A sudden and severe outage at Cloudflare, one of the world’s largest web infrastructure giants, triggered a domino effect this afternoon, crippling a swath of popular online platforms including ChatGPT, X (formerly Twitter), Spotify, and Canva. The incident served as a stark reminder of the modern internet’s fragile dependence on a concentrated set of backend providers.

The disruption, which began around midday for many users, was linked to a critical failure within Cloudflare’s global systems. The company provides a core suite of services including content delivery networks (CDN), Domain Name System (DNS) resolution, and security firewalls that millions of websites rely on to manage traffic, block threats, and ensure speedy performance.

For end-users, the failure manifested in a variety of frustrating ways. Some found themselves completely locked out of services, facing error messages and inaccessible login pages. Others experienced severely delayed loading times and intermittent functionality, making platforms unreliable for the duration of the incident.

The list of affected services read like a who’s who of the digital world. Alongside headline acts like ChatGPT and X, collaboration tool Discord, design platform Canva, learning site Udemy, and AI search engine Perplexity AI were all reported to be experiencing issues. Even the popular outage tracker Downdetector briefly went offline, overwhelmed by a surge in traffic from users seeking answers.

The incident underscores a critical vulnerability in the architecture of today’s internet. While the web appears vast and distributed, a significant portion of its traffic is funnelled through a handful of major infrastructure providers like Cloudflare, Amazon Web Services (AWS), and Google Cloud. A single point of failure at any of these entities can have catastrophic, global consequences, echoing the significant AWS outage that disrupted services just last year.

Cloudflare officially acknowledged the issue on its status page, stating that its engineering teams were actively working to restore services. As technicians raced to diagnose and rectify the root cause, users began reporting a slow and partial recovery across various regions. However, the company warned that full system stabilisation could take time, with the potential for sporadic issues to persist.