Intelligent Climate: Intersection of AI and Climate

Introduction

The history of artificial intelligence is longer than the history of most modern democracies. Yet it was introduced to the public much later in the twentieth century. In 1950, the term Artificial Intelligence (AI) was discussed by an English mathematician and computer science pioneer, Alan Turing, in his paper on computer machinery and intelligence (Mucci, n.d).
Yet, now more than ever, the discourse on AI, its ethics, and concerns has begun to shape public narrative, with generative AI taking the center stage. The big fear for many has been the potential for AI to reduce costs and thus take over jobs. A study by the U.S. Bureau of Labor Statistics on the impacts of AI on employment projections between the period of 2023-33 shows that there is a predicted percent change of 11.7% for computer occupations, 3.7% for legal occupations, 6.9% for business and financial operations occupations, and 6.8% for those in architecture/engineering occupations (BLS, 2025). These impacts are concerning, especially in global economic insecurity. Even so, a relationship that is not widely discussed is one between climate change and AI. Climate change is one of the most compelling and urgent collective challenges in recent times.
On the one hand, AI simulations and machine learning are being used for weather and climate modeling. Scholars are exploring the role that AI can have in addressing climate change by expanding scientific knowledge, improving modeling accuracy, and refining climate scenario projections (Lewis et al, 2024). On the other hand, a less discussed aspect of this relationship is the impact of AI on the environment. AI data centers significantly contribute to greenhouse gas emissions. According to some estimates, powering these data centers already accounts for 2% global energy demand, nearly the same figure as consumed by the airline industry (Marabelli & Davison, 2025). The intersection between the two offers both unprecedented opportunities and some complex challenges.
The uptake of AI in India has also been rapid, with 6 million people employed in the tech and AI ecosystem (PIB Delhi, 2025).  Indians are rapidly embracing open access to generative AI resources, with nearly 64 percent of Indians having used AI, according to a study, which is twice the global average (TOI, 2025).
This provides a good possibility for the expansion of AI in India, and the government policies are shaping up to do the same as well.  This paper aims to address both sides of this coin to present opportunities to position policy & practice to pave the way for a careful balance between using AI to drive progress in climate modelling and insisting on safeguards to ensure its development does not hamper the progress towards net zero emissions targets.

Tool for Climate Modelling

Weather forecasting and climate modelling have become big industries.  Tech giants such as Google and IBM are leading efforts to use AI models for more precise and expedited forecasting (Jamwal, 2024). The conventional climate models that are built manually by climate scientists use mathematical equations to describe the process of interaction between land, oceans, and air have been guiding global climate policy through the years (Wong, 2024). However, these models rely on powerful supercomputers that are energy-intensive, consuming up to 10 megawatt hours of energy to simulate a century of climate (Wong, 2024). Climate scientists are using machine learning (ML) and deep learning (DL) to overcome some of the challenges.
The evolution of DL has enabled scientists to utilize AI methodologies for modeling, detection, prediction, and impact assessment of extreme climate events (Camps-Valls, 2025). These models depend on the spatio-temporal Earth observation, reanalysis, and climate data. However, even though these ML algorithms can be used for deterministic extreme event prediction, they are currently limited in their regional scope.
India is eager to harness AI and ML for better weather and climate forecasting in the midst of growing challenges of extreme weather events. In India, IIT professionals and geoinformatics scholars have also been working on AI-based models, but they are faced with the challenge of compiling data (Jamwal, 2024). As the climate researchers highlight, in remote and complex regions such as the Himalayan cryosphere, in situ measurements are scarce, glacier and glacial-lake datasets are sparse, and socio-economic or health-related climate-sensitive data are hard to access or archived at low resolution (Jamwal, 2024). Presently, despite extreme events such 2023 GLOF in Sikkim, there is no seismic data collected around glaciers in the entire Indian Himalayan region and no detailed GLOF risk analysis either (Jamwal, 2024).
Since AI/ML-driven models rely heavily on large volumes of high-quality data, the lack thereof risks creating unreliable models. Nonetheless, the government-led initiative for AI can create opportunities to strengthen data-collection networks. The potential for AI climate modelling for extreme climate events in India can be a tool for disaster management on all levels of planning and precautions for states, especially those with higher vulnerability.

