Cloudflare’s Network Failure Slams ChatGPT, X, and More

On the morning of November 18, 2025, a major Cloudflare outage disrupted a huge part of the internet. Many users saw error pages, had trouble logging in, or couldn’t reach dashboards — even though Cloudflare later said the issue was resolved. The company’s network supports about 20% of global web traffic, which meant the outage had a wide impact.

The problem started around 6:40 a.m. ET when Cloudflare detected “internal service degradation.” This followed a sudden spike in unusual traffic hitting one of their services, which triggered cascading errors in their core systems. Cloudflare’s CTO admitted that a configuration change in their bot-mitigation system caused a latent bug that escalated into broader network failure. He stated clearly that this was not a cyberattack.

By later in the day, Cloudflare had deployed a fix and said the incident was under control, but they continued to monitor for lingering problems.

Because Cloudflare handles content delivery and security for so many web platforms, the outage rippled across the web. Reported affected services included big names like ChatGPT (and other AI tools), the social-media app X (formerly Twitter), design and productivity sites like Canva, and gaming platforms.

Some users, however, said things still weren’t fully back to normal: slow-loading sites, errors when saving drafts, redirect loops, and login issues persisted on dashboards and mobile apps.

The incident is being viewed as a stark reminder that even core internet infrastructure remains fragile. For many experts, this outage highlights the danger of relying too much on a single provider for performance and security.