We started the Princeton AI for Climate Network because we were looking for a space where AI researchers and climate practitioners are ready to talk to and collaborate with each other.

At Princeton, we have world‑class expertise in both. But the two communities largely operated in silos. That gap between cutting‑edge models and real‑world climate action felt too urgent to ignore.

That is why a group of us founded PAICN – a graduate‑student‑led space to ask the hard questions: How can AI accelerate climate solutions without making the planet’s energy problem worse? Who benefits, and who gets left behind?

Ahead of Earth Day, we organized our first launch reception. Over 50 participants ranging from students, faculty, practitioners, and even a local high school teacher and student, joined us. The insights from our panelists and interactive Q&A with the audience fundamentally shaped our research agenda for months to come.

Here is what we learnt, in a nutshell:

Prof. Robert Socolow, an energy systems legend and father of the “stabilization wedges”, warned that today’s data centers are like 1960s cars: inefficient and lacking market pressure to improve. His lesson from 1970s electricity forecasts: exponential growth always levels off. The question is when, and how much carbon is emitted before it does.

Prof. Ning Lin showed how reinforcement learning can design adaptive coastal defenses for NYC. But AI trained on past climate may fail under future non‑stationarity. Hybrid modeling (physics + ML) is the path forward.

Dr Colby Fisher (Founder of Hydronos Labs) exposed the translation gap: better academic models go unused because industry can’t run or understand them. His fix – openness and reproducibility – turns end users into makers.

Grace Lam (Co-Founder of Neta AI) flipped the script: the bottleneck isn’t data, it’s workflows. “What I thought the world needed, no one would pay for.” Start with a problem that’s on fire.

Gituku Ngene (PAICN founding committee, UNDP) gave a global justice edge: without care, AI will widen the adaptation gap. Our long‑term goal – extend Princeton’s expertise to the Global South as genuine capacity building, not extraction.

Sarah Boll, Princeton’s Executive Director of Sustainability, challenged students to model AI’s future energy demand on campus, which is a concrete local test.

What’s next? PAICN will be organizing a smaller‑scale graduate roundtable to turn these insights into actionable research agendas, on energy‑efficient AI, decision‑ready climate models, and equitable deployment.

Deepest gratitude to our speakers, participants, and everyone who believes that graduate students can help bridge AI and climate action. If you’re a Princeton student passionate about AI and climate, join us!

#PAICN #AIforClimate #PrincetonUniversity #ClimateAction #Sustainability