Claude, the AI assistant from Anthropic, experienced a worldwide outage on Friday, adding to the chaos already affecting Microsoft 365 services like Outlook and Teams. Users reported login failures and degraded performance, while Microsoft investigated its own widespread issues.
What Happened: Claude's Global Outage
According to Anthropic's status page, claude.ai and the Claude Cowork agent tool showed degraded performance on Friday. Many users took to social media to report problems signing in and using the service. The API, developer console, and Claude Code appeared operational at the time of reporting.
The most documented failure involved the Microsoft 365 connector. Anthropic stated that since Thursday, its connector for Microsoft services had been failing most requests to email, calendar, and search tools. The Claude plugin for Outlook was also affected.
Microsoft 365 Services Also Down
Microsoft's services had been down for several hours, with the company investigating. Azure, Microsoft 365, and other Microsoft services heavily influence the operations of many companies, including Anthropic. The interconnected nature of these platforms means that when one fails, others can quickly follow.
At the moment, the only recourse for users is to monitor the status pages of both Anthropic and Microsoft 365 for updates. Anthropic later updated that Claude had been restored and was working correctly, though some regions might still experience issues.
Why This Matters: The Ripple Effect of Cloud Dependencies
This incident highlights the fragility of cloud-dependent workflows. Many businesses rely on Microsoft 365 for communication and collaboration, and increasingly, they integrate AI assistants like Claude to enhance productivity. When either service falters, the impact can be significant.
The outage also underscores the importance of resilience planning. Companies should have contingency plans for when critical services go down, including alternative communication channels and backup AI tools. As AI becomes more embedded in daily operations, the stakes of such outages rise.
Key Takeaways for Businesses
- Monitor status pages: Regularly check the status pages of your critical service providers.
- Diversify tools: Avoid over-reliance on a single vendor for essential functions.
- Have a backup plan: Establish alternative workflows for when primary services fail.
- Communicate transparently: Keep employees and customers informed during outages.
- Review SLAs: Ensure service level agreements cover outage compensation and support.
Comparing Service Status During the Outage
| Service | Status | Impact |
|---|---|---|
| Claude.ai | Degraded | Login and usage issues |
| Claude Cowork | Degraded | Agent tool performance |
| Claude API | Operational | No reported issues |
| Claude Code | Operational | No reported issues |
| Microsoft 365 Connector | Failing | Email, calendar, search requests |
| Outlook Plugin | Affected | Claude integration in Outlook |
| Azure | Investigating | Potential widespread impact |
What to Watch Next
Both Anthropic and Microsoft are actively investigating the issues. Users should keep an eye on their respective status pages for the latest updates. Anthropic has already reported that Claude is restored, but regional issues may persist. The background to this story is set out in Microsoft 365 Outage Hits Outlook and Teams Users Worldwide. A related thread runs through Microsoft 365 Outage: Outlook and Teams Down Worldwide. Earlier coverage in Microsoft 365 Outage: Global Disruption and Recovery Timeline shows how the argument developed.
For businesses, this outage serves as a reminder to build redundancy into their systems. As AI tools become more critical, ensuring that you can operate without them—at least temporarily—is essential. Consider diversifying your AI providers and maintaining fallback options. A closer look at the same subject appears in Microsoft 365 Outage Sends Australia Back to Pre-Email Era. Anyone following the wider picture can start with Lousy Boss Doesn’t Like To Pay Employees, Gets Reported To Labor Board, Ends Up Paying More. The same theme surfaces in Office EU: European Alternative to Microsoft 365 and Google Workspace, which takes a different angle.
Additionally, this event may prompt discussions about service reliability and the need for more robust infrastructure. As AI continues to evolve, so too must the systems that support it. Context on how this played out before sits in Why centralized Microsoft 365 email signatures save more time than. A companion piece, Anthropic Launches Claude Science for Laboratory Research, approaches the same subject from another angle.
FAQ
What caused the Claude outage?
The exact cause is not yet known, but Anthropic reported that its Microsoft 365 connector had been failing most requests since Thursday, affecting email, calendar, and search tools. The Claude plugin for Outlook was also impacted.
Is Claude working again?
Anthropic updated that Claude has been restored and is working correctly, though some regions might still experience issues. Users should check the status page for the latest information.
How does this affect Microsoft 365 users?
Microsoft 365 services like Outlook and Teams were already down, and the Claude outage compounded the disruption. The Claude connector for Microsoft services failed, preventing users from accessing email and calendar through Claude.
What should businesses do during such outages?
Businesses should monitor status pages, have backup communication channels, and consider diversifying their AI tool providers to reduce dependency on a single service.