About the Project

EcoFlow was inspired by the growing use of AI in our daily lives and the fact that most people, including us, rarely think about its environmental impact. We wanted to build something that makes this impact easier to understand in a simple way.

We built EcoFlow as a lightweight web app that estimates the environmental cost of AI prompts. Users type in a prompt and get an estimate of its energy use and carbon footprint. The goal is to raise awareness, not to provide exact scientific measurements.

We used a simple model where longer prompts are assumed to use more computation, which can lead to higher energy use. In simple terms, the impact increases with the size of the input. We then estimate the carbon footprint by multiplying the size of the input by a small constant that represents average emissions per unit of text processed.

We built EcoFlow using a simple full-stack setup with a frontend for input and output and a backend that handles the calculations. We focused on keeping it fast, clear, and easy to use.

One challenge we faced was figuring out how to simplify a very complex system into something understandable without losing the main idea. Another challenge was managing time while balancing high school and college work during development.

We are proud that we were able to turn a simple idea into something users can actually see and understand in such a short span of time.

Through this project, we learned how to break down complex ideas into simple models and how to build a working prototype under very limited time constraints.

Next, we want to improve the accuracy of our estimates and coherence of our updated prompts.