Back To The Fax Machine: Microsoft's Meltdown Exposes The Agentic Economy's Weakest Link

Traders reached for fax machines, Japan's finance ministry pinned paper notices to the wall, and the AI assistants meant to automate the office went dark right along with the inbox.Hang in there clippy...For more than 18 hours, the AI agents Silicon Valley says will soon do your job couldn't read your email, because nobody could. A Microsoft 365 failure that began Thursday knocked out Outlook for businesses and governments in more than 70 countries and took Microsoft's own Copilot down with it, along with the Microsoft connectors that Anthropic's Claude and OpenAI's ChatGPT use to reach users' mail and calendars.Every one of those failures has one thing in common: Exchange Online, Microsoft's cloud email and calendar backend. In its status notice Friday morning, Microsoft said features across its 365 services "that interact with Exchange Online and email services are impacted," while "elements of those services that do not depend on interaction with Exchange Online remain functional." Teams calls and group chats mostly kept working. Anything that had to touch a mailbox did not.What Went Down With Exchange OnlineServiceWhat brokeOutlook / Exchange OnlineMailbox access through Outlook on the web, desktop and mobile sync; delayed email deliveryMicrosoft 365 CopilotCopilot Chat failed to load; Copilot cut off from email, calendar and filesTeamsCalendar, invites and meeting edits failed, along with Copilot answers about Outlook and filesMicrosoft Graph APIRequests for mail, calendar and contacts timing outClaude (Anthropic)Microsoft 365 connector's mail, calendar and search tools failing for most requests; Claude for Outlook add-in wouldn't open for someChatGPT (OpenAI)Errors in its Microsoft 365 connectorsDefender for Office 365 / PurviewDelays or failures in data-protection and policy enforcementPlanner, To Do, IntuneTask failures; device enrollment and compliance checks hitStill standingTeams calls and chats; files stored directly in SharePointSource: Microsoft service status updates of November 13, Anthropic status, FinanceXAs we reported Thursday, Microsoft first blamed elevated load on part of its infrastructure and said it was reviewing "an unexpected increase in a specific type of service traffic" to see whether it was contributing. It then ruled that out, saying the load "isn't driven by an increase in traffic" but by a large number of servers dropping out after failing automated health checks, which dumped their traffic onto the machines still running. Overnight, Microsoft warned that servers it had brought back "can become unresponsive again once they receive traffic," and several did. It paused all deployments, patched an authentication and routing component and stretched its email timeout window to 48 hours so delayed messages wouldn't be lost.By 9 a.m. ET Friday, Microsoft said email queues were "significantly reduced and approaching normal levels," though some users could still hit intermittent access problems or delayed mail. It says it has seen no evidence of malicious activity and has promised a full post-incident report with "root cause and defense-in-depth actions" once mitigation is complete. Microsoft executive Charles Lamanna said as much early Friday:We regret that today's outages caused disruptions for our customers, and we are working to address the issue. We will continue to share more information as soon as we have it, and for now I can confirm that there is no indication this is the result of malicious or cyber activity.— Charles Lamanna (@clamanna) November 13, 2026The Agents Went Down With The Inbox"Copilot experiences may be unable to access work data such as email, calendar, and files," Microsoft's status notice said Friday of its own AI assistant, and some users couldn't load Copilot Chat at all. Inside Teams, Copilot answers about Outlook, calendars and documents failed.The same notice says requests to Microsoft Graph that touch Exchange Online data, "such as mail, calendar, and contacts, may time out." Graph is the API outside software uses to read and act on a company's Microsoft 365 data, the usual on-ramp for third-party AI assistants wired into a corporate inbox. Even some Google Workspace customers who read their mail through Outlook's desktop and mobile apps reported trouble.Image via windowslatest.comAnthropic users were hosed too - opening an incident titled "Elevated errors affecting the Microsoft 365 connector" at 12:52 p.m. ET Thursday, around