This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
AI’s Mixed Impact on Sustainability
Posted on February 26, 2026 by Phoebe Ryu
#Corporate Sustainability Reporting #ESG Issues #Uncategorized
Rapid Adoption of AI
Artificial Intelligence (AI) is rapidly increasing in use by both individuals and businesses, with generative AI enabling the creation of original content in response to prompts and agentic AI creating systems that can make decisions and act autonomously. McKinsey recently conducted a survey of 1,993 worldwide participants in which almost 90% said their organizations are using AI on a regular basis. Additionally, an International Data Corporation survey found that 76% of IT decision makers worldwide find AI “critical” for sustainable development. While many companies are using AI to create tools that improve sustainability, the harmful environmental impact of greater AI use is creating a new set of sustainability challenges.
Environmental Impact of AI
AI has large negative environmental consequences due to the enormous demand for energy and water used in the data centers that house the AI servers. In 2024, the United States used around 200 terawatt hours of electricity, with an estimated 53 to 76 of those terawatt hours coming from AI-specific data centers. The estimate of 76 terawatt hours is enough to power more than seven million U.S. homes for a year. It has been found that large data centers need as much as five million gallons of water a day to properly cool their equipment. Bloomberg reports that about 65% of data centers developed since 2022 are in areas of high water stress.
These resource demands will only continue to increase in intensity as AI use continues to grow among individuals and companies.
Greyparrot’s AI Waste Sorter
Companies are using AI to develop sustainable solutions that can eliminate tedious tasks and improve worker safety. Greyparrot, based in the United Kingdom, developed a system using AI and smart cameras to enhance waste sorting at waste and recycling facilities. This job, which is typically performed manually, can be time consuming, inefficient and dangerous. Instead, Greyparrot offers a live camera feed that operators can view while the AI system identifies the waste materials of objects entering the facility. Waste facilities using Greyparrot’s system have seen a 10% increase in PET recovery.
Google’s Clashing Impact on Emissions
Google is using AI to develop tools to address climate issues. Beginning in 2021, Google incorporated an AI model into Google Maps to create routes that consume the least amount of fuel by considering vehicle type, terrain and fuel type, all while maintaining a similar estimated time of arrival. This feature was estimated to have reduced greenhouse gas (GHG) emissions by
2.9 million metric tons in 2023. Google also created Green Light, an AI tool that optimizes traffic light cycles to reduce waiting time, which has the potential to reduce GHG emissions by 10% at intersections.
At the same time, Google’s increased use of AI is expected to lead to higher levels of GHG emissions, which were 14.3 million metric tons in 2023. Google introduced their “AI Overview” feature in 2024, showing a Google Gemini-generated response to the standard Google search prompt, which uses 0.24 watt-hours of energy with each text prompt. This AI response is not optional and with about 16.4 billion Google searches being made per day, the supplemental AI response consumes large amounts of energy. While Google is making an effort to create AI tools that improve sustainability, the increased use of AI in search is more than offsetting the emissions saved.
Dell Focuses on Sustainable AI Infrastructure
Dell Technologies has focused on making data centers sustainable through energy efficiency, liquid cooling optimization, and personalized assessments for interested customers. The importance of efficient infrastructure, storage management, and material recycling is emphasized in Dell’s sustainability initiatives. They offer Pro Consult Advisory Services and the Get Efficient Assessment to individually analyze data centers and identify areas of improvement. Results from these programs showed a 72% decrease in power and cooling while having a 77% increase in performance.
Need for AI Regulations
As the rapid adoption of AI leads to higher levels of energy and water usage, governments are beginning to put regulations in place to reduce the environmental impact. The European Parliament introduced their first AI law in 2023 which touched on the environment, with member Brando Benifei saying, “we cannot just rely on self regulation.” In 2025, the Coalition for Environmentally Sustainable Artificial Intelligence was announced at the AI Action Summit in Paris. This group brings together over 90 partners, ranging from governments to businesses that want to make AI more sustainable with specific strategies that are accessible to all.
In the U.S., the Artificial Intelligence Environmental Impacts bill was introduced in Congress in 2024 to address the environmental concerns that come with the growing use of AI. However, no progress has been made on the bill as its status remains as “introduced”. The U.S. government’s AI Action Plan emphasizes that the U.S. wants to dominate the AI industry on a global level and discourages regulation, making it unlikely that environmental policies regarding AI will be passed on a federal level. The responsibility for monitoring the environmental impact of AI and taking action will increasingly fall upon companies themselves.
Future Outlook
AI can be used to create sustainable solutions to environmental challenges and AI infrastructure such as data centers can be made more sustainable through energy optimization, storage compaction and cooling efficiency. With AI being a tool used by individuals and corporations on a daily basis, ensuring it is regulated or monitored is crucial to limiting its impact on the environment. The power of AI can be harnessed to improve sustainability, but only if the consequences are acknowledged and addressed.
G&A incorporates AI responsibly into its research and advisory work by prioritizing transparency, data integrity, and human oversight. Explore our approach to applying accountable AI in our work here.
ABOUT PHOEBE RYU
Analyst-Intern, G&A Institute
Phoebe Ryu is currently a Masters student in Rutgers University’s 4+1 Environmental Sciences program, having just graduated in May with a Bachelor’s of Science in the program with a minor in Sustainability.
Prior to serving as a G&A intern she had performed as a research assistant in various labs but found her passion now lies in corporate sustainability efforts after serving as an Environmental Health and Safety intern within an office setting. Her interests are sustainability efforts related to agriculture along with environmental justice.