Anthropic investigates widespread outage affecting Claude services

Artificial intelligence company Anthropic is investigating technical issues affecting several of its Claude products, including its consumer chatbot, developer API, and programming environment.

According to an update on the company’s status page, the incident began on September 29, with elevated error rates reported across Claude.ai, including its desktop and mobile apps, Claude Code and Claude Cowork. Some API requests were also returning errors.

Anthropic later expanded the list of affected services to include Claude Console and the Claude API. The company classified Claude.ai, Claude Code and Claude Cowork as experiencing partial outages, while the Console and API were reported to have degraded performance.

Users reported unsuccessful requests, problems loading or sending conversations, and prompts to sign in again. Anthropic said retrying requests could resolve some issues.

Downdetector recorded more than 11,113 reports by 7:47 a.m. Pacific Time, with Claude Chat accounting for most reported problems.

Anthropic said it applied a mitigation at 14:41 UTC after opening an investigation at 14:21 UTC. The company reported that error rates had fallen substantially, but some sign-in attempts, account actions and Claude Code and Cowork sessions were still failing. Engineers continued working towards a full resolution.

The disruption came shortly after Anthropic rolled out Claude Sonnet 5.5, which became the default Sonnet model on its API on September 28. The company has been expanding its developer ecosystem through new model and platform updates.