Some services are experiencing issues
Increased Error Rates - Bahrain
We are investigating increased API error rates in a single Availability Zone (mes1-az2) in the ME-SOUTH-1 Region.
Increased Error Rates - UAE
We are investigating issues with AWS services in the ME-CENTRAL-1 Region.
Past Incidents ‹ May 05 - May 11, 2026 ›
May 11, 2026
Elevated error rates with GPT 5.5
We are investigating the issue for the listed services.
May 10, 2026
Users may experience elevated errors in ChatGPT uploading files and Codex Cloud creating tasks
We are investigating the issue for the listed services.
May 09, 2026
May 08, 2026
Degraded Performance with Codex Cloud Tasks
We are observing an issue with follow-up prompts to Codex Cloud tasks failing at higher than expected rate. We are investigating the issue.
Increased Error Rate for gpt-5.5 model in the API
We are investigating the issue for the listed services.
May 07, 2026
Increased Error Rate and Latency - N. Virginia
We are investigating instance impairments in a single Availability Zone (use1-az4) in the US-EAST-1 Region. Other Availability Zones are not affected by the event and we are working to resolve the issue.
Perplexity Website and API incident
A small percentage of our users may experience instability while using Perplexity Website and products. We're investigating the incident.
Elevated transcription failures affecting ChatGPT & Codex
We have identified that users are experiencing elevated errors for the impacted services. We are working on implementing a mitigation.
Increased error rate with image generation in the API
We are investigating the issue for the listed services.
May 06, 2026
Connection failures for organizations restricting GitHub access by IP address
We have identified an issue affecting organizations that restrict GitHub access by source IP address. A recent infrastructure change altered the IP addresses Anthropic uses for outbound connections to GitHub, which may cause Claude Code remote sessions, GitHub Enterprise plugin syncing, and Claude Security to fail. We are reverting the change to restore the previous IP ranges; no allowlist changes will be needed on your side. We will post an update once the fix is deployed.