<!-- Search description: A Microsoft outage disrupted Outlook, Teams, Copilot and AI connectors in ChatGPT and Claude for over 12 hours. What happened and what it exposed. --> <h2>Microsoft Outage Knocked Out Outlook, Teams And Copilot, And AI Assistants Felt It Too</h2> <p style="text-align: justify;">A major Microsoft outage disrupted Outlook, Teams, Copilot and linked AI tools for more than 12 hours from Thursday into Friday, November 13, 2026, showing how one failure inside the world's biggest workplace software provider can reach everything from email to AI assistants and corporate deals.</p> <img alt="Dark server room with rows of racks and blue lights, illustrating the Microsoft cloud outage that hit Outlook, Teams and AI connectors" src="" style="display: none;" /> <h2>What caused the Microsoft outage?</h2> <p style="text-align: justify;">The disruption began around 16:17 UTC on Thursday inside Exchange Online, the cloud service that underpins Outlook email. Microsoft found that some servers stopped responding, pushing their workload onto the machines that were still running, and when engineers restarted those machines they became unresponsive again under the renewed traffic, according to reporting carried by Invezz. Mark Gregory, an engineering academic at RMIT University, told The Guardian that the result is a cascading problem, where overloaded surviving servers fail in turn. Microsoft said it had found no indication of malicious activity, and its full explanation of the root cause is still pending.</p> <h2>How did the Microsoft outage reach ChatGPT and Claude?</h2> <p style="text-align: justify;">The AI assistants themselves stayed online, but the parts that connect to workplace data did not. OpenAI warned that new Microsoft connector connections in ChatGPT might fail and their associated tools might not work correctly. Anthropic reported failures affecting its Microsoft 365 connector's email, calendar and search functions, along with Claude for Outlook. In other words, the outage hit access to linked workplace information rather than the underlying chat models.</p> <h2>How bad was the disruption?</h2> <p style="text-align: justify;">Downdetector recorded more than 130,000 US reports of Outlook problems at the peak, with users reporting missing emails, failed logins and locked-out workplace services, according to Invezz. Government organisations in Australia, Japan and the United States reported difficulties, and the damage spread well beyond inboxes: Indian airlines IndiGo and Akasa Air said customer email support was disrupted, Reuters reported problems at Taiwanese financial institutions including a company that cancelled a planned bond repurchase transaction, and US electric-truck maker Workhorse postponed its quarterly earnings call to November 16, citing the outage directly.</p> <h2>Was the Microsoft outage a cyberattack?</h2> <p style="text-align: justify;">No. Microsoft said it had found no indication of malicious activity, and its engineers traced the failure to unavailable servers and overloaded processes handling connections inside Exchange Online.</p> <h3>Did ChatGPT and Claude stop working?</h3> <p style="text-align: justify;">The chat models stayed online. What failed were Microsoft connectors: OpenAI warned that new Microsoft connector connections might fail, and Anthropic reported email, calendar and search failures in its Microsoft 365 connector and Claude for Outlook.</p> <h3>Is Microsoft fixed now?</h3> <p style="text-align: justify;">Microsoft reported partial recovery, but Outlook and Teams issues persisted into Friday. Microsoft stock closed Thursday at $523, up 0.36 percent, suggesting investors did not immediately treat the disruption as a lasting financial problem.</p> <h2>What the outage says about depending on one provider</h2> <p style="text-align: justify;">Greyhound Research chief analyst Sanchit Vir Gogia told CIO that enterprises have priced Microsoft 365 as a licence and never as a single point of failure, and argued that companies need practical contingency arrangements rather than assuming one provider's services will stay continuously available. The episode is a reminder that as AI assistants plug deeper into workplace systems, a communications outage can now stall financial decisions and investor announcements too.</p> <h2>Further reading</h2> <p style="text-align: justify;">Read our related coverage: <a href="https://www.bhginfodesks.com/2026/11/anthropics-new-regional-rules-take.html">Anthropic's New Regional Rules Take Effect Today: Who Can Still Use Claude</a></p> <p style="text-align: justify;"><b>Sources:</b> Invezz (invezz.com), The Guardian (theguardian.com), Reuters (reuters.com), CIO (cio.com), Microsoft service status updates</p>
0 Comments