• Part I: Overview
• Part II: AI framework & applications for climate change
• Part III: Climate-AI landscape
• Part IV: Challenges, trends & prospects

Part I: AI technology & climate change problem

Climate change and digital transformation are the two most powerful trends of our century. The way in which we manage them, and their increasing interaction, will play a significant role in humanity’s future in the 21st century and beyond.

Artificial intelligence: Definition

AI, or artificial intelligence, refers to the development of computer systems that can perform tasks that typically require human intelligence, such as recognizing speech, making decisions, and identifying patterns in data. AI technologies include machine learning, deep learning, natural language processing, and computer vision, among others.

According to the OECD AI Principles, artificial intelligence encompasses any computer algorithm that makes predictions, recommendations, or decisions on the basis of a defined set of objectives. AI algorithms can be trained to learn and improve over time, allowing them to perform increasingly complex tasks with greater accuracy and efficiency. AI is increasingly being used in a wide range of applications, from virtual assistants and autonomous vehicles to medical diagnosis and financial analysis.

Currently, there is no universally accepted legal definition of AI. However, some countries and organizations have proposed their own definitions. For example, the European Commission defines AI as “systems that display intelligent behaviour by analysing their environment and taking actions — with some degree of autonomy — to achieve specific goals.”

Similarly, the United States National Institute of Standards and Technology (NIST) defines AI as “a system that can sense, reason, act and adapt to the real world, with goals, including learning and problem-solving.”

These definitions emphasize the ability of AI systems to perform tasks autonomously and to adapt to new situations, while also acknowledging the importance of goal-oriented behavior and the ability to learn from experience. As AI technology continues to evolve and become more complex, the definitions of AI may also evolve in response.

Types of AI

There are several types of AI, each with its own characteristics and capabilities. Here are some of the main types:

• Reactive Machines: These are the most basic type of AI systems that do not have the ability to learn from past experiences. They can only react to specific inputs in a pre-determined way. For example, a chess-playing program can only respond to specific moves made by the opponent.
• Limited Memory: These AI systems have some ability to learn from past experiences, but only in a limited way. They can use past data to make decisions in the present, but they do not have the ability to “remember” past experiences in the long term. Self-driving cars use this type of AI to learn from previous experiences and make real-time decisions based on the current situation.
• Theory of Mind: These AI systems have the ability to understand the emotions, beliefs, and intentions of other agents. They can predict the behavior of others based on this understanding. This type of AI is still in the early stages of development, but it has applications in fields such as psychology, sociology, and robotics.
• Self-Aware: These AI systems have a sense of self and are able to understand their own emotions and limitations. They can make decisions based on this self-awareness, and they may be able to improve their own abilities over time. This type of AI is still largely theoretical and is not yet widely used in practical applications.

Key features of AI

There are several features that are considered essential to AI systems. These include:

• Machine Learning: AI systems should be able to learn and improve their performance over time. Machine learning algorithms allow AI systems to analyze data, identify patterns, and learn from their experiences.
• Natural Language Processing: AI systems should be able to understand and interpret natural language, including speech and text. Natural language processing allows AI systems to communicate with humans in a more natural way.
• Computer Vision: AI systems should be able to interpret visual data, such as images and videos. Computer vision algorithms allow AI systems to analyze and understand visual information.
• Robotics: AI systems can be integrated with robots to create intelligent machines that can perform a wide range of tasks. Robotic technology allows AI systems to interact with the physical world and perform tasks that would be difficult or impossible for humans to do alone.
• Autonomy: AI systems should be able to operate independently and make decisions based on their programming and data inputs. Autonomy allows AI systems to perform tasks more efficiently and with greater accuracy.
• Adaptability: AI systems should be able to adapt to changing circumstances and learn from new experiences. Adaptability allows AI systems to improve their performance over time and become more effective at their tasks.

These features are not necessarily exclusive to AI, but they are often considered essential to the development of advanced AI systems that can perform complex tasks with a high degree of accuracy and efficiency.

Climate change: Overview of the problem

Climate change refers to a long-term shift in global or regional climate patterns, primarily caused by the increasing concentrations of greenhouse gases, such as carbon dioxide and methane, in the Earth’s atmosphere. These gases trap heat from the sun, which leads to a gradual warming of the Earth’s surface and atmosphere.

The effects of climate change can be seen in various ways, including rising global temperatures, melting glaciers and polar ice caps, rising sea levels, more frequent and severe weather events, such as droughts, floods, hurricanes, and wildfires, and changes in the timing and patterns of plant and animal life.

Human activities, such as burning fossil fuels, deforestation, and intensive agriculture, are the primary drivers of climate change. The consequences of climate change are far-reaching, affecting ecosystems, economies, and human health and well-being. Addressing climate change is therefore a critical global challenge that requires urgent action to reduce greenhouse gas emissions and adapt to the changes already underway.

Key strategies for combating climate change

There are many ways to combat climate change. Here are some key strategies:

• Reduce greenhouse gas emissions: The most important strategy for combatting climate change is to reduce greenhouse gas emissions, particularly carbon dioxide emissions from burning fossil fuels. This can be achieved through a combination of strategies, including transitioning to renewable energy sources, improving energy efficiency, and promoting sustainable transportation options.
• Promote sustainable land use: Deforestation and other land-use changes are major contributors to greenhouse gas emissions. Promoting sustainable land use practices, such as reforestation, conservation, and sustainable agriculture, can help reduce these emissions.
• Promote energy efficiency: Improving energy efficiency in buildings, appliances, and vehicles can significantly reduce greenhouse gas emissions while also saving energy and money.
• Adapt to the changing climate: As climate change is already happening, it’s important to prepare for and adapt to the impacts of a changing climate, such as sea-level rise, more frequent and severe weather events, and changing agricultural and ecological systems.
• Promote sustainable consumption and production: Encouraging sustainable consumption and production patterns, such as reducing waste, promoting circular economy practices, and choosing environmentally-friendly products, can help reduce greenhouse gas emissions and promote a more sustainable future.
• Support climate action at all levels: Combating climate change requires collective action at all levels, from individual action to national and international policies and agreements. Supporting and promoting climate action at all levels can help drive meaningful change.

These strategies require action and collaboration from individuals, businesses, communities, and governments to achieve meaningful progress in combatting climate change.

