Anthropic's Claude AI service suffered a major outage on Monday, coinciding with a spike in user reports regarding declining output quality.
The outage, which involved elevated error rates, affected both Claude.ai and Claude Code for approximately 48 minutes between 15:31 and 16:19 UTC.
This technical disruption follows months of growing dissatisfaction among developers. Users on Reddit and GitHub have increasingly flagged issues with the model's performance and reliability.
Data suggests rising error rates
An analysis of the Claude Code GitHub repository reveals a sharp uptick in quality-related issues. When prompted to analyze the repository's open issues, Claude itself concluded, "quality complaints have escalated sharply — and the data tells a pretty clear story."
Data from the repository indicates that April is on track to exceed the volume of issues seen in March. The model noted that April has already recorded over 20 quality issues in just 13 days, following a significant jump from the January-February baseline.
Anthropic has previously implemented measures to reduce usage during peak hours to balance capacity and demand. Some industry observers, including AMD's AI director Stella Laurenzo, have claimed the model's responses are becoming less satisfactory.
However, the validity of the rising issue count remains debated. Many reports in the repository appear to be AI-generated, a common trend among open-source developers. Additionally, Anthropic's GitHub Actions script automatically closes inactive issues, which critics say may obscure the true number of unresolved problems.
