Microsoft Restores 365 Services After 24-Hour Global Network Failure

Microsoft began restoring access across its cloud productivity suite on Friday following an extensive disruption that degraded core applications including Outlook, Teams, and Copilot for more than 24 hours.

In a statement reported by Wired, Microsoft attributed the failure to internal network infrastructure rather than malicious activity. The company stated that the disruption was introduced during an earlier network capacity upgrade and surfaced following a subsequent network configuration change. Microsoft explicitly ruled out cyberattacks as well as issues related to its artificial intelligence agents, noting that it had remediated the underlying defect and was continuing recovery across remaining regions.

The failure began shortly after midday Eastern time on Thursday, initially presenting as an Outlook disruption in North America before expanding to affect organizations in more than 70 countries and territories. Downdetector recorded more than 130,000 incident reports on Thursday afternoon as Exchange Online dependencies failed, interrupting communications across commercial and personal accounts drawn from Microsoft's base of more than 430 million users.

The outage triggered wide operational friction across technology, enterprise, and public sector infrastructure, according to reporting from CBS News. Artificial intelligence developers OpenAI and Anthropic each observed service disruptions when connecting to Microsoft 365 tools, while commercial vehicle manufacturer Workhorse Group postponed its quarterly financial results conference call. Government administrative bodies in Singapore, the Philippines, and Japan encountered communication bottlenecks, and carriers including Air India and IndiGo temporarily rerouted or adjusted customer support channels due to email delays.

During a press briefing in Sydney, Australian Home Affairs Minister Tony Burke stated that Microsoft had briefed domestic cybersecurity officials, describing the disruption as an update failure and drawing comparisons to the July 2024 CrowdStrike software incident that disabled enterprise endpoints worldwide.

To mitigate permanent message loss, Microsoft extended its internal mail queue retention window to 48 hours to ensure correspondence initiated during the downtime delivers as services stabilize. Independent security researcher Jonathan Leitschuh observed that users should expect staggered delivery spikes and concentrated notification batches as downstream queues complete processing.

Microsoft has reported ongoing recovery across North America, Europe, and the Asia-Pacific region, while cautioning that customers may experience intermittent latency as data traffic balances out. The company indicated it will publish a comprehensive post-incident technical review once internal investigations conclude.