CanopAI : MIT AI for Energy IAP 2025
Dorothy Abreu, Harvard Kennedy School MPA '25
Rachel Koh, Boston University, Business Administration and Management '25
Bruce Hecht, MIT System Design and Management MS '23
CanopAI Summary
Summary:
The CanopAI is an AI-powered tool that provides an outline of steps to improve grassland and tree canopy coverage for improved agriculture, water, and heat management.
Methodology:
The CanopAI tool is built using Stack AI to access resources including national environmental strategies, frameworks, canopy and land use data, and background presentations, videos, and documents. The user is prompted to input the location of interest and the size (in land area). Two examples provided a high-level guideline using Meta's LLAMA LLM and Open AI's ChatGPT 4.o. These two guideline outputs are then iterated with Perplexity to provide the user with a summary for future planning including a 5-step process and considerations of the financial model and analysis.
Future Work:
The interface to build on 2D mapping using examples from Meta's approach published in 2024 to use machine learning on satellite and LIDAR data for forecasting future scenarios based on the location and area.
