AI Chatbot (Enterprise)
Documentation for the AI Chatbot feature in LandOS.
Overview
The AI Chatbot provides project-level analytical summaries and cross-computation insights by applying large language model (LLM) interpretation to the quantitative outputs produced by LandOS computations. Rather than requiring users to extract and synthesize numbers from individual result widgets, the chatbot generates structured plain-language assessments of what the data means for a carbon project.
The AI Chatbot is available to enterprise users.
Regenerating Summaries
Summaries can be regenerated at any time. Users may provide custom instructions to direct the chatbot's focus. For example, requesting that it emphasize a specific land cover class or compare results against a reference polygon. Regenerating with custom instructions does not alter the underlying widget data.
After navigating away from the chatbot, all chat history will be lost.
Polygon Comparisons
The chatbot can be queried across multiple computations or analyses within a project to generate comparative insights. Example comparisons are:
- Contrasting Land Eligibility results between two candidate project areas
- Comparing REDD+ VCU estimates across different methodology variants (Verra, Hansen, MapBiomas)
- Evaluating how different Forest Loop labeling configurations affect eligible area
Limitations
AI-generated summaries are derived from the quantitative outputs of LandOS computations and should be evaluated alongside the underlying data. The chatbot does not access external databases, real-time market data, or sources beyond the computation outputs provided to it. Summaries are intended to support interpretation and communication, not to substitute for expert technical review of project feasibility.
Prefeasibility results are indicative; the chatbot's interpretations carry the same caveats as the underlying computations.
Computation types without AI summarization enabled do not currently generate chatbot output.