Organizations that are working to combat climate change:

• Intergovernmental Panel on Climate Change (IPCC): The IPCC is a scientific body established by the United Nations that assesses the science, impacts, and risks of climate change and provides policymakers with advice on response options.
• United Nations Framework Convention on Climate Change (UNFCCC): The UNFCCC is an international treaty that aims to prevent dangerous human interference with the climate system by stabilizing greenhouse gas concentrations in the atmosphere.
• World Wildlife Fund (WWF): WWF is an international conservation organization that works to protect natural habitats and promote sustainable living.
• The Nature Conservancy: The Nature Conservancy is a global conservation organization that works to protect ecologically important lands and waters.
• Greenpeace: Greenpeace is an international environmental organization that campaigns for solutions to environmental problems, including climate change.
• The Climate Group: The Climate Group is an international non-profit organization that works to accelerate climate action and promote a low-carbon economy.
• The Global Green Growth Institute (GGGI): GGGI is an international organization that supports emerging and developing countries in transitioning to a green economy and achieving sustainable development.
• The World Resources Institute (WRI): WRI is a research organization that works to address environmental challenges, including climate change, through research, analysis, and policy solutions.

These are just a few examples of the many organizations that are working to address climate change and promote sustainability.

Why is AI technology well-suited for applications related to climate change?

Climate change is a global emergency challenging scientists, engineers, and industry experts from a wide array of disciplines to use their knowledge and skills in pursuit of solutions to protect our planet. Not surprisingly, some of those solutions are likely to be made possible by artificial intelligence.

“Climate data sets are enormous and take significant time to collect, analyze, and use to make informed decisions and enact actual policy change,” says Jim Bellingham, a pioneer in autonomous underwater robotics systems and executive director of the Johns Hopkins Institute for Assured Autonomy. “Using AI to factor in elements of climate change that are constantly evolving helps us make more informed predictions about changes in the environment so that we can deploy mitigation efforts earlier.”

Artificial intelligence combines predictions based on trends and patterns with the extensive data collected. Models are at the core of prediction, but to rely on these models to make decisions, people must trust the models. Scientists are asking “what if” questions, and policymakers are weighing costs and benefits based on data collected and analyzed by researchers. AI is one tool that provides insights into where uncertainties come from related to climate change and that can help us understand what the models are telling us, which can feed back into better observation programs, improving the models, and even using AI as part of the model system. Assurance, or trust, is a key aspect of using AI.

AI technology is well suited for applications related to climate change for several reasons:

• Big data analysis: Climate change generates vast amounts of data, including data from satellites, weather stations, and other sources. AI can be used to analyze this data quickly and accurately, enabling a better understanding of the effects of climate change and informing response strategies.
• Predictive modeling: AI can be used to build predictive models of climate change impacts, such as sea level rise, crop yields, and extreme weather events. This can help policymakers and businesses anticipate the effects of climate change and develop effective response strategies.
• Optimization: AI can be used to optimize the use of renewable energy sources and reduce energy consumption in buildings and industrial processes, which can help to reduce greenhouse gas emissions.
• Automation: AI can be used to automate routine tasks in climate monitoring and response, freeing up human resources to focus on more complex tasks.
• Rapid response: AI can be used to rapidly respond to climate-related emergencies, such as natural disasters, by providing real-time data analysis and decision-making support.

Thus, AI’s ability to process vast amounts of data quickly and accurately, its ability to build predictive models, and its ability to optimize systems make it well-suited for applications related to climate change.

How is AI being used to address climate change?

Climate change is one of the most difficult scientific problems that humans have ever faced. It’s a phenomenally complex system with an enormous number of variables. When people talk about climate change, they tend to focus on the physical aspects of climate, such as the amount of carbon dioxide in the atmosphere, temperatures, precipitation levels, and wind patterns. But these characteristics are all shaped by a living planet that is constantly changing.

Climate data sets are enormous and take significant time to collect, analyze, and use to make informed decisions and enact actual policy change. Using AI to factor in elements of climate change that are constantly evolving helps us make more informed predictions about changes in the environment so that we can deploy mitigation efforts earlier.

Areas of action for governments in supporting the responsible use of AI in the context of climate change. (Source: Global Partnership on AI Report)

AI is being used in various ways to address climate change, including:

• Climate modeling and prediction: AI is being used to improve climate modeling and prediction, which can help to better understand the impacts of climate change and develop more effective response strategies.
• Renewable energy optimization: AI is being used to optimize the use of renewable energy sources, such as solar and wind power, by predicting energy production and consumption patterns and improving energy storage and distribution.
• Carbon capture and storage: AI is being used to optimize carbon capture and storage technologies, which can help to reduce greenhouse gas emissions from power plants and other industrial sources.
• Sustainable agriculture: AI is being used to improve agricultural practices by optimizing crop yields, reducing fertilizer use, and improving soil health, which can help to reduce greenhouse gas emissions and promote food security.
• Energy efficiency: AI is being used to improve energy efficiency in buildings and industrial processes by identifying areas for improvement and optimizing energy use.
• Natural disaster management: AI is being used to improve natural disaster management by predicting and mitigating the impacts of extreme weather events, such as floods and hurricanes.
• Climate finance: AI is being used to improve climate finance by analyzing financial data to identify investment opportunities and measure the impact of climate-related investments.

These are just a few examples of how AI is being used to address climate change. AI has the potential to significantly improve our ability to understand and respond to climate change, and its use is likely to increase in the coming years as the urgency of the climate crisis grows.

Part II: AI framework & applications for climate change

The changing climate will have a major impact on environmental, social, and economic systems around the globe. Mitigation is therefore critical, including efforts to achieve net-zero emissions by 2050. However, minimizing the harm will also require increasing efforts at adaptation and resilience. These efforts will require support from activities such as research, finance, and education.

