So Outlook died yesterday afternoon and apparently just… stayed dead. 18 hours and counting. Outlook, Teams, SharePoint, Copilot, Defender — the whole Microsoft 365 stack, in 74 countries, something like 140,000 Downdetector reports just for Outlook.
The wild part isn’t even the outage itself, it’s the blast radius. The US House of Representatives, Australia’s spy agency and federal police, Japan’s government IT — all hit. A Taiwanese firm had to cancel a bond deal it already scheduled. And here’s the one that got me: Claude’s and ChatGPT’s Microsoft 365 connectors went down too. The models were fine, they just couldn’t reach your work data anymore. Your AI assistant was up, your email was not.
Microsoft says it’s not a cyberattack — some servers went unexpectedly unavailable, a core process for secure connections got overwhelmed, and the rest buckled under the load. They reverted a network config change, started rerouting traffic, and as of this morning things are slowly coming back, but there’s still no full root cause and no ETA on full restoration. They’ve bumped the email timeout to 48 hours so delayed messages don’t bounce while the backlog clears.
Honestly, the thing that keeps nagging me is how single-point-of-failure this all is. One vendor’s email platform goes down and governments, hospitals, schools, and half the companies on earth just… can’t work. We all know this is the deal we’ve made, but every time it happens it still feels insane.
Did it hit you yesterday? Were you one of the lucky ones on Gmail, or were you staring at a spinning Outlook like the rest of us?
