Back

Second Edition

ICEF
Artificial Intelligence
for Climate Change
Mitigation Roadmap

ICEFArtificial
Intelligence for
Climate Change
Mitigation Roadmap

Artificial intelligence is a powerful tool for addressing one of the most pressing global challenges: climate change. The ICEF Artificial Intelligence for Climate Change Mitigation Roadmap (Second Edition, November 2024) explores how AI can contribute to the fight against climate change, offering more than 100 specific, actionable recommendations. The 17 chapters of the Roadmap and related material are below.
Scroll down
Preface

The first ICEF Artificial Intelligence for Climate Change Mitigation Roadmap was released in December 2023. Since that time, attention to artificial intelligence (AI) has continued to grow at a rapid pace. Tens of billions of dollars have poured into AI projects, policymakers around the world have considered new AI policies, and OpenAI reports that each month more than 200 million people now use ChatGPT.

Signs of a changing climate continue to grow as well. Based on global average temperatures, July 22, 2024 was the warmest day ever recorded; 2023 was the warmest year ever recorded; and the 10 warmest years on record are the past 10 years. Yet global emissions of greenhouse gases continue to climb.

Can AI help cut emissions of greenhouse gases? This Roadmap explores that question. In this second edition of the Artificial Intelligence for Climate Change Mitigation Roadmap, a team of 25 co-authors builds on last year’s roadmap—comprehensively updating all old chapters, adding six new chapters and offering 5–10 specific, actionable recommendations in each chapter. The goal is to provide a useful resource for experts and non-experts alike.

Download Authors Page Download Preface

Five Key Takeaways

The 17 chapters and 334 pages of this Roadmap explore many topics in considerable detail, including current applications of AI in reducing GHG emissions, future possibilities, risks, barriers, policy options and the limits of current knowledge. For those of you interested in a quick summary of our main messages, here are five.

Artificial intelligence (AI) has the potential to make very significant contributions to climate change mitigation in the years ahead. This includes incremental gains (such as increasing output at solar farms and improving energy efficiency in buildings) and transformational gains (such as helping discover important new materials for clean energy technologies).

Greenhouse gas (GHG) emissions from computing operations for AI are less than 1%—and perhaps much less than 1%—of global GHG emissions. These emissions will very likely increase in the years ahead, in amounts that could be modest or significant. 

The main barriers to realizing AI’s potential to help reduce GHG emissions are lack of data and lack of trained personnel. Governments, companies and educational institutions should work together to overcome these barriers.

Trust in AI systems is essential for AI to deliver substantial benefits for climate change mitigation. For AI to be trustworthy and trusted, risks related to bias, privacy, misinformation, disinformation, safety, security and other issues must be addressed.

Every organization with a role in climate change mitigation should consider opportunities for AI to contribute to its work.

Download Takeaways
Executive Summary

The ICEF AI for Climate Change Mitigation Roadmap (Second Edition) has three parts.

Part I provides basic background on artificial intelligence and climate change.

Part II explores AI’s potential to help reduce greenhouse gas emissions in eight sectors: power, the food system, manufacturing, road transport, aviation, buildings, carbon capture and nuclear power.

Part III discusses cross-cutting issues, including large language models, greenhouse gas emissions monitoring, materials innovation, extreme weather response, greenhouse gas emissions and power demand from AI and government policy. A final chapter offers findings and recommendations.

