Claude hit by outage — Claude Code, Cowork, and web apps all down
Elevated errors took down chat, Code, and Cowork for ~11,800 users during OpenAI's DevDay; Gov tier stayed up and Anthropic named no cause.
Copy markdown
Code, Cowork, and every app went dark
The partial outage hit Claude.ai on web, iOS, Android, and macOS, plus Claude Code and Claude Cowork — users saw failed requests, conversations that wouldn't load, and forced sign-ins. Claude for Government stayed up.
11,800+ reports before the status page caught up
Down Detector logged over 11,800 reports by 10:17am ET as errors spiked; Anthropic confirmed "elevated error rates" at 10:21am ET but gave no root cause or ETA. If your builds or agents stalled mid-morning, this was why.
The single-vendor reminder
It landed the same afternoon OpenAI's DevDay soaked up the headlines, so the disruption slipped by quietly. If your pipeline hard-depends on one provider, a fallback route (Bedrock, Vertex, or a local model) turns a day like this into a blip instead of a standstill.