Infrastructure provider Cloudflare confirmed it has resolved a widespread outage that caused major platforms—including ChatGPT, X (formerly Twitter), and Claude AI—to go offline on 18–19 November 2025.
What-Happened: Timeline & Major Impact
According to Cloudflare’s status updates, the incident began when an internal change triggered a latent bug in its bot-mitigation service. This then cascaded, leading to abnormal database behaviour and widespread errors across its network. :contentReference[oaicite:5]{index=5}
- The disruption started around 06:30 ET (≈11:30 UTC) on 18 November when the database permission change led to a “feature file” doubling in size, triggering the fault.
- Platforms affected included ChatGPT, X, Perplexity, Canva, League of Legends, Genshin Impact and many more.
- Cloudflare acknowledged the fix later on 19 November and stated that service levels were returning to normal.
Why This Matters
When a single infrastructure provider like Cloudflare goes down, the ripple effects are global. As noted by analysts at the Cornell Tech Policy Institute, this outage underscores how even the most innovative platforms are only as resilient as the network underneath them. :contentReference[oaicite:14]{index=14}
Key Takeaways for Businesses & Developers
- Audit third-party dependencies: services layered on top of large providers may still carry unseen operational risk.
- Design fallback strategies: multi-CDN or multi-provider routing can reduce single-point failures.
- Monitor infrastructure metrics beyond your own stack: upstream errors may originate outside your control.
Next Steps from Cloudflare
Cloudflare’s CTO, Dane Knecht, publicly apologized, stating: “Earlier today we failed our customers and the broader Internet… I’m truly sorry for the impact that we caused.” :contentReference[oaicite:16]{index=16} The company also announced it will publish a full incident retrospective and strengthen its bot-mitigation systems.
Recommended from PromakAI News
Source: Cloudflare. Compiled by PromakAI News.