Artificial intelligence (AI) is quickly becoming a cornerstone of society. Its rapid growth is startling, and it’s scary yet exciting. People are using AI for almost anything: research, art, text editing, the list goes on. The productivity that AI promises is its biggest appeal. While AI can be convenient and a masterful tool, it’s important to also acknowledge its negative environmental impacts. Everything has an environmental cost, but people tend to assume that the cyber world has no connection to the physical world because of how it is seemingly contained. This disconnect is simply untrue. AI consumes unimaginable amounts of electricity and emits so much carbon in said process. By exploring AI’s energy demands and potential solutions, we can better understand the urgent needs to address its environmental implications.
AI and The Amount of Energy it Uses
AI technology requires vast amounts of computational power. Training AI models, like ChatGPT, demands many high-performance servers operating continuously in data centers around the world. These data centers consume electricity not only to run the servers but also to cool them, ensuring optimal performance and preventing overheating.
To put it into perspective, researchers estimate that training a single large-scale AI model can emit as much carbon as five cars over their entire lifetimes, including manufacturing and fuel consumption. This environmental toll is amplified as companies race to develop more advanced models, increasing energy demands exponentially.
The environmental impact extends beyond the direct electricity consumption. Many data centers rely on non-renewable energy sources like coal and natural gas, further increasing carbon emissions. While there has been a push towards renewable energy integration in some regions, the pace of AI's growth often outstrips the progress in clean energy infrastructure.
Moreover, as AI becomes more ingrained in our daily lives, its energy usage becomes more of an uncontrollable entity. Whether generating text, creating art, or processing large datasets, each interaction with AI consumes energy. Multiply this by billions of users globally, and the environmental impact becomes immense.
The Hidden Physical Costs of AI
AI’s physical costs go beyond electricity consumption. The manufacturing of hardware required for AI, such as GPUs (graphics processing units) and TPUs (tensor processing units), also contributes to environmental degradation. Extracting the raw materials needed for these devices, including rare earth metals like lithium and cobalt, results in habitat destruction, water pollution, and human rights abuses in mining regions.
Furthermore, the disposal of outdated hardware poses a growing challenge. E-waste, or electronic waste, is one of the fastest-growing waste streams globally. Many of the materials used in electronic devices are non-biodegradable, and improper disposal can release toxic chemicals into the environment.
The convergence of these factors reveals that AI is far from being an intangible technology. Its environmental impact is as real as the fossil fuels burned to keep it running, and its consequences are felt in ecosystems around the world.
Steering Away from AI Overuse
The problem with AI and its emerging presence is how it is quickly becoming a part of society. All good things should be used in moderation, and AI is no exception to this. Our usage of generative artificial intelligence must be responsible. Although there is no seemingly physical cost of using said technology, environmental degradation is a real consequence of overuse.
Outside of environmental costs, AI abuse can result in declining attention spans and codependency. In 2004, the average attention span on a screen was approximately 2 ½ minutes. In 2023, the average declined to a startling 47 seconds. This decrease isn’t only the fault of AI; it’s short form content as a whole, such as the rise of apps like Tik Tok and Instagram. Our brains have been conditioned to crave quick and easy to consume media. Another big difference between the past couple of years and the present is AI generated content, which exponentially increases media production. There’s so much content to consume, and when time feels so scarce, our brains don’t want to spend time on just one piece of media. It all becomes a numbers game: more consumption, less depth.
Consumers, too, have a role to play. Instead of relying on AI for tasks that can be done manually or through simpler means, users should evaluate whether the convenience of AI justifies its environmental cost. For example, automating basic emails or generating simple artwork might save time but at a considerable energy expense. Critical thinking skills are integral to existing as a human, and this is something that must be worked on every day. Instead of using AI for its convenience, consider relying on and strengthening your own human intelligence. Artificial intelligence can be a powerful tool, but it is not an alternative to human intelligence. If AI replaces cognitive labor, then we risk losing a vital part of being human.
AI Alternatives
To mitigate the negative impacts of AI, environmentally and socially, there are alternatives that can be explored. Here are some things to consider instead of AI:
Strengthen Human Intelligence: People can rely on their own research skills and knowledge instead of going straight to AI when brainstorming.
Digital Minimalism: Reflecting on digital habits is something everyone should consider. Using technology sparingly will reduce overall energy consumption and helps create a more sustainable lifestyle.
Support Ethical Tech: Advocating for more ethical tech development can help with AI’s negative impacts. An example of this would be advocating for transparency in energy usage and carbon emissions when using AI.
Support Renewable Energy: Data centers utilizing renewable energy sources would significantly reduce AI’s carbon footprint.
Moving Forward
With AI and everything else in life, balance is key. AI overuse has serious ramifications that should be acknowledged. As individuals, businesses, and governments, we must acknowledge AI's environmental and social costs while working collectively to mitigate them. A sustainable and ethical future for AI demands a balance between embracing its potential and minimizing its impact on the planet and society. Ultimately, the path forward requires thoughtful, intentional action at every level. By fostering collaboration, we can harness AI’s potential without compromising the health of our planet or the vibrancy of our humanity. A future where technology and sustainability coexist is not only possible but imperative.
Comments