Download Executive Summary
0.
Part 1
Introduction
Introduction to Artificial Intelligence
AI is the science of making computers perform complex tasks typically associated with human intelligence. Modern AI systems have far-reaching capabilities in at least four areas: detection, prediction, optimization and simulation. AI requires available and accessible data.
Download Chapter
Introduction to
Climate Change
Concentrations of heat-trapping gases in the atmosphere are now higher than at any time in human history. This is changing the Earth’s climate. Storms, heat waves and droughts have increased in frequency and intensity in recent decades, consistent with scientists’ predictions as heat-trapping gases accumulate in the atmosphere. Sea-level rise from climate change threatens coastal cities around the world.
Download Chapter
0.
Part 2
Sectors
Power
System
AI is a key tool for decarbonizing the power sector. At solar and wind power plants, AI is already helping speed permitting and increase output. AI can increase the capacity of transmission lines with dynamic line rating. Virtual power plants and demand response programs are starting to rely heavily on AI tools. Barriers include inaccessible data, lack of trained personnel and poor market design.
Download Chapter
Food
System
AI has significant potential to help reduce GHG emissions from food systems, while enhancing resilience. Key AI application areas include remote sensing for agricultural monitoring, modeling to optimize farm management decisions and accelerated breeding programs for climate-resilient crops. However, significant challenges persist, such as limitations in model interpretability and transferability, data biases and the risk of exacerbating existing inequalities in food systems.
Download Chapter
Manufacturing
AI has significant potential to help decarbonize manufacturing by optimizing existing industrial processes and operations. AI can help manufacturers adapt to production issues, avoid past mistakes, improve production yields, promote recycling, minimize energy consumption, adopt alternative energy sources and optimize manufacturing schedules and supply chains.
Download Chapter
Road
Transport
AI has significant potential to help reduce GHG emissions from road transport. AI can speed deployment of electric vehicles (EVs) by improving siting of charging infrastructure, extending EV battery life and helping operate vehicle-to-grid networks. AI provides critical support for intelligent transportation systems, helps promote modal shifts and plays a central role in operating autonomous vehicles (which can reduce GHG emissions through platooning and other measures).
Download Chapter
Aviation
AI has the potential to reduce aviation emissions and climate impacts in several ways. One especially promising approach is using AI to help predict when aircraft-induced condensation trails (contrails) will form and enable minor flight route changes to avoid them. AI can also predict key properties of novel formulations of sustainable aviation fuel (SAF) and help improve engine and aircraft design to increase fuel efficiency.
Download Chapter
Buildings
AI can play an important role in reducing CO2 emissions from buildings. In the design stage, AI can help improve energy efficiency, site placement and material choices. During construction, AI can help manage wastes and identify emission-reduction opportunities on site. When a building is operational, AI can optimize HVAC and other mechanical systems. Approaches must be adapted to diverse local contexts, especially to conditions in developing economies where the vast majority of building construction will take place in the decades ahead.
Download Chapter
Carbon
Capture
AI has the potential to significantly reduce costs and accelerate deployment of CCUS, including radical improvements in performance and dramatically faster project implementation. AI can help identify new materials for carbon capture and use, including sorbents, catalysts and membranes. AI applications, such as digital twinning, could dramatically improve efficiency and costs of facility design and operations.
Download Chapter
Nuclear
Power
AI is already being used to optimize fueling and maintenance of current-generation nuclear reactors and shows promise in aiding in the design of the advanced reactors. AI has many other potential applications in nuclear power, including helping interpret scans of irradiated concrete to reduce uncertainty about its condition. But the nuclear industry lacks large volumes of accessible data. Using AI in complex safety-critical systems requires careful planning.
Download Chapter
0.
Part 3
Cross-Cutting Topics
Large Language
Models
Large language models (LLMs) have captured the public’s imagination through the human-like output of popular products like ChatGPT. These LLMs are already helping mitigate climate change – such as by helping make sense of vast repositories of climate change information in multiple languages. In the future, LLMs can do even more to fight climate change – such as by serving as tutors in climate education, advancing basic science in climate change mitigation and helping accelerate permitting processes.
Download Chapter
Greenhouse Gas
Emissions Monitoring
Historically, GHG data have been fragmented and sometimes incomplete, with significant time lags. AI is now playing a critical role in overcoming this limitation by analyzing vast amounts of data from satellites and other technologies to provide more complete, near-real-time emissions monitoring. AI’s contributions are particularly notable in monitoring methane emissions. AI is also revolutionizing CO2 emissions tracking by integrating large data sets from different sectors.
Download Chapter
Materials
Innovation
Advanced materials with special properties are vital for decarbonization. Standard methods for identifying new advanced materials through computation are slow and require large computing resources. AI can speed these processes and reduce costs. Generative AI methods have been able to propose entirely new classes of advanced materials that had not previously been envisioned as relevant to decarbonization. 
Download Chapter
Extreme Weather
Response
AI-based forecasting models are becoming increasingly accurate, using far less time and energy and costing less than conventional forecast models. Thanks to these emerging capabilities, the role of AI in enhancing forecasting and enabling better early warning systems for extreme weather events is becoming increasingly important. AI is providing vital tools for disaster preparedness and response, especially in regions with limited forecasting capabilities today.
Download Chapter
Greenhouse Gas
Emissions from AI
GHG emissions from AI computation are currently less than 1%—and perhaps much less than 1%—of the global total. Data center power demand is growing steeply in many places around the world, due in part to demand for AI. Estimates of near-term growth vary widely. In the medium- to long-term, AI could result in either net increases or net decreases in GHG emissions. In part because AI is a transformational technology in the early stages of deployment, the range of uncertainty is enormous.
Download Chapter 15a Download Chapter 15b
0.
Part 4
Conclusions
Government
Policy
Governments play an important role in using AI for climate change mitigation—collecting data used in AI models, funding clean energy research programs that use AI tools, establishing policies that shape the use of AI in the power and transport sectors, and more. Governments also play an increasing role in managing risks from AI, which is essential in promoting trust in well-functioning AI systems. Governments can help realize AI’s potential to contribute to climate change mitigation with policies and programs in a range of areas. 
Download Chapter
Findings and
Recommendations
In this chapter we offer 12 findings concerning AI and climate change and 12 recommendations for realizing AI’s potential to contribute to climate change mitigation.
Download Chapter
Appendix
Additional Reading
For anyone interested in additional reading on artificial intelligence and climate change, we especially recommend these sources from the PDF linked below.
Download Appendix
Appendix
Recommendations from Each Chapter
Review our recommended actions to take from each chapter.
Download Appendix
Full Report
ICEF Artificial Intelligence for Climate Change Mitigation Roadmap (Second Edition, November 2024)