Anthropic has confirmed a major outage affecting its Claude AI assistant, with users around the world reporting failures across web, mobile and API access.
The disruption was first formally acknowledged on 2 March 2026, when the company posted an incident notice at 11:49 UTC stating that engineers were “investigating” elevated error rates. A further status update at 12:06 UTC confirmed that the investigation was still in progress and that service had not yet been restored.
The issue is not confined to a single product or geography. According to the current incident description, the problem is “impacting users broadly rather than being limited to one app or region,” indicating that Claude’s entire platform footprint is affected. Users attempting to interact with the system may encounter failed requests, timeouts or inconsistent responses while the outage continues.
Anthropic has not yet given a timeframe for when Claude will be fully operational again. The most recent communication notes that there is currently no estimated time of resolution but stresses that the incident is actively being worked on by the technical team.
The company’s status page lists the incident as ongoing, with the core symptom described as “elevated errors” across all interfaces. That encompasses attempts to use Claude via the web interface, mobile access, or programmatically through the API.
Until a further update is posted, developers and end users relying on Claude are being advised, in effect, to expect intermittent or complete disruption to normal service. The next key development will be Anthropic’s confirmation that the root cause has been identified and a fix deployed, followed by a change of incident state from “investigating” to recovery or resolved.
