Anthropic Outage History

Anthropic Claude experienced a brief outage where chat messages and previous conversations failed to load for users. The incident lasted about 10 minutes, beginning around 4:00 AM UTC and resolving by 4:10 AM UTC. StatusGator detected the issue as an Early Warning Signal based on user reports, noting loading problems before any official acknowledgement. There were no official status page updates from Anthropic during the event, and available reports suggest the disruption was limited in scope.

The Claude outage affected Claude.ai and the Claude platform (platform.claude.com), rendering Claude chat and related services unavailable for roughly 35 minutes. Users worldwide reported being unable to access Claude and encountering errors across multiple regions. The outage began at 10:48 AM UTC and was resolved by 11:23 AM UTC.

The Claude status page showed the incident progressing from an initial disruption to full restoration, with 2 official updates published during the incident. A notable volume of user reports from around the world indicated broad reach, with issues reported across Europe, North America, and Asia. The service returned to normal following the resolution at 11:23 AM UTC.

Anthropic Claude.ai experienced a brief outage, with the Claude.ai component becoming unavailable for about 35 minutes. The outage began on Dec 5, 2025 9:03 AM UTC and was resolved by Dec 5, 2025 9:38 AM UTC. Users worldwide reported issues, including errors such as HTTP 500 and authentication-related messages across regions including Europe, North America, Africa, and Asia.

The provider issued two status updates during the incident, describing an initial investigation and subsequent restoration. The outage was confined to the Claude.ai component and affected users attempting to access Claude.ai features.

Anthropic Claude Haiku 4.5 and Opus 4.5 experienced elevated errors, affecting three components: Claude API (api.anthropic.com), claude.ai, and Claude Code. Users attempting to interact with these services may have seen failures or degraded responses during the incident.

The issue lasted about 10 minutes, beginning at Dec 3, 2025 12:23 PM UTC and resolving by Dec 3, 2025 12:33 PM UTC. The provider issued two status updates during the event, first noting an investigation and later confirming that services were returning to normal.

Anthropic Claude experienced a widespread login outage that prevented users from signing in to the web interface and mobile apps. The issue was first evident in user reports around 4:15 PM UTC, with a rapid surge in complaints through the afternoon. StatusGator issued an Early Warning Signal at Dec 2, 2025 5:09 PM UTC, noting login issues affecting web and app access, and the signal ended at Dec 2, 2025 5:28 PM UTC. The outage blocked access to Claude services, including Claude.ai, for many users until the issue appeared to abate.

The Claude AI service claude.ai experienced an outage that prevented many users from signing in and accessing Claude functionality. The incident began around Dec 2, 2025 4:34 PM UTC with elevated errors and a status of investigating, then moved to a degraded state at Dec 2, 2025 4:38 PM UTC. Reports from users around the world indicated widespread sign-in issues, with people in North America, Europe, Asia and other regions unable to sign in or access Claude. The outage lasted approximately 1 hour and 30 minutes and affected the claude.ai component.

The Claude status page documented two official updates during the incident, detailing the ongoing investigation and eventual remediation. By Dec 2, 2025 6:08 PM UTC the status returned to normal, and access was restored. A high volume of user reports over the period reflected the broad impact across multiple regions.

Anthropic Claude experienced latency across four core components: platform.claude.com, Claude API (api.anthropic.com), claude.ai, and Claude Code. Users reported slower responses and intermittent delays during the incident, with latency observations beginning around 10:00 PM UTC. The degraded performance was acknowledged when the Claude services moved to a degraded state at Dec 1, 2025 10:13 PM UTC, as teams began investigating around 10:09 PM UTC. The issue lasted about 1 hour and 30 minutes and the services returned to normal by Dec 1, 2025 11:43 PM UTC.

An outage affected Anthropic Claude's Sonnet 4.5, beginning at 7:03 PM UTC and lasting until 9:18 PM UTC. The disruption impacted four components: platform.claude.com, Claude API (api.anthropic.com), claude.ai, and Claude Code. Users globally reported slow or unresponsive chats, long response times, and timeouts across Claude's chat and development features. The incident persisted for a little over two hours with widespread user impact. The Claude status page published four updates during the incident. At 8:41 PM UTC the team stated the issue had been identified and a fix was being implemented, and by 9:18 PM UTC services were restored across all components. A large volume of user reports around 8:00 PM UTC UTC indicated the issue affected users across regions.

Anthropic Claude Haiku 4.5 experienced elevated errors affecting multiple Claude services, with users hitting platform.claude.com, the Claude API, claude.ai, and Claude Code seeing failures or slower responses. The incident began at Nov 30, 2025 2:08 PM UTC and lasted about 30 minutes, with services restored by Nov 30, 2025 2:38 PM UTC.

Anthropic's status page documented an investigation and subsequent recovery, with the incident resolved by Nov 30, 2025 2:38 PM UTC. The provider posted two official updates during the incident to communicate progress.

The Anthropic Claude platform (platform.claude.com) and Claude API (api.anthropic.com) experienced a brief degradation, with elevated errors reported for requests to Claude 4.5 Sonnet 1M. Users attempting to access Claude features may have encountered failures or degraded responses during the window.

According to the Claude status page, the incident began at 9:13 AM UTC and was resolved by 9:23 AM UTC, lasting approximately 10 minutes. The page noted that the issue had been identified and a fix was being implemented around 9:10 AM UTC. The service issued two status updates during the incident as the teams worked to restore normal operation.