A postmortem of three recent issues \ Anthropic
This is a technical report on three bugs that intermittently degraded responses from Claude. Below we explain what happened, why it took time to fix, and what we're changing. Between August and early September, three infrastructure bugs intermittently degraded Claude's response quality. We've now resolved these issues and want to explain what happened. In early August, a number of users began reporting degraded responses from Claude. These initial reports were difficult to distinguish from normal variation in user feedback. By late August, the increasing frequency and persistence of these reports prompted us to open an investigation that led us to uncover three separate infrastructure bugs. To state it plainly: We never reduce model quality due to demand, time of day, or server load. The problems our users reported were due to infrastructure bugs alone. We recognize users expect consistent quality from Claude, and we maintain an extremely high bar for ensuring infrastructure changes do
Between August and early September, three infrastructure bugs intermittently degraded Claude's response quality. We've now resolved these issues and want to explain what happened. In early August, a number of users began reporting degraded responses from Claude. These initial reports were difficult to distinguish from normal variation in user feedback. By late August, the increasing frequency and persistence of these reports prompted us to open an investigation that led us to uncover three separate infrastructure bugs. To state it plainly: We never reduce model quality due to demand, time of d
related reading
- An update on recent Claude Code quality reports \ Anthropicanthropic.com
- Investigating three real-world incidents in our cybersecurity evaluations \ Anthropicanthropic.com
- When AI builds itself \ Anthropicanthropic.com
- Tracing the thoughts of a large language model \ Anthropicanthropic.com
- Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnettransformer-circuits.pub
- Teaching Claude Whyalignment.anthropic.com
- Claude’s Constitution \ Anthropicanthropic.com
- Teaching Claude why \ Anthropicanthropic.com
- An alignment assessment of recent cybersecurity incidentsanthropic.com
- Natural Language Autoencoders \ Anthropicanthropic.com
- Best practices for Claude Code - Claude Code Docsanthropic.com
- Better Models: Worse Toolslucumr.pocoo.org