“Urgency is an inherently subjective call”: Anthropic tells TNS how it grades Claude Duet
Claude Duet, powered by Claude Opus 5.5, monitors connected applications to flag urgent items without user prompts. Anthropic is using internal graders and employee workflows to test the agent's ability to distinguish between urgent and non-urgent tasks. The company admits that evaluating 'urgency' is difficult because it is inherently subjective.
Key points
- Claude Duet autonomously monitors apps to prioritize tasks.
- The tool is powered by the Claude Opus 5.5 model.
- Anthropic uses internal graders to assess the correctness of task prioritization.
Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. The New Stack did the reporting; read the original for the full story.
Read the full story at The New Stack
