Claude Outage History

Claude Opus 4.5 experienced elevated errors beginning at 11:33 AM UTC, affecting Claude API (api.anthropic.com), claude.ai, platform.claude.com, and Claude Code. Users attempting to access these services encountered errors and degraded responses. The Claude status page noted an investigation into the issue, and the outage lasted for about one hour before services returned to normal at 12:33 PM UTC.

The service posted two status updates during the incident. Reports came from multiple regions, including Europe and India, with some users reporting overloaded errors. The issue affected four components and generated a moderate level of user-reported activity before resolution.

Claude Sonnet 3.7 experienced elevated errors across multiple endpoints, causing degraded performance for Claude API (api.anthropic.com), claude.ai, platform.claude.com, and Claude Code. The incident began around 10:13 AM UTC with investigation, the status was raised to a degraded state at 10:18 AM UTC, and services were fully restored by 10:38 AM UTC. Claude's status page documented the investigation and subsequent recovery, with updates reflecting the progression from issue to resolution.

Claude Console experienced degraded performance beginning at Jan 27, 2026 10:28 PM UTC and ended at Jan 27, 2026 11:08 PM UTC, affecting platform.claude.com (formerly console.anthropic.com). Users may have encountered slower responses or intermittent issues when accessing the Claude Console. The outage lasted about 40 minutes.

Claude Haiku 3.5 experienced elevated errors and degraded performance across multiple access points, including Claude API (api.anthropic.com), claude.ai, platform.claude.com, and Claude Code. Users reported slow page loads and upgrade-related issues during the incident.

Claude AI app and the web interface experienced a global disruption, with users worldwide reporting the service not loading, prompts failing, or chat interactions timing out. The incident began with an Early Warning Signal at Jan 25, 2026 12:36 PM UTC indicating issues with the Apple app, and the signal remained in place until {{time:18:42:00Z}}. Activity from user reports surged through the afternoon, indicating widespread impact across many regions.

Claude experienced a brief degradation in Opus 4.5 with increased error rates impacting four components: Claude API (api.anthropic.com), claude.ai, platform.claude.com, and Claude Code. Users interacting with the API and related tools may have experienced intermittent failures and slower responses. The outage began with an investigation around 8:05 AM UTC and the status changed to degraded at Jan 25, 2026 8:08 AM UTC.

The issue was resolved by 9:58 AM UTC, with the outage lasting approximately 1 hour 53 minutes. The Claude service issued 2 status updates during the incident, describing the ongoing investigation and the subsequent recovery.

Claude API (api.anthropic.com) experienced elevated errors starting at Jan 23, 2026 5:13 AM UTC due to an invalid signature in the thinking block, causing API calls to fail or return errors. The incident lasted until Jan 23, 2026 7:48 AM UTC, a total duration of 2 hours and 35 minutes. The issue affected users attempting to use Claude's API endpoints, with responses that were not as expected.

The provider posted two status updates during the incident: the initial notice of elevated errors and ongoing work to fix the issue, and a follow-up stating that the service had returned to normal. Engineers implemented a fix and the Claude API returned to standard operation by the end of the window.

An outage affected Claude API (api.anthropic.com), claude.ai, and Claude Code, causing elevated error rates and degraded performance for a brief period. The incident began at 10:23 PM UTC when the Status page indicated a degraded state and was resolved by 10:28 PM UTC. The service posted two status updates during the event—one noting elevated error rates and ongoing monitoring, and a follow-up confirming restoration. With user reports indicating widespread impact across multiple regions, the outage lasted about 5 minutes.

Claude.ai began experiencing elevated errors at 4:08 PM UTC, with Claude Code joining the disturbance shortly after. Users reported errors when sending prompts, getting responses, or accessing chats, including non-responsive chats, error messages, blank screens, and login or prompt failures. The issue affected both claude.ai and Claude Code and persisted for about 45 minutes before services returned to normal, with restoration indicated by 4:53 PM UTC.

The provider issued 3 status updates during the incident, communicating ongoing investigation and eventual restoration: at Jan 22, 2026 4:08 PM UTC, Jan 22, 2026 4:13 PM UTC, and Jan 22, 2026 4:53 PM UTC. A broad set of users across multiple regions reported trouble with chat and coding workflows, reflecting a brief, global impact on Claude AI services.

Claude.ai experienced a global outage that disrupted chat and degraded responses on claude.ai and Claude Code. Users reported that messages were not being accepted and conversations were unreliable, with a large volume of user reports (147 total) indicating widespread impact. StatusGator detected the issue at Jan 22, 2026 3:31 PM UTC, about 37 minutes before the first official acknowledgment at Jan 22, 2026 4:08 PM UTC. The outage affected two components, claude.ai and Claude Code, and lasted until Jan 22, 2026 4:53 PM UTC, for a duration of roughly 1 hour 22 minutes.

The Claude status page issued 3 updates during the incident (at Jan 22, 2026 4:08 PM UTC, Jan 22, 2026 4:13 PM UTC, and Jan 22, 2026 4:53 PM UTC UTC). This indicates the incident was actively managed, and services were restored to normal operation after the final update at Jan 22, 2026 4:53 PM UTC.