AI-Driven Climate Change Solutions and Their Future

Japanese technology plays pivotal role post-COP29

10 March 2025

Japanese version: 30 January 2025

Motoki Tajima
Business Data Analytics Division

The world saw its hottest average temperature1, and a barrage of extreme weather, in 2024. How we tackle climate change will determine the future of the planet. Digital technologies were the focus of attention for the first time at the United Nations Climate Change Conference's 2024 session (COP29) in Baku. Among them, AI-driven climate change solutions represent a burgeoning field spanning both the physical and digital realms. Japanese technologies have much potential to play a pivotal role in bridging AI solutions and COP29 discourse—early warning systems and satellite imagery for climate monitoring in particular.

AI adoption accelerates electricity consumption

Analysis of AI solutions first requires consideration of their negative impacts. In recent years, the massive electricity consumption of data centers, which serve as the backbone of AI, has become a growing concern. Running AI models requires intensive computational resources, and as generative AI becomes more widespread, processing demands have surged, significantly increasing power consumption.

The International Energy Agency estimates that electricity consumption in data centers could increase 2.3-fold by 2026 when compared to 2022 levels.2 Without intervention, this will lead to an increase in fossil fuels to match the increase in power generation, resulting in greater emissions of greenhouse gases (GHG).

The issue of electricity consumption by data centers was a key topic at COP29, reflecting AI’s growing global presence. While no definitive solutions were reached, the discussions highlighted the need to focus on how the sources of GHG emissions will change over time. Mitsubishi Research Institute, Inc. advocates for the use of energy-efficient AI and the development of semiconductor technology to both continue using AI and reduce electricity consumption.*

AI-driven climate change measures also show promise

Climate change measures fall in two broad categories: mitigation measures and adaptation measures. Mitigation focuses on reducing and absorbing emissions, such as enhancing energy efficiency and expanding renewable energy. Adaptation, on the other hand, strengthens societal and economic resilience against unavoidable climate impacts. Adaptation measures include urban infrastructure improvements and agricultural products fit to future climate change. While mitigation features heavily in the bulk of climate change measures, the swell in global temperature and extreme weather necessitate a comprehensive approach including adaptation.

AI’s capacity for navigating swathes of data positions it well for developing data-driven solutions, including adaption. AI solutions exist in both the digital and physical realms. In the digital domain, AI output creates value directly as information. In the physical domain, AI contributes indirectly: through hardware such as robots and through the development and operation of products.

For digital solutions, AI fills roles spanning recognition, analysis, prediction, and optimization.
  • Recognition: Extract features and patterns to understand a given situation and identify semantics
  • Analysis: Organize enormous amounts of complex data and identify patterns
  • Prediction: Forecast future trends and outcomes based on historical data
  • Optimization: Discern the most appropriate solution for given circumstances
AI-driven solutions cover a wide range of climate change measures (Figure 1). Both climate change measures and AI-related technology are rapidly evolving, and each country should consider where their opportunities lie on the canvas of solutions—the first step in considering climate change solutions.
Figure 1: Overview of AI-driven climate change solutions
Overview of AI-driven climate change solutions
Source: Mitsubishi Research Institute, Inc.

Japanese technology bridges COP29 discourse and future AI solutions

COP29 featured active discussion of adaptation measures. COP30 aims to complete a list of specific indicators for assessing progress towards adaption-measure goals. While frameworks for adaption measures are still in their early stages, digital technology appears to be a frontrunner for innovation in combating climate change.

COP29 did however highlight two key adaptation solutions: early warning systems and satellite imagery. Both are areas of strength among Japanese technologies.

Early warning systems

Early warning systems provide alert of potential natural disasters, such as sudden heavy rainfall and floods, making timely evacuations possible. At COP29, the Early Warnings for All initiative emphasized the importance of making early warning systems accessible to all, including those in the world’s poorest regions.

Mobile communication networks cover 96% of the global population, making them the most viable solution. Yet, three billion people still lack mobile access; this is due to economic constraints rather than technical limitations. Developing alternative communication channels for this population is crucial.3

Many developing countries also face a lack of data on past disasters. Early warning systems under such circumstances will necessitate use of existing infrastructure and innovative methods to make accurate predictions amid modest data.

Due to its geographical characteristics, Japan is prone to a variety of natural disasters, including typhoons, floods, heatwaves, and droughts. This unique situation has led Japan to proactively collect data and develop technologies over many years to effectively address these challenges. With this extensive experience and a wealth of solutions, Japan is well-positioned to play a central role in building innovative early warning systems by sharing its expertise globally.

Satellite imagery for climate monitoring

Comprehensive climate assessment requires extensive data collection. Satellite and aerial imagery are playing a larger part in the monitoring of environmental conditions. Image recognition is closely associated with deep learning, which has fueled the recent AI boom, and is expected to make further advances.

Early warning systems can predict the timing and magnitude of sudden, heavy rainfall, and adding satellite imagery to the mix opens the door to areas other than weather. For example, by analyzing land cover and atmospheric conditions, solutions can identify areas where virus-carrying mosquitos are likely to appear and forecast the outbreak of infectious diseases. Satellite imagery can be used for both adaptation and mitigation measures, such as monitoring deforestation and analyzing the atmospheric concentration of CO2.

A leader for years in industrial robotics and the automation of manufacturing, Japan has forged advanced sensing technologies. Further innovation in climate solutions is possible by combining these cutting-edge sensors with AI capabilities, accumulated data, and satellite images. There is much potential in bringing Japanese technology to a wider global audience and putting it to work in novel ways.

What Japan needs to do

Japan has been working for years to strengthen its resilience by promoting the development of societal infrastructure for disaster prevention and mitigation. The region is characterized by frequent natural disasters, lending to Japan’s strengths in physical measures. AI and IoT can enhance Japan’s digital measures, enabling it to develop unique and cost-effective climate solutions featuring high reliability and long-term support. Through this work, Japan has a chance to play a leading role in the field of adaptive solutions.

For this to happen, the private sector will need to develop solutions and business models that meet global needs and technological trends. Meanwhile, the government must facilitate the wider use of digital technology through policy support and IT infrastructure investment. Other key factors will be whether the country can build an open innovation platform that promotes collaboration across data sets and research projects, and whether it can leverage its advanced technological capabilities, accumulated experience, and data.

Through such initiatives, digital technologies can be positioned at the heart of sustainable climate change measures. Japan can lead the effort by collaborating with countries worldwide and setting a global precedent for AI-driven climate change solutions.
Photo at COP29 by author.
Photo at COP29 by author.

Works Cited

1Copernicus Climate Change Service (2024) “Copernicus: 2024 virtually certain to be the warmest year and first year above 1.5°C”
https://climate.copernicus.eu/copernicus-2024-virtually-certain-be-warmest-year-and-first-year-above-15degc (accessed 6 December 2024)

2International Energy Association (2024) “Executive summary”
https://www.iea.org/reports/electricity-2024/executive-summary (accessed 16 December 2024)

3Shanahan et al. (2024) The State of Mobile Internet Connectivity 2024 GSMA

Author profile

Author

Motoki Tajima

Business Data Analytics Division

Motoki Tajima helps clients redesign their businesses based data utilization. His consulting work is informed by several years spent working on the business-operator side. He strives to propose solutions standing in the shoes of his clients.