AI technology has the potential to be implemented in many ways to combat climate change. AI has the potential to play a significant role in reducing greenhouse gas emissions and mitigating the effects of climate change. AI as a tool is uniquely positioned to help manage these complex issues. Due to its capacity to gather, complete, and interpret large, complex datasets on emissions, climate impact, and more, it can be used to support all stakeholders in taking a more informed and data-driven approach to combating carbon emissions and building a greener society. It can also be employed to reweight global climate efforts toward the most at-risk regions. AI can play a significant role in addressing the climate change problem in several ways:

• Data Analysis: AI can analyze large amounts of data on climate patterns, greenhouse gas emissions, and other factors that contribute to climate change. By analyzing this data, AI can help researchers better understand the causes and effects of climate change and identify potential solutions.
• Energy Management: AI can optimize energy consumption in buildings and other structures. By analyzing data on energy usage, weather patterns, and other factors, AI algorithms can predict energy demand and adjust heating, cooling, and lighting systems accordingly. This can lead to significant reductions in energy usage and greenhouse gas emissions.
• Renewable Energy: AI can optimize the use of renewable energy sources such as solar and wind power. By analyzing data on weather patterns, energy generation, and energy storage and distribution systems, AI algorithms can increase the reliability and efficiency of renewable energy systems, making them more competitive with traditional energy sources.
• Carbon Capture: AI can improve the efficiency and effectiveness of carbon capture technologies. By analyzing data on factors such as temperature, pressure, and gas flow, AI algorithms can optimize the capture and storage of carbon dioxide emissions.
• Agriculture: AI can optimize agricultural practices and reduce greenhouse gas emissions from farming. By analyzing data on soil quality, weather patterns, and crop yields, AI algorithms can optimize planting, fertilization, and irrigation, leading to increased crop yields and reduced use of fertilizers and other inputs that contribute to greenhouse gas emissions.
• Transportation: AI can optimize transportation systems and reduce greenhouse gas emissions from vehicles. By analyzing data on traffic patterns, road conditions, and weather, AI algorithms can optimize route planning and reduce congestion, leading to reduced fuel consumption and emissions from vehicles.

AI applications for climate change.

Boston Consulting Group (BCG) recently surveyed global public- and private-sector climate and AI leaders about their views on the potential of AI as a tool in the fight against climate change as well as the roadblocks that prevent its adoption. Most respondents had authority over both topics. Respondents represented private- or public-sector organizations with at least 1,000 employees or at least $100 million in global annual revenues. The survey sample covered 14 countries: Argentina, Australia, Brazil, Chile, China, France, Germany, India, Japan, New Zealand, South Africa, Spain, the UK, and the US.

In response, 87% replied that advanced analytics and AI is a helpful tool in the fight against climate change today. In addition, 67% of those in the private sector stated that they want governments to do more to support the use of AI in fighting climate change.

Public- and private-sector climate and AI leaders who oversee the AI and climate topics support using AI to combat climate change, but only 43% have a vision of how to use it. (Source: BCG)

The questions addressed respondents’ views regarding the use of advanced analytics and AI to fight climate change, as well as associated organizational obstacles.

Leaders cite insufficient expertise, availability, and confidence as obstacles to using AI in their climate change efforts. (Source: BCG)

While there are some areas where AI solutions are well-established and ready for broad application — such as emissions measurement and the monitoring of natural carbon sinks — existing AI-related climate-change solutions today are more typically scattered, difficult to access, and lacking the resources to scale. In addition, not all organizations are yet actively engaged in climate and AI topics, while even leaders already engaged in this space face many obstacles to climate AI use. In fact, 78% of those surveyed by BCG cite low access to AI expertise, whether inside or outside their organization, as an obstacle; 77% report a lack of available AI solutions; and 67% say they face a lack of organizational confidence in AI data and analysis.

Framework of using AI to combat climate change

The use of AI offers an opportunity to make meaningful change in this critical moment, whether through mitigation, adaptation, and resilience or by supporting the fundamentals of overall climate efforts:

• Mitigation. AI can be employed to help measure emissions at both the macro and micro levels, reduce emissions and greenhouse gas (GHG) effects, and remove existing emissions from the atmosphere. In BCG’s experience, for example, AI can be used to help reduce GHG emissions equal to 5% to 10% of an organization’s carbon footprint, or a total of 2.6 to 5.3 gigatons of CO2e if scaled globally.
• Adaptation and Resilience. AI can be applied to improve hazard forecasting for regionalized long-term events, such as sea-level rise, and for immediate, extreme events, such as hurricanes, among other possibilities. These applications include the management of vulnerability and exposure, such as by developing infrastructure that can minimize the impact of climate hazards.
• Fundamentals. AI can also be used to bolster efforts across climate research and modeling; climate finance; education, nudging, and behavior change, such as by powering personalized tools to estimate an individual’s carbon footprint or making recommendations for environmentally friendly purchases.

Framework of using AI to combat climate change. (Source: BCG)

Mitigation

One of the most critical uses of AI is in the measurement, reduction, and removal of emissions and GHG effects. Increased efforts in these areas are crucial to ensuring that we meet the goals of the Paris Agreement.

Measurement: Measuring emissions, both in the overall environment and at the level of individual products and actions, allows to take stock of current situation and forecast future trends. It also allows us to prioritize the abatement efforts with the highest potential to reduce emissions, at both the macro level and the micro level:

• Macro Level. Measurements of overall environmental emissions are important inputs to models that simulate future climate scenarios. Solutions can use AI to support such models through more precise measurements, for example, or through the processing of remote-sensing data from satellites. They can also use it to monitor the impact of new and existing climate policies, thereby enabling teams to refine their climate models and mitigation strategies.
• Micro Level. Emissions measurements at the level of individual products or actions — including emissions created at any point in a supply chain — can help producers understand their products’ carbon footprint; track their progress toward environmental, social, and governance (ESG) targets; or identify the leading ways to abate their scope 1, 2, and 3 emissions.

According to the Carbon Disclosure Project’s Global Supply Chain 2021 report, only 38% of companies are engaging with their suppliers on climate change — even though supply chains cause on average 11.4 times as many emissions as companies’ direct operations.

Reduction: Rapid efforts to cut the level of ongoing emissions and the consequent GHG effects — from optimizing transportation networks to supporting research on new technologies — are essential to change the climate. There are three components to reduction:

• Reducing Emissions Intensity. AI solutions can be used to support the switch toward new energy sources. Supply forecasting for solar energy can help us identify areas where there is potential for increasing the use of solar energy, thus reducing GHG emissions.
• Reducing Emissions-Generating Activities. AI can also reduce emissions by optimizing supply chains through improved demand prediction (to combat overproduction) or efficient transportation of goods (such as shortening delivery times and minimizing energy use). This can be done using data to generate models that predict demand or optimize transportation routes.
• Reducing the Greenhouse Effect. If policymakers turn towards geoengineering solutions to curtail the effects of climate change, AI will be an essential tool for accelerating chemistry research and can help us to develop new materials and processes that result in less greenhouse gas emissions. Additionally, encouraging behavioral change can reduce energy consumption and lower emissions.

