Friday the 13th delivers tech horror show
A Friday-the-13th technology pile-up unfolded overnight as a Microsoft Exchange Online outage disrupted Outlook and Microsoft 365 connectors, triggering intermittent failures in Anthropic’s Claude and OpenAI’s ChatGPT – although Microsoft ruled out a cyber attack.
The issues started around 6pm SAST last night when users began experiencing problems accessing their mailboxes through Outlook on the web, as well as desktop and mobile apps, while some e-mail messages were also delayed.
Microsoft subsequently reported a wider problem affecting access to multiple Microsoft 365 services.
Tom’s guide, at 6.30am, was saying that “something is seriously wrong with Microsoft Outlook and other popular Microsoft services like Teams, Microsoft 365, Copilot and others, as more than 100 000 users are flocking to the popular outage-reporting website Downdetector to report issues”.
The website adds that this number has “dropped a bit since then, but it’s still a very widespread outage”, while Downdetector, which counts user-submitted problem reports for each service rather than users, showed 908 reports for Microsoft 365 as of 9am.
Traffic jam
Microsoft says the outage persisted for more than 12 hours because traffic was redirected away from servers that had become unavailable, placing additional pressure on the remaining operational infrastructure. The increased load overwhelmed those servers, which also began intermittently failing to process requests and complicated Microsoft’s efforts to restore the service.
Microsoft had redirected some traffic away from affected infrastructure in the US to reduce pressure, identified the servers that were intermittently becoming unavailable, and then started to redistribute Outlook traffic across its global network, says the University of Pennsylvania’s Penn Computing.
The software company says it is pursuing multiple workstreams to pinpoint the underlying cause of the traffic influx and advance the recoverability of the backend servers. “We’re taking a measured approach to identify and repair affected infrastructure.”
Microsoft initially reported positive results from restarting affected infrastructure, but at 5am it said the changes intended to improve service availability were not working as expected and that it was looking at alternative ways to increase capacity, while dealing with imbalances in how traffic was being managed.
“As of now, we are not aware of any malicious activity that is causing this incident. We will continue to treat this incident with urgency and priority until full resolution,” Microsoft said in its latest update as of the time of writing at 8am.
Microsoft adds that, “once mitigated, we will provide full details of the root cause, defence-in-depth repair items, and more within a Post-Incident Report”.
- Nicola Mawson, Contributing journalist, iTWeb
By this year, attackers were using AI to scale and accelerate cyber crime, which extends from generating code and automating attacks, to crafting convincing phishing and deepfake scams. The AI Incident Database lists more than 7 000 incidents in which AI was used as a hacking tool.


