Anthropic is investigating a major Claude outage in September 2026 after users started facing errors across several of the company’s services. The issue is affecting Claude.ai, Claude Code, Claude Cowork, and parts of the Claude API.
Users have reported failed requests, problems loading conversations, and difficulties sending messages. Some users have also been asked to sign in again. Anthropic says that retrying a request may work in some cases.
The disruption is important because Claude is now used for more than simple AI chats. Its tools are also used by developers, businesses, and third-party applications.
What Is Affected by the Claude Outage?
The current problem is not limited to the main Claude website. According to Anthropic’s status updates, several products are experiencing elevated error rates.
| Claude Service | Reported Issue |
|---|---|
| Claude.ai | Failed requests and conversation errors |
| Claude mobile and desktop apps | Loading and access problems |
| Claude Code | Service errors |
| Claude Cowork | Service disruption |
| Claude API | Some requests returning errors |
The Claude API issue could have a wider effect because other apps and services can use Anthropic’s technology through the API. If those applications depend heavily on Claude, users could also notice problems without directly using the Claude website.
Anthropic Is Still Investigating
Anthropic initially marked the incident as “Investigating” at 14:21 UTC. Just seven minutes later, the company confirmed that the Claude API was also affected.
At the time of the update, Anthropic had not explained what caused the problem. It also had not provided a specific time for when all services would return to normal.
For users, the main option is to wait and try again later. Some requests may work after being retried, but repeated errors can continue while the investigation is underway.
This Is Not Claude’s First Recent Outage
The Claude outage September 2026 comes less than a week after another service incident.
On September 22, Anthropic reported increased errors affecting several Claude models, including Opus 5, Mythos 5.1, and Fable 5.1. That earlier incident was resolved in around one hour and 38 minutes, based on archived status information.
The latest disruption appears broader because it involves the main Claude experience, developer tools, business products, and the API.
Why the Claude API Problem Matters
An API outage can be more noticeable than a normal website issue. Developers can build their own applications around an AI API, so an interruption can potentially affect several products at once.
For businesses using Claude in their daily workflows, even a temporary outage can slow down coding, writing, research, customer support, or other AI-powered tasks.
What Happens Next?
Anthropic is continuing to investigate the incident. Until the company provides more information, the exact cause remains unknown.
For now, users experiencing errors can retry their requests and check Anthropic’s service status for updates. The bigger question is how quickly all affected Claude services can return to normal.
Internal linking opportunities: Link phrases such as “AI tools for developers,” “AI productivity tools,” “Claude API,” and “latest AI news” to relevant pages on your own website.