Removal: Removing emissions from the atmosphere is a crucial pillar of the effort to achieve a net-zero future, whether by monitoring natural carbon sinks, such as forests or by supporting technological removal efforts, such as direct air capture (DAC). There are two main types of removal:

• Environmental Removal. Natural ecosystems such as forests, algae, and wetlands play a central role in atmospheric carbon removal. Monitoring these ecosystems requires gathering and processing large amounts of data, a situation in which AI is very effective.
• Technological Removal. Another way to remove carbon from the atmosphere is through scientific advances such as DAC, which filters and captures CO2 from the air as it passes through a machine. The captured CO2 is either stored underground — in saline aquifers, for example — or prepared for industrial use. An alternative to DAC is bioenergy with carbon capture and storage (BECCS), which generates energy out of biomass, including wood and agricultural waste, and captures the resulting CO2 for storage underground or for industrial use. Solutions can apply AI to assess capture and storage locations, monitor leakage, and optimize the industrial processes and materials used for carbon capture.

Adaptation & Resilience

Adapting to climate change is a critical undertaking for policymakers and the public, as it boosts resilience to the effects of long-term climate trends and extreme events. Solutions can apply AI to promote adaptation and resilience, particularly in hazard forecasting and vulnerability and risk-exposure management.

Hazard Forecasting: AI is well suited to help in forecasting and projecting climate-related hazards, whether in the form of improved long-term projections of regionalized events, such as sea-level rise, or in the form of early warning systems for extreme weather events:

• Projecting Regionalized Long-Term Trends. Accurate projections of regionalized events can support efforts by decision-makers to adapt to climate change. Being able to plan ahead and know which areas are most likely to be affected can help planners devise long-term strategies to secure local livelihoods and support sustained economic growth. Solutions can use AI in this vein to create large-scale regional-level climate models, predict crop and fishing yields, or provide city-level climate indicators to local governments.
• Building Early Warning Systems. In a shorter time frame, forecasting individual extreme weather events can limit their devastating effects and thereby save lives. The use of AI in such forecasts has revolutionized forecasting accuracy and permitted real-time data processing and analysis — such as in reviewing satellite images to spot forest wildfires and even to predict their likelihood.

Vulnerability and Risk-Exposure Management: Another key component of adaptation and resilience efforts is the strategic management of vulnerability and risk exposure, such as by planning the development of infrastructure in threatened areas so as to minimize the impact of climate hazards. AI can be employed in the following ways:

• Managing Crises. Disaster response is a critical aspect of vulnerability management, enabling public- and private-sector leaders to make more informed decisions under circumstances where they might previously have had to operate with sparse or limited information. In such instances, AI can be used to efficiently combine natural-language processing with operations that involve gathering and combining information from multiple sources.
• Strengthening Infrastructure. The increasing frequency and intensity of natural disasters caused by climate change are putting existing infrastructure to the test. The resilience of health systems, transportation infrastructure, education, and information media is essential if we are to avoid the direst consequences of natural disasters. Solutions can use AI to strengthen infrastructure through predictive maintenance of structures such as roads, bridges, and electricity distribution systems; through monitoring of water quality and air pollution; and even through support for design projects and risk identification.
• Protecting Populations. On average, 20 million people every year must leave their homes and relocate to other areas within their countries as a result of extreme weather disasters, according to the UNHCR.
• Preserving Biodiversity. Humans are part of a vast ecosystem. When a natural disaster strikes, that ecosystem’s biodiversity is put at risk of imbalance and collapse. AI can be enlisted to control intelligent irrigation systems that minimize the adverse impact of weather events on agricultural production, identify crop diseases early, or promote the transfer of knowledge between biomes.

Fundamentals

AI can be used to support research and education efforts on climate change, helping stakeholders understand the risks and implications and share what they learn. All such efforts reinforce and magnify progress toward mitigation and toward adaptation and resilience.

Climate Research and Modeling: Modern climate research and modeling lie at the heart of all policy discussions about mitigation and about adaptation and resilience. They underpin our scientific understanding of the future consequences of climate change at the local, regional, and global levels.

Climate Finance: The broad field of climate finance includes investing in and financing climate-change initiatives — funding that is essential if these initiatives are to achieve the scale necessary to have a meaningful impact. Another angle of climate finance involves carbon markets, which create monetary incentives for climate-friendly behavior by allowing firms to trade carbon offsets — instruments that represent the reduction, avoidance, or sequestration of CO2e in metric ton units.

Education, Nudging, and Behavior Change: Educating the public about the practical realities and threats of global warming can foster change from the ground up, reshaping individuals’ behavior. AI can be used in this vein to help power personalized tools for estimating carbon footprints and for making individual recommendations for online learning courses or climate-friendly purchases. AI is already ubiquitous in the domain of generating algorithmic recommendations for videos to watch or products to buy; similar algorithms can be applied to understanding behaviors and nudging citizens toward climate-friendly activities.

Public- and private-sector leaders see the greatest business value in reducing and measuring emissions. (Source: BCG)

Key AI applications & capabilities for climate

Within each of the uses (mitigation, adaptation & resilience, fundamentals), AI can be employed as part of five key applications:

• Gather, complete, and process complex datasets on emissions, climate effects, and more: AI-powered data approaches, such as the Internet of Things (physical objects with embedded technology that exchange data with devices or systems over the internet), give rise to novel data on consumer behaviors and preferences, industrial processes, and supply chains. This data can be used to train algorithms that uncover behavioral patterns and preferences at scale. Moreover, AI allows the use of new types and combinations of data for analysis, including images, text, and sounds.
• Strengthen planning and decision-making: AI can be used to analyze and compare complex, multilayered problems — for example, when civil servants need to design an optimal agricultural policy to deal with more frequent droughts or more severe flooding. The impact of such policies is manifold and interlinked, making this application ideal use of AI.
• Optimize processes: AI has long been used to solve optimization problems in various domains across the private and public sectors. For example, it may be employed to optimize a supply chain by lowering the carbon footprint of individual products or to optimize the coordination of crisis response teams in the wake of extreme weather events.
• Support collaborative ecosystems: Global cooperation is essential if we are to respond effectively to the challenges posed by climate change, and AIpowered language tools can be used to support global communications and coordination on climate goals. Further, vertical data sharing between entities can be made faster and more secure through AI. For example, sharing data on GHG emissions between firms can contribute to a comprehensive view of the carbon footprint of entire value chains — a critical monitoring step for companies and individuals. And using AI can help automate this reporting, impute missing values based on similar production processes, and analyze the information on a wide scale.
• Encourage climate-positive behavior: AI can be applied in recommendation systems to automatically weight suggestions toward climate-friendly options. Similarly, AI-powered optimizations can encourage climate-friendly processes by imposing penalties for alternatives that have a larger carbon footprint.

