Claude for Excel (www.claude.com)

Claude for Excel has launched as a beta research preview (waitlist open to 1,000 Max, Team and Enterprise customers), bringing Claude’s LLM capabilities directly into Excel to analyze entire workbooks—nested formulas, multi-tab dependencies and calculation flows. It provides cell-level citations with every explanation, can update assumptions across a model while preserving formulas and formatting, and highlights changes with real-time explanations. Use cases include rapid answers about specific cells or worksheets, scenario testing without breaking dependencies, automated debugging of #REF!, #VALUE! and circular reference errors, and generating or populating financial models and templates. Supported file types are .xlsx and .xlsm; file-size limits vary by plan. For the AI/ML community this is significant because it operationalizes explainability and provenance inside a ubiquitous analytical tool: cell-level citations and traceable edits improve auditability and reproducibility of model logic, while preserving formula integrity reduces risky manual interventions. It also showcases hybrid workflows where LLMs assist with debugging, assumptions testing and template population rather than replacing domain review. Important caveats: Claude can make mistakes and should be reviewed before finalizing outputs; advanced Excel features (pivot tables, conditional formatting, data validation, data tables, macros/VBA) are not yet supported. The feature is designed to work within enterprise compliance frameworks, signaling a measured rollout focused on security and correctness.