AI: Climate Problem or Solution?
The author is a rising Sophomore at Laurel School in Ohio.
From essays crafted by Chat GPT to Google Gemini answers appearing after a search, it is almost impossible to avoid the increasing presence of artificial intelligence (AI). An article published by McKinsey in March of 2025 even reported that “78 percent of respondents [in their survey] say their organizations use AI in at least one business function, up from 72 percent in early 2024 and 55 percent a year earlier.” As both AI and climate change continue to grow rapidly, questions arise about the relationship between the two. Unfortunately, the link isn’t as simple as one might think; it’s complex and relies greatly on how we use AI.
First, AI generates an immense amount of greenhouse gas emissions in various ways, from the training of models to the use of chatbots, which exacerbates the climate crisis. The training process of AI is one of the main contributors to this issue. An MIT News article explains that “[i]n a 2021 research paper, scientists from Google and the University of California at Berkeley estimated the training process alone consumed 1,287 megawatt hours of electricity (enough to power about 120 average U.S. homes for a year), generating about 552 tons of carbon dioxide.” Much of this energy comes from fossil fuels. The US Energy Information Administration reported that, in 2023, around 60% of electricity from “utility-scale electricity generation facilities in the United States” came from non-renewable sources. Adding to this are the electricity demands of AI data centers.
In fact, a paper from the MIT Sloan School of Management corroborates that “[a]lready, data centers account for 1% to 2% of overall global energy demand, similar to what experts estimate for the airline industry.” This number is projected to grow exponentially due to bigger, more advanced, AI models and the increased usage of AI. Furthermore, questions or comments typed into an AI chatbot or other model use energy and produce emissions. A report from The Guardian specifies that “generating AI queries could require as much as 10 times the computing power as a regular online search.”
AI doesn’t just hurt the world around us through its vast energy consumption and emissions; it also “demands land and water, often at the expense of local communities,” according to a review by the Brookings Institution. This is due to data centers’ dependence on water. An article from the University of Illinois Urbana-Champaign's Center for Secure Water reported that AI’s water usage can be funneled into three main categories: water needed for cooling purposes throughout the process of creating AI models, water utilized in the generation of electricity for powering data centers, and water consumed in relation to AI supply chains, such as in the manufacturing of microchips. Data centers consume large amounts of water in the face of growing water scarcity issues, harming vulnerable communities even further.
After learning all of this, it may seem impossible that AI could aid us in mitigating climate issues. However, there are quite a few ways society can use artificial intelligence to fight back against environmental crises.
First, many of the problems relating to emissions from AI can be solved. Instead of relying on fossil fuels to power data centers, renewable energy, such as solar or wind power, could be used. Google, for example, “partnered with a clean-energy startup to develop a geothermal power project in Nevada that is now contributing carbon-free energy to the electric grid.” This would drastically reduce the emissions impact of AI since the AI would no longer rely on fuels that pollute the Earth. Moreover, more effective, efficient, and sustainable policies for data centers’ dependence on water and land could prevent the devastating impacts on at-risk populations.
In fact, AI may even have the potential to greatly reduce climate issues if it is used to “accelerate low-carbon technologies” as described in a World Economic Forum piece. An article from the Columbia Climate School continues that “AI is increasingly used to manage the intermittency of renewable energy so that more can be incorporated into the grid; it can handle power fluctuations and improve energy storage as well.” This means renewable energy cannot only be used in the process of AI production, but AI can also extend the use of renewables to entire grids. For example, Norway has integrated AI into its electricity grids, helping to utilize more environmentally-friendly energy and promoting sustainability. Indeed, another report from the World Economic Forum found that “AI has the potential to reduce global greenhouse gas (GHG) emissions by 5-10% – an amount equivalent to the annual emissions of the entire European Union.” AI has the potential to change the global landscape, helping to preserve the Earth for years to come.
Looking at all of this data and information, is the net impact of AI on the environment positive or negative? Well, it’s up to companies, entrepreneurs, and AI users. The choice is whether to work towards making AI training and data centers more sustainable or to allow AI to continue to emit large amounts of greenhouse gases. It remains uncertain whether AI will be used to power a transition to renewable energy and drastically reduce climate impacts, or if the environmental issues related to climate change will be dismissed and remain largely unaddressed. Artificial intelligence could help save the world, but it comes down to usage and awareness. It is imperative to think critically and carefully about AI applications in order to ensure the Earth is protected instead of destroyed.