There are also several key themes for how AI can accelerate climate action:

• Distilling raw data into actionable information. AI can identify useful information within large amounts of unstructured data, often by scaling up annotations that humans could provide more laboriously. For example, AI can analyze satellite imagery in order to pinpoint deforestation or identify areas of cities vulnerable to coastal inundation, or can filter large databases of corporate financial disclosures for climate-relevant information.
• Improving predictions. AI can use past data to predict what will happen in the future, sometimes also incorporating auxiliary information. For example, AI can provide minute-level forecasts of solar power generation to help balance the electrical grid, or predictions of agricultural yield as extreme weather threatens food security.
• Optimizing complex systems. AI methods are good at optimizing for a specific objective given a complicated system with many variables that can be controlled simultaneously. For example, AI can be used to reduce the energy needed to heat and cool a building, or to optimize freight transportation schedules.
• Accelerating scientific modeling and discovery. AI can accelerate the process of scientific discovery itself, often by blending known physics-based constraints with approximations learned from data. For example, AI can suggest promising candidate materials for batteries and catalysts to speed up experimentation, and can quickly simulate portions of climate and weather models to make them more computationally tractable.

In these ways, AI can help stakeholders — from businesses and governments to NGOs and investors — pursue a more informed and data-driven course, while offering them opportunities to create meaningful change in this critical moment.

Key areas where AI can facilitate climate action

Most of the applications are the subject of active development, and many are already beginning to be deployed.

Selected AI-for-climate applications within buildings and cities (Rolnick et al. 2019)

• Electricity systems. AI can enable significant emissions reductions in electricity systems, across a wide range of applications. To balance power grids efficiently, and thereby enable the integration of large amounts of renewables, it is essential to forecast both electricity supply and demand, a function that AI can provide. AI can also improve algorithms for electricity scheduling and storage, as well as the management of microgrids in areas with decentralized systems. AI can improve the operations of renewable energy generators such as wind turbines and solar panels and can pinpoint methane leaks in natural gas pipelines. AI is also being used to accelerate the discovery of new energy-relevant materials, such as those used in photovoltaic cells, batteries, and electrofuels.
• Buildings and cities. AI can increase the efficiency of energy use in buildings and urban environments. In cases where certain data on the built environment has not been collected, AI can be used to label infrastructure in satellite imagery. AI can infer energy use from building properties, as well as interpret data from smart meters. Within smart buildings, AI can optimize building functions such as heating and lighting to conserve energy. For city-scale optimization, AI can be used in soft sensor systems and data mining. AI can also help cities with waste management to reduce methane emissions associated with landfill and Wastewater.
• Transportation. AI can help decarbonize transportation in many ways. AI can improve estimates of transportation usage, as well as model demand for public transportation and infrastructure. AI can optimize freight routing and scheduling and can increase the utilization of low-carbon options such as trains. To advance the adoption of electric vehicles, AI can optimize charging protocols and locations, as well as inform the design of batteries and next-generation fuels. AI is also an integral part of autonomous vehicle technologies, though the climate impacts here are uncertain: self-driving personal vehicles may increase emissions by making driving easier, while autonomous buses could decrease emissions by pooling passengers and integrating with public transportation.
• Heavy industry and manufacturing. AI can reduce emissions associated with heavy industry in numerous ways. AI can be used in adaptive control and process optimization to reduce the energy consumed by industrial processes, as well as in demand response to schedule such processes to reduce emissions intensity. AI-enabled predictive maintenance and digital twins can also boost efficiency and in some cases reduce leaks of greenhouse gases such as methane. Increasingly, AI is being used in the discovery of materials such as catalysts, which may reduce the energy needs of certain chemical processes. AI can also help optimize recycling processes and waste sorting for energy-intensive materials such as aluminum and steel, which can help avoid emissions associated with mining and processing virgin material.

Selected AI-for-climate applications within agriculture, forestry, and other land use. (Rolnick et al. 2019)

• Agriculture. AI can support both mitigation and adaptation efforts in agriculture. Precision agriculture involves the use of AI in automated tools that are responsive to variability within a crop, offering the potential for increased efficiency and reduced greenhouse gas emissions associated with agricultural chemicals and land use. (It is worth noting that incentives in agriculture are not always aligned with GHG emissions reduction, though there can be significant overlap.) From the standpoint of adaptation, remote sensing tools for crop monitoring and yield prediction can advance food security in the face of droughts and other extreme weather.
• Forestry and other land use. AI can facilitate responsible land use practices and nature-based solutions for carbon sequestration, in several ways. AI tools are being used together with satellite imagery in carbon stock estimation for informing land management decisions and for calculating carbon offsets. AI is also being used to help track deforestation and other land use changes, as well as in drones to accelerate afforestation. There are also numerous uses of AI in predicting the risk and spread of wildfires.
• Climate science. AI can advance models of climate, weather, and other Earth systems. It can help provide data for such models by calibrating sensors or by inferring properties such as ice cover from raw data like satellite imagery. AI can also provide fast approximations to certain physics simulations within climate and weather models that would otherwise be prohibitively time-intensive to run. Such approximate simulations can be useful both in improving overall models and in increasing the spatial resolution at which they can practically be run, thereby providing more localized predictions of risk.
• Societal adaptation. AI can aid in societal resilience to the effects of climate change. AI tools can pinpoint vulnerable locations and target infrastructure improvements where they are most needed, as well as enable predictive maintenance to avert failures. Methods from AI for healthcare can improve public health models as well as a societal response to climate-influenced pandemics and other diseases. In the event of disasters such as storms, floods, and fires, AI can inform relief efforts by improving maps and identifying at-risk individuals.
• Ecosystems and biodiversity. AI can support biodiversity preservation in the face of a changing climate. AI methods are increasingly being incorporated into sensors used to monitor wildlife, remote sensing tools for assessing ecosystem change, and recognition systems used to identify species from visual or audio data. AI is also beginning to be used in parsing ecological information, for example from citizen science databases.
• Climate policy. Many of the above applications of AI can be valuable for policymaking by providing data useful in informing policy decisions. However, AI can also inform climate policy in other ways. AI can be incorporated in models used for assessing policy options, and can also be used in causal inference to help assess the efficacy of policies that have been executed.

