‘Sorry for the disruption’: Microsoft outage causes 2 days of trouble

Outlook email and other Microsoft services were returning to normal Friday, following an outage that began Thursday morning and disrupted office work worldwide.

Microsoft said there were no signs that a cyberattack or rogue artificial intelligence agents were to blame. Rather, it was a technical issue caused by the company trying to upgrade network capacity. 

By Friday morning, as businesses and agencies on the West Coast opened, most Outlook inboxes were back up and running, and Microsoft was working on clearing the backlog of emails sent into the digital abyss. 

“We are deeply sorry for the disruption this incident has caused our customers and partners,” a Microsoft spokesperson said. “We know the trust they place in our services and take that responsibility extremely seriously.”

Microsoft said at noon on Friday that it did not have an estimated time for full recovery. The company later said the outage was not linked to air traffic control problems that halted departing flights across the U.S. on Friday.

The Microsoft outage created difficulties for businesses and government agencies worldwide, spurred online jokes about the loss of email and led to speculation about what had gone wrong.

Microsoft Executive Vice President Charles Lamanna, who oversees Copilot and the platform behind Outlook, tried Thursday to snuff out speculation of cyberattacks or AI troubles.

“I can confirm that there is no indication this is the result of malicious or cyber activity,” Lamanna said in a social media post Thursday night.

Lamanna disclosed more about the technical failures in a social media post Friday afternoon, and again rebuffed speculation.

“This was not a cyberattack or an issue caused by AI tooling,” he said. “Our amazing team was able to work through a very challenging situation to restore service.”

The company will issue a full postincident report once a review is complete.

‘An economic vulnerability’

Starting around 9:30 a.m. Thursday, tens of thousands of users reported connectivity issues with various services under the Microsoft 365 umbrella, including Outlook, Copilot, the messaging app Teams, Word, Excel and other productivity software. As the day progressed, disruptions primarily affected Exchange Online, the engine behind Outlook and other Microsoft messaging services.

The situation did not look good Thursday evening. While one side of the globe turned away from their abnormally inactive inboxes and apps, Microsoft’s updates said fixes weren’t working to resolve the outage, which stretched beyond normal downtimes. 

Because the outage mostly affected Microsoft’s messaging services, it didn’t cause the same havoc as a major Windows outage did in July 2024, when a cybersecurity update pushed out by CrowdStrike completely crashed computers around the world. During that earlier episode, libraries and clinics closed, and delays caused chaos at airports as millions of people were completely locked out of their computers.

Hits to productivity from this week’s incident resulted in estimated global insured losses between $300 million and $500 million, according to Morningstar DBRS. That’s well below the top end estimate of $1.5 billion from the CrowdStrike-caused outage. 

Jay Aldebert, chief growth officer for Illinois-based consultancy International Services Inc., said the financial damage tallied up after outages like Thursday’s are usually the losses among Fortune 500 companies and what happens on the stock market. Daylong outages can create chaos for firms like his that are overly dependent on services from a limited number of tech companies.

“On Thursdays, some people send their payroll to companies like ADP through email, and when you have to wait another day, what do you think that does to productivity and morale?” he said. “And if I perform a service for you and send you an invoice, we’d be in a situation where you wouldn’t be able to pay it.”

James Vrachas, an executive director at Houston, Texas-based investment firm Post Oak Group, said the outage shows an overreliance by millions of businesses on a relatively small number of providers.

“That’s efficient when everything works, but it creates significant concentration risk when something goes wrong,” Vrachas said. “A disruption at one company can quickly affect businesses, financial institutions, and government operations across the globe. This isn’t just a technology issue anymore. It’s an economic vulnerability that companies and investors need to take seriously.”

‘Had to go back to talking’

The Outlook outage caused headaches for a wide range of workers. A spokesperson for the Seattle mayor’s office said Thursday that while the city government didn’t grind to a halt, the primary system for documents and other information was offline.

For some local governments, the outage struck as officials were working on next year’s budget. 

It’s “a pretty intense time for us,” said Everett City Councilmember Paula Rhyne. “There’s a lot of questions and communications that need to happen before budget votes, so to have all of that on pause during a really critical time is also really difficult.”

The outage also affected Rhyne’s ability to respond to constituents’ questions or concerns. “I can’t even pull up their old email to see if they left a phone number that I could give them a call,” Rhyne said. 

King County Superior Court spokesperson Amy Roe said not having Outlook available didn’t have much effect on trials and hearings already in progress. But it disrupted the Ex Parte department, where people contact the court seeking civil protection orders and other emergency relief.

Sponsored

“These orders are often sought by people who fear for their personal safety and may be heard by the court on the same day or in a matter of hours,” Roe said. “Petitioners typically file them electronically, so email is part of the process. Because this wasn’t possible, it was necessary to communicate over the phone.”

Roe said communicating sensitive personal information over the phone rather than through a digital form or email is not ideal.

“We don’t yet know the full impact of the outage, but we are concerned for court users,” Roe said. “Their matters are always highly consequential and often urgent. When people are hindered in their ability to reach us in the ways they are used to, it affects their access to justice.”

Critical operations at other agencies were spared, even if quick communication was hindered. 

At the state Employment Security Department, which runs Washington’s unemployment insurance program, the biggest outage-related problem was that some customers “couldn’t reset their passwords” on their accounts, said agency spokesperson Chris Barron.

Customers could still contact the agency via phone and through its online portal, he said. With a relatively short outage, “people can adjust,” Barron said. 

“If it happened on a Sunday or Monday when people were filing their weekly unemployment claims or claims for paid leave, that might’ve been another situation where we saw larger customer impacts,” he said.

Many firms have contingency plans for communication breakdowns. 

When the outage interrupted client communications at Milliman, a Seattle-based actuarial and consulting firm with offices worldwide, the company’s “business continuity plan” pointed employees to a simple solution: the phone. 

“We had to go back to talking to people, and that worked,” said Craig Reynolds, a principal and consulting actuary.

Reynolds, who is also on the Mercer Island City Council, said Milliman, like most companies, routinely shares files via email. Had the outage lasted longer, Reynolds said Friday, “we would have had to look at other options.”

King County sheriff spokesperson Brandyn Hull said the outage didn’t affect policing, and department office workers still had access to Teams.

“Honestly, other than email, we were business as usual, and we are back up,” Hull said on Friday. 

UW Medicine spokesperson Susan Gregg said operations at hospitals managed by UW Medicine, including Harborview Medical Center, weren’t disrupted. 

As residents in Asia, Europe and Africa began rising for the day Friday, Outlook systems still weren’t fully online. 

Britain’s National Health Service said early Friday it was “working to ensure critical services and appointments continue for patients, despite the global Microsoft issue having a disruptive impact.” The European Court of Human Rights and the Supreme Court of the Philippines also warned of email disruptions and delays.

As morning arrived in the U.S., Microsoft’s updates looked more positive, and the company said it was slowly bringing email service back. Full restoration of emails sent Thursday may lag as the company aims for a gradual and stable recovery.

Microsoft said early on that it was aware of the issues and had begun investigating. The company provided updates throughout the day Thursday as it scrambled for a fix. Microsoft initially determined that overloaded servers may have contributed to the problem, but later said it wasn’t because of extraordinarily high traffic. Servers, which undergo automated health checks, were failing and dropping offline, straining the remaining infrastructure.

Microsoft Philanthropies underwrites some Seattle Times journalism projects.