Anthropic outage history

Users of Anthropic Claude experienced degraded calendar functionality as the Google Calendar connector encountered elevated errors. The issue affected Claude AI, Claude Code, and Claude Cowork, with attempts to list or search calendar events failing. The problem first appeared as an elevated-issues notice at Oct 28, 2026 4:23 PM UTC and had been affecting services since October 26. A single user report around 6:00 PM UTC described a server not responding experience.

The Gmail connector within Anthropic's Claude suite faced elevated errors affecting users who rely on search and read operations for Gmail. This issue was isolated to the Gmail connector and did not impact other Claude connectors (claude.ai, Claude Code, Claude Cowork). The impact began at 4:03 AM UTC and lasted until 5:53 AM UTC.

Official updates confirmed the investigation into elevated Gmail-related errors and described the Gmail connector issue as the scope of the disruption; by 5:53 AM UTC the affected components were returned to a healthy state with claude.ai, Claude Code, and Claude Cowork moving from warn to up, and user experiences gradually returning to normal.

Anthropic Claude Sonnet 5.5 experienced elevated errors, causing degraded performance across Claude services. The issue affected multiple components under Claude Sonnet 5.5, including claude.ai, Claude API (api.anthropic.com), Claude Code, and Claude Cowork, likely impacting users attempting to run prompts, code tasks, or workspace features. The incident began at 3:13 PM UTC and was resolved by 3:53 PM UTC as services returned to normal.

Status updates from Anthropic indicated the affected components moved from nominal to degraded (warn) and then back to up, with overall service restoration occurring after roughly forty minutes.

Users attempting to use Claude Sonnet 5.5 across Claude Cowork, claude.ai, Claude API (api.anthropic.com), and Claude Code encountered elevated errors beginning at 9:18 AM UTC. The issue caused a brief outage with requests failing for about 25 minutes. Anthropic investigated and restored services, with all affected components back online by 9:43 AM UTC. The incident affected multiple Claude services and impacted user workflows temporarily as the team worked to resolve elevated errors.

An incident briefly affected Claude Fable 5.1 with elevated errors on requests across the Claude suite. Impact began at 8:18 AM UTC, spanning Claude Cowork, claude.ai, Claude API (api.anthropic.com), and Claude Code, causing degraded responses for a short period while a fix was developed. The provider noted that the cause had been identified and that a remediation was in progress, with updates to follow as needed.

The issue was resolved by 8:33 AM UTC, and affected components returned to normal operation. The outage was brief and contained to these interfaces, with services restoring promptly after the fix.

Elevated errors affected several Claude models (Mythos 5.1, Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5) and related endpoints, causing failures or timeouts for requests to the Claude API (api.anthropic.com) and tools like Claude Cowork, claude.ai, and Claude Code. Users interacting with these models may have experienced error responses or degraded performance during the incident window, with impact across multiple Claude services.

Anthropic Claude users began experiencing elevated errors across multiple Claude endpoints starting at 5:18 AM UTC, including Claude Cowork, claude.ai, Claude Code, and the Claude API (api.anthropic.com). Requests to other Claude models returned errors or degraded responses for about 25 minutes, affecting core model endpoints and user workloads that rely on Claude services. The issue progressed through a degraded state before recovery began, with services returning toward normal after a partial restoration window.

An outage for Anthropic's Claude Opus 5.5 caused elevated errors on requests, impacting several Claude services: Claude Cowork, claude.ai, Claude API (api.anthropic.com), and Claude Code. The incident began at 12:33 AM UTC and was resolved by 12:43 AM UTC. Status page updates indicated engineers were investigating the elevated errors and working to restore normal operation across the affected components.

Claude Fable 5.1, Sonnet 5.5, Haiku 5.5, and fast mode requests to Opus 5.5 began experiencing elevated errors, starting at 11:43 PM UTC. The affected components—Claude Cowork, claude.ai, Claude API (api.anthropic.com), and Claude Code—saw degraded performance and failures for impacted tasks during the incident window.

Around 11:58 PM UTC, engineers reported recovery progress as success rates began returning toward normal and monitoring continued. By 12:13 AM UTC, all affected components were back up, marking the incident’s resolution. The status updates indicate the cause had been identified and remediation was being implemented.

Starting at 3:53 AM UTC, Anthropic reported elevated errors in its Connectors, affecting Claude Cowork and claude.ai and causing degraded connectivity for workflows relying on the Connectors. The issue manifested as elevated errors across the Claude Connector surface, limiting integration functionality for users.