Anthropic vs OpenAI: reliability
In the last 60 days, Anthropic had 41 incidents (median fix 1 min) and OpenAI had 90 incidents (median fix 1 min). Anthropic broke less often.
Side by side, last 60 days
Anthropic's worst incidents
| Date | Incident | Severity | Time to fix | Status page |
|---|---|---|---|---|
| Anthropic: Elevated errors for Claude Fable 5.1, Sonnet 5.5, and Haiku 5.5 Cause: The cause of elevated errors on requests to Claude Fable 5.1, Sonnet 5.5, and Haiku 5.5 has been identified as a configuration issue in fast mode requests to Opus 5.5. | critical | 1 min | first | |
| Coding Agents Are Broken | critical | ongoing | not yet | |
| Anthropic: Elevated errors affecting the Microsoft 365 connector | high | 28 h | first | |
| Anthropic: Some users unable to sign up with Apple on the Claude iOS app | high | 3 days | first | |
| Anthropic: Elevated purchasing errors on claude.ai | high | 2 h | first |
OpenAI's worst incidents
| Date | Incident | Severity | Time to fix | Status page |
|---|---|---|---|---|
| OpenAI: Unable to start new Workspace Agent runs | critical | 11 h | first | |
| OpenAI: Issues with Codex | critical | 50 min | +50 min | |
| Anthropic AI Service Crashes | critical | 1 min | never | |
| OpenAI-Hugging Face Incident Causes Production Outage | critical | 1 min | never | |
| OpenCode V2 Security Regression | critical | ongoing | not yet |