Anthropic Claude AI suffers 90-minute global outage

A widespread outage disrupted Anthropic's Claude artificial intelligence platform on Monday, June 22, 2026, leaving thousands of users globally unable to access its chatbot, developer tools, and API services for nearly 90 minutes. The incident, one of the broadest to affect the service in recent months, highlighted growing reliability concerns as businesses and individuals become increasingly dependent on AI for daily tasks.

The disruption began at 10:19 p.m. ET on June 21 (0037 UTC on June 22), according to the company's status page, and impacted multiple Claude models simultaneously, including Opus, Sonnet, and Haiku. Users reported a range of problems, from the chatbot returning blank responses and failing to load conversations to complete unavailability of the API and coding tools. The issues affected core products like Claude.ai, Claude Code, Claude Cowork, and the Claude API, disrupting both personal and professional workflows.

Anthropic's engineering team identified the root cause of the problem by 11:11 p.m. ET (0111 UTC) and began implementing a staged fix. Service was restored model by model, with Opus 4.8 coming back online first, followed by Haiku 4.5 and Opus 4.7....