What is it about?
Students learning about sustainability often get plenty of general information and little help turning it into a decision. GreenBot is a free web app I built for this. Instead of one all-purpose chatbot, it uses 5 AI experts, each covering one area: biodiversity, emissions, energy, waste, and everyday consumption. A question about a coffee shop's carbon footprint starts with the emissions expert, then moves to the energy expert and the waste expert, with each one keeping the earlier context. The student ends up with one practical plan and the reasoning behind it.
Featured Image
Photo by Ant Rozetsky on Unsplash
Why is it important?
GreenBot uses 5 domain experts that hand off to each other as a question shifts topic. Keeping each answer within one domain gives students focused guidance, and the context carries between experts so the final advice holds together. It runs on swappable model providers (OpenAI, Gemini, Grok), so a school isn't tied to one company. The article reports on a working, free tool that anyone can use today.
Perspectives
I built GreenBot during my PhD, partly out of frustration that students could find plenty of sustainability information online and still not know what to actually do. Writing this reflection made me think harder about why the 5-expert split works: narrowing the framing changes the questions students ask. I'd like to see it used in real courses next, and I'm curious whether the handoff between experts feels natural to learners or gets in the way.
Farhan Ahmed
Vaasan Yliopisto
Read the Original
This page is a summary of: Reflections on GreenBot: An AI-Powered Platform for Sustainability Education, interactions, June 2026, ACM (Association for Computing Machinery),
DOI: 10.1145/3816261.
You can read the full text:
Contributors
The following have contributed to this page
