Microsoft 'Deeply Sorry' as Global Outage Hits 70+ Countries and Recovery Begins

At 17:54 GMT on Thursday, Microsoft acknowledged that users in the Americas could not get into their Exchange Online mailboxes. Shortly afterward it warned that Microsoft 365 might be affected worldwide. Nearly 24 hours later, on Friday afternoon, services were broadly coming back, and the company was apologising.

"We are deeply sorry for the disruption this incident has caused our customers and partners," a Microsoft spokesperson said, adding that the company knows the trust placed in its services and takes that responsibility extremely seriously. Microsoft described "encouraging progress" but said it had no estimate for when everything would be fully restored. A full post-incident report will follow once its review is complete.

The company's account of the cause is specific. It says a technical issue was introduced during a network capacity upgrade, and that the outage was not connected to a cyberattack or to artificial intelligence agents. The recovery was deliberately gradual, it said, because of the scale of the problem and to keep services stable once restored. By 04:36 on Friday it reported only a partial recovery in some regions. By 13:32 it described broader improvements in North America, sustained improvements across much of Europe, and encouraging recovery signals in Asia Pacific.

The reach was wide. The outage touched more than 70 countries, and over 430 million people use the affected products. Institutions hit by Friday included the Japanese government, the Philippines Supreme Court and the Australian spy agency, along with trains, hospitals and courts in North America. Singapore's government acknowledged some of its systems were affected, and work was disrupted for many people in India. At the COP31 climate summit in Antalya, Turkey, advisers and officials were frustrated at being unable to open Outlook.

In the United States, House of Representatives staff told Politico that the outage set back a full day of legislative work, because everything ran through Outlook. In Canada, the national rail network warned that passengers might not receive emailed updates about their trips, and hospitals across the country had email problems.

In the UK, some parts of the NHS, including GPs and hospitals, were affected, as were some schools and police forces. Cyber minister Liz Lloyd said the impact on critical frontline services such as 999 calls was "limited" and that the transport network was not disrupted.

Smaller businesses and individuals felt it too. Peter Pentz, who runs a winery on South Africa's Cape West Coast, said his entire business was effectively brought to a standstill. Communication with exporters stopped, and what he called critical production work was halted because the winery relies on these systems for parts of its daily operations. In Staffordshire, a woman named Georgia was due to complete her first house purchase on Friday after eight months of waiting. Her solicitor warned of a likely delay because other agencies could not be reached. She noted that hundreds of other purchases were completing the same day, making it unfeasible for everyone to be contacted in time.

Experts read the day as a warning. Rob Demain, chief executive of cybersecurity firm e2e-assure, called it a stark reminder of how heavily the world relies on a handful of big tech providers. Matt Saunders, field chief technology officer at Adaptavist, said classic disaster planning suggests using a second service for resilience, but in practice it is almost impossible for a business to function without tools like Microsoft's. Organisations, he said, need a plan for working without them.

What to watch next is Microsoft's post-incident report, which should say more about how the upgrade fault happened, and whether it offers any firm timeline for full recovery.