About a year ago, I wrote a blog post on the environmental impacts of Starlink. It became one of our most-read pieces, and I couldn’t be prouder. A small part of me hopes that maybe, just maybe, we helped shed a little light from our sunny corner of the world.
This post follows a similar vein, though the topic is a little different. Over the last couple of years, AI has been the topic on everyone’s lips. It’s everywhere, all-encompassing, and it’s transforming the way we engage with content—for better and for worse.
One area that isn’t getting the attention it deserves, however, is the very real and very tangible impact of AI’s energy demands on public health. Inspired by a recent New Scientist article titled “Energy-hungry AI is already harming health – and it's getting worse,” this post dives into how the emissions from data centers powering AI models are causing far-reaching health consequences. The research backing these claims, conducted by the University of California, Riverside, makes for sobering reading, and I think it’s a discussion we need to have.
The Unseen Consequences of AI’s Energy Appetite
AI might be revolutionizing industries, but powering this revolution comes with a hefty price tag. Training large AI models demands vast amounts of electricity, and much of it still comes from burning fossil fuels. The resulting emissions aren’t just a climate issue—they’re a public health crisis in the making.
The numbers are eye-opening. The research estimates that by 2030, data center emissions could cause 600,000 asthma cases and $20 billion in public health costs annually in the U.S. alone. To put that in perspective, that’s roughly equivalent to the health cost of the entire U.S. steelmaking industry. What’s more, pollutants generated by these facilities don’t stay local—they can travel across state lines, impacting communities far removed from the data centers themselves.
In Virginia's "Data Center Alley," where many facilities rely on gas-powered backup generators, the impact is already being felt. Even under conservative estimates, these generators contribute to 14,000 asthma cases annually, with public health costs reaching $300 million. If they were used more frequently, costs could escalate to $3 billion per year.
A Dual Crisis: Climate and Health
While companies like Microsoft, Google, and Amazon are investing in renewable energy and exploring nuclear power, the reality is that many data centers remain reliant on fossil fuels. AI’s soaring popularity and demand are only exacerbating the problem. Without meaningful changes, we risk deepening health inequalities, as the burden of emissions disproportionately impacts low-income communities living near these facilities.
Benjamin Lee, a researcher from the University of Pennsylvania, describes this work as groundbreaking in quantifying AI’s health impacts. Yet, he also notes the importance of validating these findings further to ensure accuracy and refinement. Regardless, the implications are clear: AI’s environmental and health costs need to be part of the conversation.
What Can Be Done?
The good news is that there are tangible steps we can take to address these challenges:
Clean Energy Commitments: Tech companies must double down on renewable energy investments to offset their electricity usage. This includes building green energy infrastructure near data centers to minimize emissions at the source.
Energy Efficiency: From optimizing algorithms to using more energy-efficient hardware, developers can play a critical role in reducing AI’s energy footprint.
Responsible Siting: Companies need to consider public health impacts when deciding where to build data centers. Locating facilities near clean energy sources or away from densely populated areas could mitigate harm.
Policy Action: Governments have a role to play in encouraging greener practices, whether through stricter emissions regulations, tax incentives for clean energy use, or penalties for heavy polluters.
Transparency and Accountability: Businesses should openly share their energy usage and associated health impacts. Only through transparency can we hold organizations accountable and drive change.
A Moderate Perspective
AI is an extraordinary tool, but its potential should not blind us to its costs. From climate change to public health, the ripple effects of its energy demands are too significant to ignore. At the same time, this is not an unsolvable problem. By prioritizing responsible innovation, we can ensure AI continues to improve lives without jeopardizing the health of our communities.
The original research, conducted using the EPA’s COBRA tool, is a compelling read and underscores the urgency of this issue. You can explore it further here.
Let’s not wait until it’s too late. The future of AI shouldn’t come at the expense of our planet—or our health.
Link to the New Scientist Article: https://www.newscientist.com/article/2459657-energy-hungry-ai-is-already-harming-health-and-its-getting-worse/