In 2025, AI and climate change, two of the biggest social disruptors we face, will collide.
Summer of 2024 Record broken For Earth’s hottest day since data collection began, it sparked widespread media coverage and public debate. This is also the year when both Microsoft And GoogleTwo of the major tech companies investing heavily in AI research and development missed their climate targets. While it has also made headlines and spurred outrage, the environmental impacts of AI are still far from common knowledge.
In fact, the current “bigger is better” paradigm of AI—characterized by the search for bigger, more powerful big language models offered by tech companies as the solution to every problem—comes with very significant costs to the environment. These range from generating massive amounts of energy to power data centers to running tools like ChatGPT and Midjourney to the millions of gallons of fresh water that are pumped through these data centers to ensure they last longer. Not hot and many rare earth metals. The hardware they contain is required to make them.
Data centers are already using 2 percent of global electricity. In countries such as Ireland, this figure rises to a fifth of the electricity generated, causing the Irish government to Effective deterrence on new data centers by 2028. While much of the energy used to power data centers is officially “carbon-neutral”, it relies on mechanisms such as renewable energy credits, which technically offset the emissions from generating this electricity. , but it doesn’t. Do not change the way it is prepared.
places like Data center street‘ In Virginia most are powered by non-renewable energy sources viz natural gasAnd energy providers are delaying retirements to keep coal power plants running Increased demands of technologies such as AI. Data centers are polluting large amounts of fresh water from scarce aquifers, pitting local communities against data center providers. Arizona to spain. in the TaiwanDuring the worst drought the country has seen in more than a century, the government chose to allocate precious water resources to chip manufacturing facilities to keep ahead of rising demands instead of letting local farmers use it to irrigate their crops. .
My latest research shows that the transition from old standard AI models – trained to perform a single task such as question-answering – to new generative models can be used. 30 times More energy to answer the exact same set of questions. Tech companies that are increasingly incorporating generative AI models into everything from search engines to text-processing software aren’t even disclosing the carbon cost of these changes — we still don’t know what ChatGPT’s conversation is about. How much energy is used during or in production. An image with Google’s Gemini.
Much of the Big Tech discourse surrounding AI’s environmental impacts has followed two trajectories: either it’s not really an issue (according to Bill Gates), or an energy will come with breakthroughs and magically fix things (according to Sam Altman). What we really need is more transparency around the environmental impacts of AI, such as through voluntary initiatives. AI Energy Star The project I’m leading will help users compare the energy efficiency of AI models to make informed decisions. I predict that in 2025, from national governments to intergovernmental organizations such as the United Nations, such voluntary initiatives will begin to be enacted by law. In 2025, with more research, public awareness, and regulation, we will finally begin to understand The environmental footprint of AI and take necessary actions to mitigate it.
