Anthropic’s Claude Faces July 28 Performance Issue
Anthropic’s Claude AI services experienced a performance issue on July 28, 2026, affecting multiple features including Claude Code on the Web, Claude Code Review, and document creation capabilities in claude.ai. The company’s status page confirmed the incident was under investigation before later marking it as resolved. The outage adds to a pattern of service disruptions that have affected the AI platform throughout July 2026.
According to Anthropic’s official status page, the company began investigating the performance degradation affecting several key features. While specific timing details were limited, third-party monitoring service StatusGator reported that users submitted 25 outage reports within a 24-hour period ending July 28, though the service was later confirmed operational.
Affected Services and Features
The performance issue impacted several of Claude’s core offerings. Document creation functionality within the main claude.ai interface experienced problems, alongside both web-based and remote versions of Claude Code, the company’s AI-assisted software development tool. Claude Code Review and Cowork Remote, collaborative features designed for team environments, were also affected by the degradation.
Despite these disruptions, not all Claude services experienced issues. The Claude Console and Claude for Government platforms appeared to remain unaffected during the incident, maintaining normal operations throughout the period.
Recent Service Reliability Concerns
This latest incident follows a series of outages that have affected Anthropic’s services in recent weeks. On July 6, 2026, the company dealt with a more widespread service disruption that affected multiple Claude models. The company’s status page flagged the issue at 07:36 UTC on July 6, initially listing elevated errors across several models. By 08:45 UTC, the issue had spread to Fable 5.
Four of six monitored services on status.claude.com showed “degraded performance” during that incident: claude.ai, the Claude API, Claude Code, and Claude Cowork. Claude Console and Claude for Government appeared unaffected by the problem. The July 6 outage was resolved by 09:13 UTC.
Anthropic has attributed previous outages issues to demand outpacing available compute capacity, driven by enterprise adoption of Claude Code and a surge in consumer signups. The company has seen significant growth in its Claude Code product, which became a viral sensation during the 2025-2026 winter holidays when users, including many non-programmers, experimented with the AI-assisted development tool.
Uptime Performance
According to the official status page, over the past 90 days, claude.ai has maintained 99.46% uptime, the Claude API 99.53%, Claude Code 99.45%, Claude Cowork 99.59%, Claude Console 99.77%, and Claude for Government 99.96%. These figures indicate that while the services generally maintain high availability, sporadic performance issues continue to affect user access.
The status page also documented other recent incidents, including a resolved issue on July 24 that required monitoring after a fix was implemented, and another incident on July 22-23. On July 14, the company investigated an outage affecting document creation in claude.ai, Cowork Remote, Claude Code, Claude Code Remote, and Claude Design.
Broader Security and Privacy Concerns
Beyond service reliability, Claude has faced other challenges in recent months. A separate incident saw public Claude pages and conversations containing sensitive data, including API keys, credentials, and resumes, appear in search results before many were removed, raising privacy concerns for enterprise and individual users.
Security researchers have also raised concerns about AI-generated applications. Sygnia, a cybersecurity firm, reported finding critical vulnerabilities in a Claude-built customer onboarding application that handled sensitive personal and financial data, highlighting potential risks in trusting AI-generated code without thorough verification.
Additionally, geopolitical tensions have emerged around Claude Code. China issued warnings about certain versions of the tool potentially posing back-door vulnerabilities and sending sensitive information to remote servers. Separately, Alibaba banned Claude Code for employees after accusing Anthropic of a distillation attack, placing the tool on a high-risk software list.
Product Development Continues
Despite these challenges, Anthropic continues to develop its product line. Recent updates to Claude Code include version 2.1.218, released on July 22, 2026, which introduced enhanced code review functionality running as background subagents, improved accessibility features, and fixes for Windows path handling issues. The company has also been working on expanding capabilities such as nested subagent spawning and updated model defaults.
The ongoing service disruptions come at a critical time for Anthropic as it competes in the rapidly evolving AI market. The company must balance rapid innovation and growing user demand against the need for stable, reliable infrastructure capable of supporting enterprise deployments and maintaining user trust.
Key Facts
- Anthropic investigated a performance issue on July 28, 2026, affecting Claude Code, document creation, and other features
- StatusGator received 25 user-submitted outage reports within a 24-hour period, though the service was later confirmed operational
- Claude services have maintained uptime ranging from 99.45% to 99.96% over the past 90 days
- Previous outages on July 6 affected four of six monitored services before being resolved
- Anthropic has attributed capacity issues to enterprise adoption of Claude Code and consumer signup surges
Sources
- Claude Status – Anthropic Official Status Page
- Claude Status Monitoring – StatusGator
- Claude Outage Currently Affecting Multiple Models – MacRumors
- Claude Code Changelog – Official Documentation
