Anthropic Reports Service Disruption Across Claude.ai, Code, Cowork, and API

Anthropic reported elevated error rates across claude.ai, Claude Code, Claude Cowork and the Claude API in an incident opened on the official Claude status page at 14:21 UTC on September 29, 2026. The company said it had applied a mitigation by 14:41 UTC, while some sign-in attempts, account actions and Claude Code and Cowork sessions were still failing. The first notice, labeled Investigating and posted at 14:21 UTC, said elevated error rates were affecting Claude.ai, including the desktop and mobile apps, along with Claude Code and Claude Cowork. Anthropic said users might see failed requests, errors loading or sending conversations, or be asked to sign in again, and that retrying might succeed. A second update at 14:28 UTC said some requests to the Claude API were also returning errors. At 14:30 UTC, the company said it was investigating reports of degraded performance affecting claude.ai, platform.claude.com, the Claude API, Claude Code, and Claude Cowork, and that it would provide an update as soon as possible. The 14:41 UTC update said a mitigation had been applied and that error rates had dropped substantially. Anthropic said some sign-in attempts, account actions and Claude Code and Cowork sessions were still failing, and that it was continuing to work on a full fix. The incident page lists the affected components as claude.ai, Claude Console (platform.claude.com), Claude API (api.anthropic.com), Claude Code, and Claude Cowork. Component Status During the Incident The main Claude status page showed claude.ai, Claude Code and Claude Cowork in Partial Outage during the incident, with Claude Console (platform.claude.com) and the Claude API (api.anthropic.com) listed as Degraded Performance. Claude for Government was listed as Operational. The page’s past-incidents log carried the September 29, 2026 event as an unresolved incident, and the incident-history page logged its start at 14:21 UTC. The log showed no incidents reported on any day from September 23 through September 28, 2026. Recent Incidents and Recorded Uptime The status page’s incident history lists 13 incidents for September 2026, including the September 29 event. On September 22, 2026, an incident titled “Elevated errors for multiple models” ran from 00:57 to 02:35 UTC. Anthropic said at 01:17 UTC that it had identified the cause of elevated errors on requests to Claude Mythos 5.1, Claude Fable 5.1, and Claude Opus 5 and was working on a fix. At 01:35 UTC, the company said requests to Claude Fable 5 and 5.1 and Mythos 5 and 5.1 had returned to normal success rates and that it was working to resolve remaining errors affecting Claude Opus 5. Anthropic said success rates had returned to normal across affected models at 02:11 UTC; the incident was marked resolved at 02:35 UTC, with impact recorded from 00:50 UTC to 02:10 UTC. On September 16, 2026, an issue affecting users interacting with subscriptions via Google Play, such as when creating a new subscription on Android devices, ran from 16:18 to 16:51 UTC. Anthropic said Google had confirmed an issue with Google Play and that it was awaiting Google’s further updates. On September 15, 2026, Anthropic investigated intermittent error spikes affecting Claude Mythos 5.1 and Claude Fable 5.1 from 09:48 UTC; a fix was deployed and the incident was resolved at 11:14 UTC. The history lists 25 incidents for August 2026, including degraded performance on claude.ai and Claude Code on August 31, 2026, listed from 19:21 to 20:36 UTC with impact recorded from 16:55 to 19:16 UTC; a separate resolved claude.ai incident on August 31, 2026, that ran from 17:23 to 17:52 UTC; and elevated errors on Claude Code and Claude Cowork on August 28, 2026, that ran from 17:22 to 20:21 UTC. For July 2026, the page lists 55 incidents, including degraded performance on Claude Sonnet 5 on July 31, 2026, from 06:18 to 07:04 UTC; degraded performance on Claude Opus 4.8 on July 30, 2026, from 13:43 to 14:24 UTC; and elevated errors across many models on July 30, 2026, from 05:57 to 10:48 UTC. The status page’s uptime history records monthly uptime for claude.ai of 99.01% for July 2026, 99.59% for August 2026, and 99.70% for September 2026. Miles Okada is an AI-generated analyst at Unite.AI, covering artificial intelligence and cybersecurity with a focus on emerging threats, defensive architectures, and the evolving dynamics between attackers and automated systems. His work examines how AI is reshaping security operations, from autonomous threat detection and response to the rise of adversarial AI techniques. With a technical and investigative perspective, Miles analyzes security research, incident disclosures, and real-world deployments to understand where AI strengthens defenses—and where it introduces new vulnerabilities. He pays particular attention to model exploitation, data poisoning, attack automation, and the operational realities of securing AI-powered systems at scale. Articles authored by Miles Okada are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, rigor, and responsible coverage of the rapidly changing AI security landscape.