Anthropic Investigates Elevated Error Rates on Claude.ai Platform
Key Takeaways
- ▸Anthropic reported elevated error rates on Claude.ai starting March 3, 2026 at 03:15 UTC
- ▸The incident specifically affects the claude.ai web interface, with investigation currently underway
- ▸Users can subscribe to incident updates via email or SMS through Anthropic's status page
Summary
Anthropic reported elevated error rates affecting its Claude.ai web interface on March 3, 2026, according to a status page incident report. The company posted an initial investigation notice at 03:15 UTC, acknowledging service disruptions for users accessing the consumer-facing Claude chatbot platform. The incident specifically impacts claude.ai, Anthropic's direct-to-consumer interface for its Claude family of large language models.
The status page indicates that Anthropic's engineering team is actively investigating the root cause of the elevated errors. Users attempting to access Claude.ai during the incident window may experience failed requests, timeout errors, or inability to complete conversations with the AI assistant. No estimated time for resolution was provided in the initial incident report.
Anthropic has set up notification systems allowing users to subscribe to real-time updates via email or SMS across multiple international phone networks. This incident follows Anthropic's pattern of transparent communication about service reliability, using Atlassian Statuspage infrastructure to keep users informed during outages. The company has not yet disclosed whether the issue affects its API services or enterprise deployments, or if the problem is isolated to the consumer web interface.
- No timeline for resolution or details about API service impact have been disclosed
