Anthropic ChatGPT Outage
An outage affecting the Anthropic platform, specifically the ChatGPT service. No official cause or resolution has been provided.
since
Here's why. Incidents from Friday, September 4, 2026 (UTC), ranked byBreakage Velocity.What is Breakage Velocity?A live score of how fast an outage or regression is spreading. The more people hit it, and the faster, the higher it climbs; it cools off as things calm down.Open-ended scale: under 25 low · 25+ elevated · 100+ high · 300+ extreme.
8 outages, 1 still active today. Worst: HashiCorp (<1 min).
An outage affecting the Anthropic platform, specifically the ChatGPT service. No official cause or resolution has been provided.
since
Errors occurred when creating new VCS workspaces with the Bitbucket cloud provider in HashiCorp Configuration Manager (HCP). The issue has been resolved, but users are advised to open a ticket if they continue to experience problems.
– UTC
Customers are experiencing increased latency for API requests to VCR, Blob, and Sandbox. The issue is currently being investigated and a fix is in the rollout phase.
– UTC
Network issues affecting CircleCI's Mac infrastructure have caused job delays. Customers may experience queuing during this time, and we are working with our provider to resolve the issue.
– UTC
The npm audits endpoint is experiencing issues, affecting all users. The issue was resolved after a fix was implemented.
– UTC
Between 10:32 UTC and 10:36 UTC, increased errors affected traffic on the High Performance Edge Network in the FRA region due to unexpected traffic spikes. Error rates have since returned to normal.
– UTC
Users in the APAC region experienced increased error rates across various services including ChatGPT, Work, image generation, file upload, Voice, and Codex Cloud. The issue has been resolved.
– UTC
Error rates for Supabase Project Lifecycle Actions have increased across all regions and are now back to normal. The incident was resolved on September 4, 2026.
– UTC