Part III: Climate-AI landscape

Global climate-AI initiatives

AI for the Planet

AI for the Planet is a global initiative launched by Microsoft in 2019. The initiative is focused on using AI to address some of the world’s most pressing environmental challenges, including climate change, biodiversity loss, and water scarcity. AI for the Planet brings together a community of organizations, researchers, and individuals who are committed to using AI to drive positive environmental impact. The alliance was created by Startup Inside, with Boston Consulting Group and BCG GAMMA as knowledge partners, in collaboration with four other organizations: the AI for Good Foundation; the United Nations Development Programme (UNDP); the United Nations Educational, Scientific and Cultural Organization (UNESCO); and the UN Office of Information and Communications Technology (OICT). The initiative supports research and development of AI-based solutions for environmental challenges, as well as collaborations between different stakeholders to share knowledge and best practices.

The coalition has several core objectives:

• Promote innovation in applying advanced analytics and artificial intelligence (AI) to climate challenges, supported by global experts from academia, startups, and the public and private sectors.
• Act as a global platform for identifying and prioritizing the leading tools and uses for AI in addressing the climate crisis.
• Identify and champion the most promising solutions to address climate change mitigation, as well as adaptation and resilience, especially in the Global South, giving visibility and recognition of the solutions.
• Ensure impact at scale through concrete and measurable actions such as building access to funding and to practitioners on the ground.
• Facilitate the development of networks among project teams, investors, and experts in the field — including startups, corporations, and the public sector.

As part of the AI for the Planet initiative, Microsoft has committed to becoming carbon negative by 2030 and removing all of its historical carbon emissions by 2050. The company is also investing in AI-based solutions to reduce its own carbon footprint and to help other organizations do the same.

AI for Good Foundation

AI for Good Foundation is a non-profit organization that is dedicated to using artificial intelligence to address some of the world’s most pressing problems. The foundation was established in 2015 by David and Daniel Feinberg, two brothers with a background in technology and philanthropy. The AI for Good Foundation’s mission is to support the research and development of AI-based solutions that can have a positive impact on society. The foundation focuses on a variety of areas, including healthcare, education, and environmental sustainability.

In the area of environmental sustainability, the AI for Good Foundation supports the research and development of AI-based solutions to address climate change, biodiversity loss, and other environmental challenges. The foundation works with a variety of partners, including academic institutions, non-profit organizations, and private companies, to develop innovative solutions that leverage the power of AI to create a positive environmental impact.

The AI for Good Foundation is an important organization that highlights the potential of AI to drive positive social and environmental impact. By supporting research and development of AI-based solutions, the foundation is helping to create a more sustainable and equitable future for all.

Climate Change AI (CCAI)

Climate Change AI (CCAI) is a non-profit organization that aims to accelerate the use of machine learning and AI to tackle climate change. The organization was founded in 2019 by a group of researchers and practitioners from leading universities and technology companies, including Google, Microsoft, and MIT. CCAI’s mission is to mobilize the AI community to develop and deploy AI solutions to address climate change, and to ensure that these solutions are accessible and beneficial to all. To achieve this mission, CCAI is working on a variety of initiatives, including:

• Research: CCAI conducts research on how machine learning and AI can be used to address climate change, and shares its findings with the wider community through publications and events.
• Education and outreach: CCAI provides training and education on AI and climate change, and works to raise awareness of the potential of AI to address this global challenge.
• Collaboration: CCAI facilitates collaborations between researchers, practitioners, and policymakers to develop and deploy AI solutions to address climate change.
• Advocacy: CCAI advocates for policies and investments that support the development and deployment of AI solutions to address climate change, and works to ensure that these solutions are ethical, transparent, and equitable.

CCAI is an important organization that is helping to accelerate the use of AI to address climate change and to ensure that these solutions are accessible and beneficial to all.

The Global Partnership on AI (GPAI)

The Global Partnership on AI (GPAI) is a multilateral initiative launched in 2020 to support the responsible development and deployment of AI. The initiative is comprised of a group of leading AI research institutions, companies, and policymakers from around the world. In December 2020, GPAI proposed the creation of a Committee on Climate Action and Biodiversity Preservation to start addressing questions regarding AI’s potential to support climate action.

GPAI’s mission is to promote the development and deployment of AI technologies that are trustworthy, transparent, and socially responsible. The initiative focuses on a range of AI-related issues, including privacy, security, and bias, and seeks to promote international cooperation and collaboration on these issues. To achieve its mission, GPAI has established working groups focused on specific areas of AI policy and development, including data governance, the responsible use of AI in healthcare, and the use of AI to address climate change. The initiative also conducts research, organizes events and workshops, and engages with stakeholders from a wide range of sectors and regions.

GPAI is an important initiative that highlights the need for international cooperation and collaboration on AI-related issues. By bringing together leading researchers, policymakers, and stakeholders from around the world, GPAI is helping to ensure that AI is developed and deployed in a way that is responsible, ethical, and beneficial to all.

The Coalition for Digital Environmental Sustainability (CODES)

The Coalition for Digital Environmental Sustainability (CODES) is a global initiative launched in 2019 that aims to leverage digital technologies to address environmental sustainability challenges. The initiative focuses on a range of issues, including climate change, biodiversity loss, and resource efficiency, and seeks to promote the use of digital technologies to address these challenges. The initiative is a collaboration between a range of stakeholders, including tech companies, civil society organizations, and academic institutions. CODES has established working groups focused on specific areas of digital environmental sustainability, including sustainable finance, sustainable cities, and sustainable agriculture. The initiative also conducts research, organizes events and workshops, and engages with stakeholders from a wide range of sectors and regions.

