Artificial intelligence (AI) has gained popularity in recent years in numerous arenas: workplace, healthcare, transportation, creativity, communication, social media, entertainment, and more. Some may claim it is even taking over the world! Today AI has helped enhance task automation, user experience, driving, digital assistance and more. So, what exactly is AI? Put simply, the term is used to describe technology that strives to reason, learn and act in ways that are as close to "human" as possible. AI software absorbs large amounts of data in order to identify patterns, predict possible outcomes, and make decisions.
AI shows up in our user experience in the form of personalized recommendations, spam detection, 24/7 chat bots and IT support, to name a few. The self-driving cars commonly seen throughout San Francisco, Waymo, are an example of how AI has integrated into transportation and driving. Apple’s big launch of the new iPhone 16 includes many AI features, such as Genmoji, image playground and an improved Siri. Other digital assistance devices, like Amazon Alexa and Google, utilize AI to improve responses and detect user patterns. Within the umbrella of AI are different types of AI; traditional and generative are the two most commonly known. Traditional AI analyzes data in order to make predictions, whereas generative AI, such as ChatGPT or Google Gemini, focuses on creating entirely new content.
While it can be beneficial in some ways, is AI sustainable? How is it impacting the environment? These are relevant questions when a new technology emerges, especially one that has been widely adopted as quickly as AI has.
The amount of energy it takes to use an AI tool is more complex than simply plugging in your device, typing a prompt, and receiving the solution. The data centers, including cloud service providers, required to maintain, train and run the AI learning models use enormous amounts of energy. Scientists have estimated that the power requirement for AI at the end of 2023 in North America was over 5,300 megawatts (MIT News). To put things in perspective, the human brain uses approximately 15-20 Watts of power. Generative AI has shorter shelf-life due to its high demand in various applications, resulting in newer versions being released every few weeks. A single “ChatGPT query consumes about five times more electricity than a simple web search” (MIT News). Beyond that, the water consumption required to cool data centers is about two liters of water for every kilowatt hour of energy. Aside from AI’s energy use, there is also the physical side. All data centers produce electronic waste through the disposal of equipment or hardware, which usually contains hazardous material – mercury and lead, further contributing to pollution (UNEP.org). The hardware can require natural resources such as cobalt, silicon, gold and other metals (Environment.yale). AI vastly increases the quantity of this discarded material.
In the long haul, generative AI’s impact is extremely significant. Its increasing electricity demand, water consumption, use of minerals and other materials in manufacturing, and e-waste is contributing to climate change. While AI definitely has its benefits, AI was created to serve as a tool that supports humans. It should not be intended to take over, replace humans, or "do our jobs" per se. As AI continues to advance, it has the potential to revolutionize sustainability efforts by optimizing supply chains, identifying wasteful processes and improving efficiency. To ensure we contribute to creating a more sustainable and equitable future we need to push the AI industry to integrate these values into every aspect of their AI products.
Sources and further reading:
https://news.mit.edu/2025/explained-generative-ai-environmental-impact-0117
https://environment.yale.edu/news/article/can-we-mitigate-ais-environmental-impacts
https://hbr.org/2023/07/how-generative-ai-can-augment-human-creativity
Photo by Júnior Ferreira on Unsplash
Replies