How Can We Use AI to Help People and the Environment?

Why It Matters

AI’s ability to provide reliable, scalable insights makes it a pivotal tool for meeting climate targets and reducing global poverty, shaping policy and industry decisions worldwide.

Key Takeaways

• AI can quantify uncertainty to improve sustainable decision‑making.
• Climate change reduces snowfall, threatening recreation and local economies.
• UN Sustainable Development Goals guide AI applications for poverty and conservation.
• AI-powered wildlife cameras enable accurate species population monitoring.
• Trustworthy, transparent AI models are essential for clean‑energy and climate solutions.

Summary

The video from Science Journeys features Caltech PhD candidate Chris Yay discussing how AI can be harnessed to address environmental and social challenges, framing the talk around sustainability and the UN’s 17 Sustainable Development Goals.

Yay highlights concrete data – snowfall in Southern California’s mountains has fallen 3‑4 inches per decade due to a one‑degree Fahrenheit rise in winter minimum temperatures, and Alaska’s glaciers have vanished over a century. He explains that AI models that quantify uncertainty can improve climate‑action, clean‑energy, poverty alleviation, and wildlife monitoring.

He illustrates AI’s reach with vivid examples: AI‑tagged camera traps counting moose in Idaho forests, AI‑driven weather forecasts, spam filters, virtual backgrounds, and a personal anecdote about breaking a $400 cuvette that redirected his path from chemistry to computer science. He also recounts a Chinese tea‑farmer’s extreme‑poverty story to underscore social dimensions.

The talk underscores that trustworthy, transparent AI is critical for scaling sustainable solutions, from optimizing renewable‑energy grids to protecting endangered species, and it aims to inspire the next generation of scientists to pursue interdisciplinary, impact‑focused careers.

Original Description

Sustainability is the challenge of meeting the needs of the present without compromising the ability of future generations to meet their own needs. Solving the world's biggest sustainability challenges—like eliminating extreme poverty, protecting wildlife, and making clean energy more reliable—often means making tough decisions with limited information. For example, where should food and medical aid go to help the most people? How many endangered animals are really left in the wild? How much electricity will people need next week, or next year?

Scientists like Caltech graduate student Chris Yeh are using artificial intelligence (AI) to answer many of these questions and make predictions about what will happen to Earth's resources in the future, so that everyone, from electric companies to farmers to lawmakers, can make smarter sustainability decisions. But there's a catch: Today's AI systems generally cannot tell us how confident they are in their predictions. That becomes a problem when high-stakes decisions, like where to send aid or when to turn on or off power plants, depend on getting it right.

Chris works on building AI systems that not only make predictions but also say how certain or uncertain those predictions are. Quantifying uncertainty helps scientists, conservationists, and engineers understand when they can trust AI and when they should be more cautious about using the information AI provides. By designing AI to work better with uncertainty, we can help make smarter, safer choices for a more sustainable future.

About the Speaker:

Chris Yeh is a PhD student in computing and mathematical sciences at Caltech, co-advised by Professors Yisong Yue and Adam Wierman. His research focus is on artificial intelligence (AI) algorithms that combine uncertainty quantification and decision-making, particularly with applications in sustainability and energy systems. Growing up in Southern California, he first learned to program computers by designing websites, and he explored his love for the outdoors through camping and backpacking trips throughout California. Chris received bachelor's and master's degrees in computer science from Stanford University in 2019. He found his passion for using AI to solve sustainability challenges as part of Stanford's Sustainability and AI Lab. Outside of his research, Chris plays ultimate frisbee on the Caltech Aftermath club team and is a cellist in the Caltech Orchestra.

About the program:

In Science Journeys, Caltech graduate students and postdoctoral scholars share their research to inspire scientific curiosity. Programs are designed for middle and high schoolers, but all are welcome. These events are made possible through the generosity of the Friends of Beckman Auditorium and sponsorship from the Caltech Employees Federal Credit Union.

https://events.caltech.edu/series/science-journeys?utm_medium=web&utm_campaign=science-journeys&utm_source=youtube&utm_content=&utm_term=

If you have questions, please email [e-mail]

Recorded on March 31, 2026.

Produced in association with Caltech Academic Media Technologies.

©2026 California Institute of Technology