Three strategic shifts outlined in the CODES action plan launched in June 2022: enable alignment, through networks and sharing of best practices; mitigate negative impacts, by promoting responsible AI; and accelerate innovation, by helping bring effective climate AI solutions to scale.

CODES is an important initiative that highlights the potential of digital technologies to support environmental sustainability. By bringing together a diverse range of stakeholders, CODES is helping to promote collaboration and innovation in this important area and is contributing to the global effort to address environmental challenges.

Companies working on AI to combat climate change

“Companies that put AI at their core are far more likely to be contributing positively to climate resilience, adaptation, and mitigation efforts than those who do not,” states James Hodson, CEO of AI for Good Foundation.

There are many companies that are working on AI solutions to help address the climate change problem:

• IBM: IBM is working on a variety of AI solutions to address climate change, including a system that uses machine learning to optimize wind turbine performance and a tool that uses AI to help cities reduce their greenhouse gas emissions.
• Microsoft: Microsoft is developing an AI tool that uses machine learning to optimize energy usage in buildings. The tool analyzes data on energy consumption and weather patterns to predict energy demand and adjust heating, cooling, and lighting systems accordingly.
• Google: Google is using AI to improve the efficiency of its data centers, which are major energy consumers. The company is also developing an AI tool to optimize the use of renewable energy sources in its data centers.
• CarbonCure Technologies: CarbonCure Technologies is using AI to optimize the use of carbon dioxide in concrete production. The company’s AI tool analyzes data on factors such as temperature, humidity, and carbon dioxide concentration to optimize the use of the gas in the concrete production process.
• Cognitivescale: Cognitivescale is developing an AI tool that uses machine learning to predict the impact of climate change on various industries, such as agriculture and transportation. The tool analyzes data on climate patterns, economic trends, and other factors to identify potential risks and opportunities.
• ClimateAi: ClimateAi is using AI to develop climate risk models that can help companies and governments better understand and prepare for the effects of climate change. The company’s models use machine learning to analyze data on climate patterns, sea level rise, and other factors to predict the likelihood and severity of climate-related events.
• Blue Sky Analytics: based in The Hague, Netherlands, is a climate-tech company specializing in converting satellite data into environmental intelligence. The company’s API-based collection of environmental datasets uses satellite data, AI, and the cloud to provide insights into various topics related to the planet and its health.
• One Concern: based in California, USA, uses AI to estimate damage from natural phenomena. The company takes a holistic approach to uncover risk exposure and building resilience, not only considering the climate risk and disaster exposure of a single building but also the networks it depends on, such as transport links and power grids.
• Cloud to Street: based out of New York, is a company that uses satellites and AI to track floods in near real-time anywhere on Earth. The company runs a global flood database offering insights into flood exposure worldwide.
• Prospera: a Tel-Aviv-based company, is a developer of machine vision technologies designed to monitor and analyze plant development, health, and stress. The company’s technology captures multiple layers of crop field data, including climate and visual data, to spot anomalies sooner.
• EXCI: EXCI, based in Maroochydore, Australia, is a bushfire detection technology company that uses AI models to fuse data from satellites and ground-based sensors. This provides persistent systematic surveillance of wildfires, empowering firefighters with the intelligence to efficiently manage and fight them. The company is based in Maroochydore, Australia.
• Kuzi: Kuzi is a Kenyan company that uses artificial intelligence to predict the breeding, occurrence, and migration routes of desert locusts across the Horn of Africa and Eastern African countries. The company’s AI-powered tool uses satellite data, soil sensor data, ground meteorological observation, and machine learning to make its predictions. These are just a few examples of the many companies that are working on AI solutions to help address the climate change problem.

These solutions are just a few illustrative examples of how AI is being used to adapt to and mitigate the effects of climate change today.

“The next frontier in AI for climate will be decision support tools and behavioral incentivization — pushing people, companies, and governments to do the right thing because it’s in their best interests,” according to James Hodson.

AI startups working on the climate change

There are numerous AI startups working to address climate change in a variety of ways. These startups are typically focused on developing innovative AI-based solutions to help reduce greenhouse gas emissions, mitigate the effects of climate change, and accelerate the transition to a more sustainable economy. Some examples include:

• Verdigris: This startup uses AI to monitor energy consumption in commercial buildings, identifying areas where energy can be saved and reducing greenhouse gas emissions.
• WattTime: This startup uses AI to predict the carbon intensity of electricity in real-time, allowing consumers and businesses to make more sustainable energy choices.
• Ecolytiq: This startup uses AI to analyze supply chain data and identify opportunities to reduce emissions, helping businesses to transition to more sustainable practices.
• Tomorrow.io: This startup uses AI to provide hyperlocal weather forecasts, helping businesses and governments to better prepare for extreme weather events caused by climate change.

These are just a few examples of the many AI startups working to address climate change. As the urgency of the climate change problem continues to grow, we can expect to see more and more startups focused on developing innovative solutions that leverage the power of AI to create a more sustainable future.

European AI startups acting for climate

The European Applied AI Alliance is embarking on its first initiative to identify European champions of AI for Environment.

European AI & Climate Landscape (Source: appliedAI )

• Clim8 Invest: A UK-based startup that has developed an AI-powered investment platform focused on sustainable investing.
• Enerbrain: An Italian startup that has developed an AI-powered building management system that reduces energy consumption and improves indoor air quality.
• Hepta Airborne: A Swedish startup that uses AI and drones to monitor and analyze forest health and predict wildfire risk.
• Plan A: A German startup that has developed an AI-powered platform to help businesses measure, reduce, and offset their carbon footprint.

These startups are leveraging AI to develop innovative solutions that support environmental sustainability and contribute to the global effort to combat climate change.

Part IV: Challenges, trends & prospects

Challenges in using AI to address climate change

While AI has the potential to play a critical role in addressing climate change, there are also several challenges associated with using this technology in this context. Some of the key challenges include:

• Data availability and quality: AI relies heavily on data, and to effectively use this technology for climate change, there needs to be access to high-quality data that is relevant and up-to-date. However, in many cases, such data is not readily available or may be incomplete, which can limit the effectiveness of AI algorithms.
• Interpretability and transparency: In many cases, AI models are complex and difficult to interpret, which can limit their usefulness in the context of climate change. It can be difficult to understand why an AI model is making a particular decision, which can make it challenging to take action based on its recommendations.
• Bias: AI models can be biased, either because of biased training data or because of the inherent biases of the algorithms themselves. This can be particularly problematic in the context of climate change, where decisions based on biased data could have significant environmental and social consequences.
• Adoption and implementation: Even if effective AI solutions are developed to address climate change challenges, there may be challenges associated with their adoption and implementation. This could include challenges related to cost, infrastructure, and regulatory frameworks.
• Limitations of AI: AI is a powerful technology, but it also has limitations. There may be some climate change challenges that simply cannot be effectively addressed using AI, and in some cases, other approaches may be more effective.

