AI service outages have become high-stakes quickly — a ChatGPT or Claude outage now disrupts developers, writers, support teams, and whole product workflows. This page tracks the main AI platforms, their current reachability from our probe, and any recent incidents reported by users or flagged on the provider's own status page. We don't rate-limit this page and we don't cache it aggressively — the data you see is within 30 seconds of live.
AI services tend to fail in two ways: capacity (rate limits or queue backlog during peak load) and infrastructure (model-server incidents, authentication outages). Capacity problems show up as slow or intermittent responses; infrastructure ones as full unreachability.
If you depend on an AI service in production, set up a WebsiteDown monitor on the provider's API endpoint — the probe will tell you when availability returns before the provider updates their status page.
If your product depends on one of these services, set up a WebsiteDown monitor and get an email when the probe flips. Free plan includes 1 monitor with 5-minute checks.