Claude AI is down again. Downdetector shows the Anthropic-owned service experiencing widespread disruptions, with over half of reported issues hitting Claude Chat specifically. Users are seeing login failures and that familiar "Claude is currently experiencing a temporary service disruption" message. The problems started showing up around 10:51 AM EDT according to Downdetector's tracking.
This isn't a one-off. User comments on the outage tracker paint a picture of intermittent problems stretching back months. One paying Max customer complained that Claude Code keeps getting stuck on "Moseying... Getting a wriggle on" endlessly. Another reported needing 8 to 9 attempts just to get output.
A January 25 incident marked as "Increased rate of errors" remains unresolved. The outages hit different parts of the platform inconsistently. Some users report Claude Code working while the main app stays down. Others get CSS loading failures or cryptic UUID errors. This suggests backend strain rather than a single point of failure.
Anthropic runs Claude across Google Cloud's TPU infrastructure and AWS, which should provide redundancy. Clearly it's not working as intended. The company hasn't commented publicly on today's outage. Their status page still shows "All Systems Operational" even as Downdetector reports climb. For anyone building workflows around Claude, these recurring problems are becoming a real headache.