Research Update · 2023 – 2026
From optimization to real-world impact
Fengqi You — Roxanne E. and Michael J. Zak Professor, Cornell University
A few visual updates on where the research has gone since we last met — AI for Sustainability and AI for Science.
The bridge your gift created — operations research meeting energy and chemical engineering — is now an engine turning AI into real, scalable technology.
A lot has happened
Since our last meeting in 2023
2023
- AAAS Fellow
- Co-Director, AI for Science Institute
2024
- NSF AI-for-Sustainability traineeship
- Cecil & Amazon Research Awards
2025
- Nature — perovskite recycling
- Nature Sustainability — AI data centers
2026
- Nature — wearable electronics
- EMSeek · battery-electrolyte AI
A selection of recent honors and awards — IEEE, AAAS, AIChE, ASEE, and more.
The lab at a glance
Two engines, at real scale
0
Google Scholar citations
0
h-index
0
Journal articles since 2024
0
Nature · Science · PNAS papers
0
Journal covers & frontispieces
0
Media outlets worldwide
Breakthroughs by theme
What the engine is producing
Sustainability intelligence
AI agents for life-cycle assessment and supply-chain footprints.
Automating LCA, from wearables to AI servers
AI for batteries & materials
Generative, interpretable models for electrolytes and materials.
65%+ lower electrolyte-prediction error
Perovskite solar & recycling
Low-toxicity aqueous recovery for next-gen solar.
Scalable aqueous recycling — in Nature
AI for agriculture
AI control for plant factories, greenhouses, agrivoltaics.
~25% less energy in indoor farming
Flagship · Nature Sustainability 2025
The energy cost of the AI boom — and how to avoid it
First open, state-by-state roadmap of AI data centers' water and carbon footprint.
AI isn't just software — it's energy, water, land, and grid.
5–10M cars
CO₂-equivalent by 2030
6–10M people
equivalent water use by 2030
Carbon emissionsup to −73%
Water useup to −86%
NPRpublic radio
APglobal wire
The Guardianinternational
WIREDtechnology
Fast Companybusiness & design
Inc.business
POLITICOpolicy / E&E
Inside Climate Newsclimate desk
AI for Science · case study
Autonomous AI for scientific discovery
EMSeek turns electron-microscopy images into structured materials insight in minutes.
2–5 min
image to report
~50×
faster workflow
Microplastics × AI + quantum
From a public-health alarm to an AI design problem
We ingest microplastics constantly — on the order of a credit card's worth each week.
deepPolyNet — AI that detects and classifies microplastics (with N. Abbott, Science Advances).
AI and quantum optimization then design plastic-binding peptides to capture them.
(Chemical Science · PNAS Nexus · Science Advances)
Energy security & critical materials
Where energy meets national security
EU-wide modeling of the energy transition, carbon capture, and storage.
From clean-fuel production to grid and industrial decarbonization.
61%
of EU natural gas displaceable by renewables by 2050
32–51%
of gas use cut through electrification
~0.95 Gt
CO₂ avoided via green hydrogen in heavy industry
Lithium
& critical-materials supply-chain risk, mapped spatially
The same optimization tools align the energy transition with security of supply — EU strategies, transcontinental corridors, and critical-mineral sourcing. (Nature Communications · ES&T)
Cornell Chronicle research story →Science on the cover
Journal covers since 2023
In the news
Reaching the public — 300+ outlets since 2023
The New York Timesnational daily
The Washington Postnational daily
The Wall Street Journalbusiness daily
APglobal wire
Reutersglobal wire
NPRpublic radio
Bloombergmarkets & tech
The Guardianinternational
CNNbroadcast
TIMEnewsweekly
Forbesbusiness
Fortunebusiness
Fast Companybusiness & design
HBRmanagement
WIREDtechnology
Scientific Americanscience
Newsweeknewsweekly
Al Jazeeraglobal news
VOGUEculture
Natureresearch news
Looking ahead
From papers to platforms
The strongest opportunities are research-to-impact pathways — AI-infrastructure planning, autonomous science workflows, materials intelligence, smart agriculture, and sustainability decision tools.
105NRT doctoral trainees
16graduate fields
5Cornell colleges
11 patents & applications
7Quantum-computing AIOptimization, molecular & peptide design, detection.
2AI ag-techEnergy-efficient controlled-environment growth & plant factories.
2EV control & microplastic detectionEV fleet control; ML multi-object detection.
Institutional capacity: faculty, trainees, and research domains across four arms.
Thank you
The bridge you built is now an engine — and it's accelerating.
From O.R. to AI for sustainability and science — shaping how the world builds, powers, feeds, and cleans up after itself.
Happy to go deeper on any of these →
AI infrastructureData-center footprint & planning
Carbon, water, siting, and a mitigation roadmap for AI-infrastructure growth.
Cornell story →AI + quantumPeptides for microplastic cleanup
Generative AI and quantum-assisted optimization for plastic-binding peptide design.
Cornell story →Materials discoverySmarter, faster AI models
Reusable models for molecular and materials discovery, from batteries to sustainable chemicals.
Cornell story →Food systemsAI control for indoor agriculture
AI-regulated light and climate for energy-flexible plant factories and greenhouses.
Cornell story →Energy systemsOffshore wind & green hydrogen
Systems analysis of offshore wind-based hydrogen for coastal U.S. decarbonization.
Cornell story →Research trainingNSF AI4S NRT ecosystem
A five-year Cornell platform across AI, sustainability, energy, agriculture, materials, and climate.
Cornell story →