Key points
- Microsoft 365 outage began around noon ET Thursday, with Outlook reports on Downdetector, and spread to more than 70 countries, per Wired.
- Microsoft says a network capacity upgrade and a later network change caused it, and it was not a cyberattack or AI agents.
- Email queue time was extended to 48 hours so delayed messages are delivered; a full post-incident report is promised.
- The Rundown reports OpenAI and Anthropic saw broken 365 connectors, and that Exchange Online-based features such as Copilot were hit.
Why it matters for marketers: Email, Copilot and connected AI tools shared one point of failure, so campaign sends, approvals and agent workflows built on Microsoft 365 need a fallback plan.
Microsoft 365 suffered a global outage that began around midday Eastern time on Thursday and ran into a second day. Wired reports that Outlook problems first showed up on Downdetector just after 12 pm ET, and the disruption grew from North America to more than 70 countries.
Microsoft says the cause was a technical issue tied to an earlier network capacity upgrade that surfaced after a recent network change. A spokesperson said it was not related to cyber or AI/agents. The Rundown AI reports recovery was uneven: servers restored overnight failed again under traffic, and Microsoft began rolling back a network change on Friday. Per that report, the failure centered on Exchange Online and features built on it, including Copilot, SharePoint and Defender XDR, while core Teams chat and meetings mostly kept working.
For marketers, the practical effect is delayed communication. Microsoft extended the email queue window to 48 hours so messages sent during the outage are delivered rather than bounced. Independent researcher Jonathan Leitschuh warned that users may see a flood of notifications as subsystems come back. Expect late replies, stale approvals and bunched-up send times for campaigns and client work.
The Rundown also reports that OpenAI and Anthropic saw broken Microsoft 365 connectors, which shows how agent workflows inherit the weaknesses of the systems they plug into. Teams that route content review, scheduling or reporting through Microsoft 365 should check what stalled and identify a backup channel. Microsoft has promised a full post-incident report, and that document is the thing to watch for a confirmed root cause.
Sources
This brief is our own summary. Read the full reporting at the source.