Training AI: GHG and Energy Use

When addressing the connection between AI and climate impact, Tachovsky states it well: ‘AI can help or harm the Planet. It’s Up to Us’ (WRI, 2025). The training, deployment, and use of AI are financially and environmentally expensive. The process requires high-performance computing (HPC) infrastructure with thousands of GPUs equipped with tensor processing units. This specialized chip enhances the speed of machine learning tasks, along with CPUs that run in parallel for weeks and months, consuming massive amounts of electricity (Kandemir, 2025).
According to the International Energy Agency, a typical AI-focused data center can consume as much electricity as 100,000 households, and larger centers that are under construction will consume 20 times that amount. Beyond energy utilization, data centers also use excessive water for advanced cooling systems. OpenAI researchers have found that since 2012, the amount of computing power required to train cutting-edge AI models has doubled every 3.4 months (Kanungo, 2023). By 2040, it is expected that the emissions from the Information and Communications Technology (ICT) industry as a whole will reach 14% of the global emissions, with the majority of those from ICT infrastructure, including data centers and communications networks (Kanungo, 2023).
The environmental cost of AI runs deeper. With increasing use of electricity & thus the dependence on fossil fuel, the AI expansion risks localized air pollution and thermal pollution in water bodies, as well as production of solid wastes, including hazardous materials (Ren & Wierman, 2024).

Keeping in mind India’s pressing environmental challenges, AI data centers can exacerbate the issues of energy use and water scarcity. The draft National Data Centre Policy of 2025 has the potential to open the market in India for the development of data centers for global tech leaders. U.S.-based companies, including OpenAI, Google, and Microsoft, have announced plans to set up data centers in India (Tomas, 2025; Google Cloud, 2025). The policy proposed by the Ministry of Electronics and Information Technology provides incentives for tax exemptions for up to 20 years for operators who meet specific capacity and efficiency targets, and a simplified process for obtaining the multiple licenses and approvals needed to set up data centers. These policies may fast-track the setting up of data centers, but operational maintainability must be taken into account. For Indian data center designers, reliable electricity is a major issue due to erratic grid quality and inconsistent load sanctions (Kongaleti, 2025).

Although this has the potential to unlock AI sector tech jobs, this also reflects the overarching policy that tech giants in the West have used time and again to create low-cost, less environmentally efficient technologies and push them to countries with weaker protections. This is also showcased by the operations of data centers in the developing world as compared to the developed world. For instance, Google operated its data centers with 97% carbon-free energy in Finland and only 4-18% in Asia (Ren & Wierman, 2024). Without sustainability checks, this could jeopardize India’s current trajectory to reduce its emission intensity and achieve climate goals.

Conclusion

The high and increasing demand for generative AI presents significant challenges for sustainable growth in infrastructure for expansion, yet at the same time, the responsible use of AI modelling can help prepare for extreme climate events in the future. Thus, like most technological advancements, AI creates room for growth and concerns for overuse. As India opens its data markets and attracts global investment for digital infrastructure.
Although not linked entirely, these aspects of AI from machine learning, climate modelling, and generative AI are different frontiers of the same element. Energy use is already a significant challenge for India with its overloaded power grid and large populous. Data centers are often also built in more remote and rural areas, which are severely undersourced for electricity in the country. The current approaches to managing AI and the deployment of resources for data centers create a significant risk of compounding socioeconomic disparities in regions and exacerbating the issues of environmental inequity (Ren & Wierman, 2024).
To sufficiently support energy-intensive data centers, future research & design must integrate renewable energy solutions and dual source power with off-grid solutions (Kongaleti, 2025).
The duality that this intersection offers the opportunity for India to position itself as a data center hub, and the government must also position itself for the same. Yet, at the same time, considering early action cannot only reduce potential harm but also be another opportunity to leverage global knowledge to strengthen data collection where gaps exist. The upcoming policy, therefore, must consider the opportunities along with the challenges for the Indian context and create protections. The path forward must consider a careful balance between harnessing AI’s transformative capacity while institutionalizing safeguards to ensure its development and deployment contribute to meaningful and sustainable climate solutions.

Turing, A. (n.d.). The history of AI. IBM. https://www.ibm.com/think/topics/history-of-artificial-intelligence

BLS (2025, February). Incorporating AI impacts in BLS employment projects: occupational case studies. U.S. Bureau of Labor Statistics. https://www.bls.gov/opub/mlr/2025/article/incorporating-ai-impacts-in-bls-employment-projections.htm

Lewis, J. et al (2024, September 28). Climate change and artificial intelligence: assessing the global research landscape. Discover Artificial Intelligence. https://doi.org/10.1007/s44163-024-00170-z


Marabelli, M. & Davison, R. (2025, September). Artificial Intelligence and the environment: ethical challenges and strategic opportunities for organizations. Journal of Strategic Information Systems. https://doi.org/10.1016/j.jsis.2025.101921