the time Microsoft posted its first public alert. By evening, its status page said the connector's mail, calendar and search tools were "still failing for most requests," that some users "may not be able to open" the Claude for Outlook add-in, and that the cause was "a Microsoft 365 service incident." OpenAI also reported errors affecting its Microsoft 365 connectors, FinanceX reported, and Calcalist said both companies had trouble reaching Microsoft services.Microsoft opens its Ignite conference in San Francisco next week with a program built around Agent 365, its system for managing fleets of AI agents across a company's apps and data. As we reported in February, Microsoft AI chief Mustafa Suleyman told the Financial Times that "most of the tasks will be fully automated by an AI within the next 12 to 18 months." Nine months later, Japanese firms were trading by fax.Japan's Finance Minister Satsuki Katayama speaks to media in Tokyo, Japan, April 21, 2026. REUTERS/Kim Kyung-HoonPhones, Fax Machines And Paper NoticesDesks fell back on phones, WhatsApp and fax machines, Bloomberg reported: Deutsche Bank's email access was patchy, traders in Singapore said some banks couldn't close financing deals, and Euronext and Nasdaq both reported email problems. The outage halted at least one bond repurchase in Europe, according to Reuters, while bankers at Goldman Sachs and Morgan Stanley in Hong Kong told CNN they couldn't reach email or Teams from the moment they signed on.At Goldman itself, TMT specialist Sean Johnstone opened his Friday morning note with "Paraskevidekatriaphobia meets its modern upgrade," a nod to the fear of Friday the 13th, before describing the floor: "MSFT here is still down and most operating off backup email so reading email has never felt more mythological... or more dependent on tech that clearly didn't get the memo about not being the unlucky guest."Cartier owner Richemont delayed its first-half results, due at 0600 GMT Friday, before publishing later in the morning. Electric truck maker Workhorse pushed its results to Nov. 16, British insurer Aviva set up backup arrangements for its earnings calls, and Maersk warned customers that email updates could be delayed.In Tokyo, the finance ministry couldn't receive email for most of the day, and ministries pinned up paper notices to give journalists the day's schedule. Chief Cabinet Secretary Minoru Kihara confirmed that communications between ministries were disrupted. In Australia, the spy agency ASIO and the Australian Federal Police were among those hit, Service NSW centres in regional New South Wales closed and the state's online bail court was disrupted. Tony Burke, Australia's home affairs and cyber security minister, said Microsoft itself likened the disruption to the 2024 CrowdStrike outage.Singapore's government email went down, the Philippines' Supreme Court said some operations were affected, and New York City Emergency Management warned that email to city agencies could be delayed. In Toronto, the Unity Health hospital network declared a code grey, its signal for an infrastructure failure, while saying patient care was not affected. Police forces, hospitals, GP surgeries and councils across southern England posted workarounds, and Air India rerouted its email while Alaska Airlines and IndiGo said their email-based customer service channels were hit.Outage reports peaked at 19,650 a minute across five Microsoft services on Downdetector, against fewer than 15 on a typical Thursday. Meanwhile, Iran war hawk fossil Mark Levin is still using AOL.My AOL email is back and working great. How's it going with Microsoft Outlook email users?— Mark R. Levin (@marklevinshow) November 13, 2026The Scam WindowThe failure also took out part of the security layer around the inbox. Standard Chartered and ING Australia warned customers about scammers posing as tech support, noting that the email security filters that normally catch such messages were disrupted too. Microsoft's notice says some Defender for Office 365 features that rely on Purview data-protection signals face "delays or failures in policy evaluation and governance action enforcement."Organizations that run Microsoft services on their own servers rather than in Microsoft's cloud came through better: Reuters reported they were "somewhat spared."The outage follows a Microsoft 365 failure of nearly 10 hours on Jan. 22. "The bigger worry is that this is not the first Microsoft 365 outage this year," eToro analyst Josh Gilbert told Reuters. Sorry Clippy...