Despite energy concerns, AI offers substantial potential to boost environmental sustainability—we need to take advantage of it.
Artificial intelligence (AI) is often entangled in environmental debates, painted as an energy monster and climate hazard. While these concerns must be addressed, they fail to capture the transformative potential AI holds for environmental sustainability. This potential, backed by data and ongoing developments, far outweighs the initial challenges.
Yes, training complex AI models requires significant energy, which raises concerns about carbon emissions. However, focusing solely on this overlooks the strides being made. Tech giants, who are heavily involved in AI, are investing in renewable energy sources for their data centers, aiming for carbon-neutral operations. Google, for example, has committed to matching its entire electricity consumption with carbon-free energy by 2030. Moreover, researchers are actively developing more energy-efficient algorithms and hardware. Techniques like “model pruning” and “quantization” are already showing promise, aiming to reduce the computational power needed for AI tasks. These advancements are crucial to separating AI’s progress from its carbon footprint.
The growing e-waste generated by AI-powered devices is a valid concern. Yet, dismissing AI based on this fact overlooks the solutions on the horizon. The EU’s Circular Economy Action Plan, for instance, proposes stricter regulations on manufacturers, making them responsible for the collection and recycling of their products. This incentivizes designing devices for easier disassembly and reusing valuable materials, like the rare earth elements needed for AI components. Additionally, advancements in recycling technologies like hydrometallurgy offer promising solutions for extracting valuable materials from e-waste, minimizing its environmental impact. Companies like RecycLiCo are developing processes that can recover up to 99 percent of critical minerals from e-waste, creating a closed-loop system for these resources.
Beyond mitigating its footprint, AI’s true strength lies in its ability to drive positive environmental change.
AI-powered systems can predict weather patterns with unprecedented accuracy, enabling communities to prepare for climate events and optimize energy grids. For example, DeepMind’s weather forecasting system has demonstrated skill exceeding traditional methods, potentially saving billions of dollars in damage from extreme weather events.
It can also guide farms in using less water and pesticides while maximizing yields, significantly reducing their environmental impact. Studies show AI-powered tools have decreased water usage by up to 25 percent and pesticide use by 9 percent. This translates to billions of cubic meters of water saved and millions of tons of harmful chemicals kept out of the environment.
Additionally, AI can promote the integration of renewable energy sources into existing grids, accelerating the transition to cleaner energy sources. By predicting energy fluctuations from wind and solar power, AI can optimize grid management and balance supply and demand more effectively. This can lead to a significant reduction in reliance on fossil fuels and their associated emissions.
AI’s impact isn’t theoretical. Peanut farmers in India who took part in a program that used AI to help production saw a yield increase of 30 percent, and a McKinsey study suggests that AI can save 180 billion cubic meters of water by 2030. Furthermore, a project conducted by American Airlines, Google and Breakthrough Energy used AI to forecast plane contrails to help optimize flight paths. This research can help mitigate 35 percent of the aviation industry’s climate impact.
Dismissing AI based on initial hurdles would be closing doors for many environmental breakthroughs. By investing in sustainable infrastructure, implementing responsible e-waste management and acknowledging its potential, we can unlock a future where technology amplifies, not hinders, our environmental goals. The evidence is clear: AI can make waves, we just have to use it right.
Acknowledgement: The opinions expressed in this article are those of the individual author.