PIB Delhi (2025, October 12). Transforming India with AI. https://www.pib.gov.in/PressReleasePage.aspx?PRID=2178092

TOI (2025, February 11). More than 60% Indians use AI, more than twice the global average: Microsoft. https://timesofindia.indiatimes.com/business/india-business/more-than-60-indians-use-ai-more-than-twice-the-global-average-microsoft/articleshow/118153012.cms

Jamwal, N. (2024, April 26). Data drought: the challenge of AI weather forecasting in India. Dialogue Earth. https://www.preventionweb.net/news/data-drought-challenge-ai-weather-forecasting-india

Wong, C. (2024, March 26). How AI is improving climate forecasts. Nature. https://www.nature.com/articles/d41586-024-00780-8

Camps-Valls, G. et al. (2025). Artificial intelligence for modeling and understanding extreme weather and climate events. Nat Commun 16, 1919 https://doi.org/10.1038/s41467-025-56573-8

Tachovsky, E.  (2025.) AI Can Help or Harm the Planet. It’s Up to Us. World Resources Institute. https://www.wri.org/insights/ai-promise-and-peril-for-planet

Kandemir, M. (2025, September 08). Why AI use so much energy-and what we can do about it. Penn State Institute of Energy and the Environment. https://iee.psu.edu/news/blog/why-ai-uses-so-much-energy-and-what-we-can-do-about-it

Google Cloud News (2025, October 14). Google Announces First AI Hub in India, bringing the company’s full AI Stack and Consumer Services to the Country. https://www.googlecloudpresscorner.com/2025-10-14-Google-Announces-First-AI-Hub-in-India,-Bringing-Companys-Full-AI-Stack-and-Consumer-Services-to-Country#:~:text=NEW%20DELHI%2C%20Oct.%2014%2C,network%2C%20all%20in%20one%20place.

Kanungo, A. (2023, July 18). The Green Dilemma: Can AI fulfill its potential without harming the environment? Earth.Org. https://earth.org/the-green-dilemma-can-ai-fulfil-its-potential-without-harming-the-environment/


Kongaleti, M.K. (2025, September 21). The unspoken trust about designing data centers in India. DCD. https://www.datacenterdynamics.com/en/opinions/the-unspoken-truth-about-designing-data-centers-in-india/#:~:text=With%20AI%20workloads%20now%20demanding,a%20political%20and%20regulatory%20maze.

Ran, S. & Wierman, A. (2024, July 15). The Uneven distribution of AI’s Environmental Impacts. HBR. https://hbr.org/2024/07/the-uneven-distribution-of-ais-environmental-impacts

Default Author Image

Anusha Arif

Found this post insightful? Share it with your network and help spread the knowledge.

Suggested Reads

COP 30: Triple Adaptation Finance and Just Transitions

The Conference of the Parties under the United Nations Climate Change Conference met for the annual meeting in Belem between November 10 to November 22, 2025. In a time of climate uncertainty and a rapid call for urgency to recognize the urgency of climate change, the COP30 brings an opportunity for global collaboration and a […]

The Need to Digitise Diplomacy: Politics of Connectivity

India, today, finds herself in a sundry of diplomatic paradoxes, marked by multi-alignment and diverging stands on issues such as Israel-Palestine. The inconsistencies in our foreign policy have been cleverly brushed off under the term ‘strategic autonomy’—perpetually undefined and unboxed. Of course, as the world’s most promising emerging market with a labour surplus and the […]

From Make in India to Create in India: Charting India’s ‘Orange’ Economy Frontier

During India’s first World Audio Visual and Entertainment Summit (WAVES) 2025 in Mumbai, Prime Minister Narendra Modi announced the dawn of the ‘Orange Economy.’ He highlighted how creators—from Guwahati musicians and Kochi podcasters to Bengaluru game designers and Punjabi filmmakers—are powering India’s march towards a $5 trillion economy. The orange economy, also known as the […]

Small Island Nations and Great Power Politics

‘Power in world politics’ is a phrase that directs our imagination towards the big powers of the world: huge armies, vast populations, and rich economies. The numerous small island nations rarely enter the picture. But when you look closely, these islands are not just remote holiday spots. These small dotted lands on the map of […]

Development Finance Is Reshaping Africa’s Economic Reality– But not for the better

Fueling growth in an economy primarily requires investments, whether for technology, infrastructure, or government interventions. The global financial system has evolved significantly: from the use of cash and traditional banking tools such as cheques and demand drafts, to more advanced systems like digital payments (e.g., UPI), and more recently, toward decentralized technologies like cryptocurrencies. While […]