Anthropic confirms Claude is down in major outage affecting multiple services

A significant disruption has impacted various services offered by Anthropic, the artificial intelligence company behind the Claude suite of tools. Users have been encountering widespread issues, including difficulties logging in and a noticeable decline in performance across the platform. This service interruption commenced on August 16, 2026, around 21:58 UTC, and has specifically affected Claude.ai, a sophisticated conversational AI, Claude Code, designed for programming assistance, and Claude Cowork, a collaborative AI workspace.

Initially, Anthropic’s status page, a crucial resource for real-time updates, indicated an investigation into authentication problems preventing users from accessing these core services. Shortly thereafter, the scope of the problem expanded, with reports confirming degraded performance across both Claude.ai and platform.claude.com. For those attempting to utilize these services, the outage has manifested in several ways: an inability to sign in, the AI failing to load or respond, and requests remaining incomplete, among other operational failures.

As of the current assessment, Anthropic’s status page continues to categorize Claude.ai, Claude Code, and Claude Cowork as experiencing a major outage. Interestingly, other critical components such as Claude Console, typically used for managing AI deployments, and the Claude API, which allows developers to integrate Claude’s capabilities into their own applications, appear to be functioning normally. The precise cause of this widespread disruption remains undisclosed by Anthropic, and investigations are ongoing.

The company provided a series of updates to keep users informed. The first update, issued at 21:58 UTC on August 16, acknowledged the initial authentication issues. A subsequent update at 22:07 UTC confirmed the broader performance degradation. Encouragingly, by 22:40 UTC, Anthropic announced that all affected services had been fully restored, bringing an end to the outage. This incident underscores the inherent complexities and potential vulnerabilities in maintaining high availability for advanced AI services.