While AI holds significant promise in the context of climate change, there are also many challenges that must be addressed to effectively leverage this technology in this critical area.

Challenges in using AI to address climate change.

AI’s negative impacts on climate

While AI has the potential to be a powerful tool in combating climate change, it can also have negative impacts on the climate if not used responsibly:

• Energy consumption: AI requires significant amounts of energy to power its algorithms and computing systems, and this can contribute to greenhouse gas emissions if the energy used is not from renewable sources.
• Data centers: The large-scale computing systems required to support AI applications, such as data centers, can also have a significant carbon footprint due to their high energy consumption and cooling requirements.
• E-waste: As AI technologies become more advanced and newer systems are developed, there is a risk of e-waste accumulation from outdated or discarded equipment, which can release harmful chemicals into the environment.
• Bias and inequality: There is also a risk that AI systems may perpetuate existing social and economic inequalities, which can contribute to further environmental degradation and harm marginalized communities that are often most impacted by climate change.

To address these potential negative impacts, it’s important for those working with AI to take a responsible and sustainable approach. This includes using renewable energy sources, designing more efficient data centers, minimizing e-waste, and addressing issues of bias and inequality in the development and implementation of AI systems.

Trends in using AI to address climate change

• Climate modeling: AI is increasingly being used to improve climate modeling, which can help predict future climate patterns and inform decision-making around mitigation and adaptation efforts.
• Renewable energy: AI is being used to improve the efficiency and output of renewable energy systems, such as wind turbines and solar panels, through real-time monitoring and optimization.
• Energy efficiency: AI is being used to improve energy efficiency in buildings and other infrastructure through automated systems that optimize energy use and reduce waste.
• Natural resource management: AI is being used to monitor and manage natural resources, such as forests and water sources, to help mitigate the impacts of climate change on these critical ecosystems.
• Carbon capture and storage: AI is being used to develop more effective carbon capture and storage technologies, which could play a critical role in reducing greenhouse gas emissions.
• Climate risk assessment: AI is being used to assess climate-related risks, such as flooding and droughts, and develop strategies to mitigate these risks.

The trend is toward using AI to develop innovative solutions that can help address the complex and interconnected challenges associated with climate change. While there are still many challenges to be addressed, AI is increasingly being recognized as a critical tool in the global effort to combat climate change.

Predictions of AI’s impact on combatting climate change

• Greater use of AI for climate modeling: AI will likely play an increasingly important role in climate modeling, which can help predict future climate patterns and inform decision-making around mitigation and adaptation efforts.
• Increased focus on renewable energy optimization: AI is expected to continue to be used to improve the efficiency and output of renewable energy systems, such as wind turbines and solar panels, through real-time monitoring and optimization.
• Advancements in carbon capture and storage technologies: AI is expected to help drive innovation in carbon capture and storage technologies, which could play a critical role in reducing greenhouse gas emissions.
• Greater emphasis on natural resource management: AI is expected to be increasingly used to monitor and manage natural resources, such as forests and water sources, to help mitigate the impacts of climate change on these critical ecosystems.
• Increased global collaboration: AI is expected to help facilitate global collaboration and information sharing, which is critical for developing effective strategies to address climate change.

While the exact impact of AI on combating climate change is uncertain, it’s clear that this technology is expected to play an increasingly important role in the global effort to address this critical issue. AI solutions need to be designed for user-friendliness, regardless of whether they are developed for corporations, governments, or the general public. They need to be easily accessible, offer tangible benefits to the user, and provide clear information to guide user action.

Despite its promise, AI cannot be used to solve the climate crisis in isolation. Rather, it is one of many tools that should be employed to address this global challenge. Individuals, communities, and organizations who have a part to play in combating the climate crisis — regardless of their formal role in AI or climate topics — should therefore consider how other emerging technologies can help and assist in removing obstacles to scale for those as well.

References

• Aligning artificial intelligence with climate change mitigation by Lynn H. Kaack et al. in Nature
• AI and Climate Change: How they’re connected, and what we can do about it, by Dobbe and Whittaker on Medium
• AI can help us fight climate change. But it has an energy problem, too, by Annette Ekin in Horizon
• BCG, How AI Can Be a Powerful Tool in the Fight Against Climate Change by Hamid Maher, Hubertus Meinecke, Damien Gromier, Mateo Garcia-Novelli, Ruth Fortmann
• BCG, AI Is Essential for Solving the Climate Crisis by Hamid Maher, Hubertus Meinecke, Damien Gromier, Mateo Garcia-Novelli, Ruth Fortmann
• BCG, Reduce Carbon and Costs with the Power of AI
• BCG, Responsible AI Builds Trust in Government
• BCG, Six Steps to Bridge the Responsible AI Gap
• Computational Sustainability: Computing for a better world and a sustainable future, by Gomes et al. in Communications of the ACM
• DeepMind/Google, Machine Learning Can Boost the Value of Wind Energy
• Global Partnership on AI Report, Climate Change and AI. Recommendations for Government Action
• How AI can help combat climate change, by Megan Mastrola in Science+Technology
• How AI can help the world prepare for climate adaptation by Tim van den Bergh
• How To Fight Climate Change Using AI by Mark Minevich in Forbes
• IPCC, Climate Change 2022: Mitigation of Climate Change
• Microsoft x BCG, Ten Guidelines for Product Leaders to Implement AI Responsibly
• OECD, OECD AI Principles, OECD AI Policy Observatory
• Tackling Climate Change with Machine Learning by David Rolnick et al. in ACM Computing Surveys, Vol.55, Issue 2
• UNESCO, Steering AI and Advanced ICTs for Knowledge Societies
• UNESCO, Using Drones to Reduce Impacts of Floods in The Gambia

Your feedback is highly appreciated! 🐼