfaced after a recent network change, not a cyberattack or AI agents^1,4,7,8,26. Ars Technica, writing Friday before the all-clear, said Microsoft described servers failing automated health checks and becoming unavailable, then a core process becoming overwhelmed so that some restored servers failed again under traffic³. Microsoft executive vice president Charles Lamanna said there was no indication of malicious or cyber activity, MercoPress reported¹⁸.

BBC said Microsoft first acknowledged Exchange Online mailbox access trouble in the Americas at 17:54 GMT Thursday and later warned of possible global Microsoft 365 impact¹. TechCrunch said the company initially described elevated load and later a large number of servers unexpectedly unavailable, with Teams, SharePoint, Copilot, Defender XDR and consumer OneDrive also down⁹. DW, citing Ookla, put the start around 1725 UTC Thursday². ClickOrlando, carrying Associated Press reporting, said Downdetector logged reports from 74 countries and territories within an hour of problems appearing midday Thursday in the United States¹¹.

BBC said the outage reached more than 70 countries and that more than 430 million people use the affected products¹. DW and Rappler said Downdetector logged more than 7.4 million problem reports worldwide^2,5. Rappler said reports peaked near 20,000 a minute⁵. MercoPress said Microsoft 365 has more than 450 million corporate users and 95 million consumers, and Teams more than 320 million¹⁸.

Governments, hospitals, courts, schools and businesses reported disruption^1,3,18,25. BBC said effects included Japan's government, the Philippines Supreme Court, Australia's spy agency, some UK NHS sites and a lost legislative workday for U.S. House staff using Outlook¹. MercoPress said Australia's tax office, some courts, Federal Police, ASIO and a New South Wales services platform were affected, and that Home Affairs Minister Tony Burke activated a digital-resilience clause with Microsoft for priority restoration¹⁸. Nation Africa said Kenyan banks, insurers and the Communications Authority of Kenya lost email, with Standard Chartered Bank Kenya directing customers to call centers²⁵. The Korea Herald said South Korean office workers turned to phones and in-person visits²⁸. Ars Technica said some doctors could not send medication updates and that bank transfers stalled where verification codes arrived by email³.

Nation Africa and The Conversation Australia said OpenAI and Anthropic reported problems with tools that connect their services to Microsoft 365^25,27. France 24 said the U.S. homeland security chief confirmed there was no Microsoft link to a separate U.S. aviation stoppage¹⁶. DW also said the Microsoft failure was believed unconnected to that ground stop².

Microsoft said it restored services gradually because of demand, validating stability at each stage^4,7,26. Ars Technica said Microsoft extended email timeout to 48 hours to preserve delayed mail³. CNBC said Friday that the underlying cause had been addressed while some customers were still being restored⁸.

A Microsoft spokesperson said, "We are deeply sorry for the disruption this incident has caused our customers and partners"¹. The company said it would issue a full post-incident report after its review^1,3,4,26. DW said the event was one of Microsoft's longest outages and its second this year². Rappler cited a Morningstar DBRS estimate of $300 million to $500 million in global insured losses, below figures it associated with the 2024 CrowdStrike incident⁵. Taipei Times, carrying AP reporting, said Burke described Microsoft comparing the episode to CrowdStrike as an update gone awry, not a cyberattack¹⁵.

The Local Germany said Deutsche Bahn's email was down, making it harder to send timetables for diverted trains, and that Microsoft reversed its latest configuration update after a process used to establish secure connections became overloaded³⁴. The Boston Globe, carrying Associated Press reporting, said more than 430 million people use Microsoft 365 apps and that Microsoft posted Saturday it had restored remaining customers³⁰. AP's Thursday dispatch said nearly 130,000 Downdetector users reported Outlook problems just after 1 p.m. EST³². The Guardian reported Australian experts warning of possible billion-dollar local costs and that the Australian Medical Association said patient care was not affected³¹.

What remains unknown is Microsoft's complete root-cause narrative, including whether Azure virtual-network configuration changes cited by a Verge correspondent will appear in the promised report⁴.