Since 2025, we have worked with Microsoft’s Azure AI Foundry - Microsoft’s unified platform for building, deploying, and managing AI applications - to automate carbon footprint calculations and help guide suppliers through decarbonization strategies.
Being cited in Microsoft’s annual report illustrates the work we’ve done thus far.
Microsoft has an ambitious goal to become carbon negative by 2030, but more than 85% of its footprint sits in Scope 3 - and nearly three-quarters of that comes from purchased goods and services and capital goods, meaning its suppliers.
Large enterprise buyers are increasingly asking suppliers to measure, report, and reduce their own emissions. A shift many suppliers lack the internal expertise or tooling to navigate on their own. Suppliers across the board face several real, well-documented barriers to decarbonization: inconsistent methods for measuring emissions, limited access to carbon-free electricity, and no clear way to identify where reduction efforts would have the greatest impact.
For a company managing supplier relationships at this scale, calculating and reviewing carbon footprint data manually across a large supplier base is also not practical, and inconsistent methodology can make it hard to compare progress or prioritize action.
Green Project built custom models in Microsoft Azure AI Foundry that automate carbon footprint calculations for suppliers and help guide them through their decarbonization efforts. This work supports Microsoft's broader effort to help its supplier base measure and reduce emissions at scale, alongside Microsoft's own internal AI-enabled tools for reviewing supplier carbon data.
These models power an intelligent chatbot that walks suppliers through tailored decarbonization strategies, alongside automated corporate- and product-level carbon footprint calculations and AI-driven quality checks
After implementing these models in 2025, Microsoft has been able to automate carbon footprint calculations across 100+ suppliers, saving an estimated 7,000+ hours of analyst time performing manual data entry and validation across Microsoft's supplier cohort.
Our models have identified decarbonization opportunities across this supplier base, and participating suppliers now have clearer visibility into where emissions concentrate and how to best position themselves alongside Microsoft's sustainability priorities.
This work reflects a small portion of the AI-enabled tooling that supports Microsoft’s supplier decarbonization efforts at scale, and Green Project looks forward to building on that foundation.
