Hanley Sustainability Institute

By Rose Olson

As a student leader working with the Hanley Sustainability Institute on my college campus, I spend a lot of time thinking about how we can make systems more sustainable — whether that’s food waste, water use or campus-wide energy practices. But lately, I’ve found myself in a moral tug-of-war over a growing issue: artificial intelligence. 

Like many people, I have been amazed to see all of the things AI can do, especially in the world of sustainability. It can help predict climate trends, help scientists analyze huge sets of data and so much more. While it can do such big and important things, most of my peers use it in smaller ways to help with school work. Whether that is to help create practice tests to study for exams or help explain certain homework problems to better understand a topic, all of these things significantly help students to learn and maximize their time and effort put towards courses.

But at the same time, I’ve also learned how much environmental harm this technology can cause — harm that is invisible behind sleek interfaces and algorithms. As I witness individuals benefiting from this technology, with zero thought about how it affects our planet, I am forced to feel the conflict on my own use of AI in academic pursuits, the workplace and personal curiosity.

After researching and better understanding the hidden environmental impacts of AI, I can now see how much bigger of a problem it is than I realized. Training large AI models — like ChatGPT, Google’s Gemini or image generators — requires enormous computational power. These processes take place in data centers that consume vast amounts of electricity, often coming from fossil fuels. For example, training a single large model can emit over 284,000 kg carbon dioxide, which is roughly equivalent to five round-trip flights between New York and London. This is a huge amount of fossil fuel emissions that are irreversible. 

And the energy demands don’t stop after training. AI models must constantly run to serve millions of users in real time. This increases the need for massive server farms, which require constant cooling and maintenance, further increasing energy use and water consumption. In 2023, a report revealed that Google’s data centers consumed over 5.6 billion gallons of water — mainly for cooling AI systems. These are resources that are becoming more scarce in a warming world. 

To be fair, AI also helps the environment in many ways. It’s used to monitor deforestation, model climate scenarios, optimize energy use, and detect illegal fishing or poaching. AI-driven efficiency can reduce emissions in logistics, agriculture and manufacturing. These are not small gains. But ultimately the moral conflict arises: can we justify environmental damage in the name of innovation, even if that innovation is supposed to help the environment? 

It’s a question of means versus end. If the energy used to power AI systems comes from fossil fuels, are we just shifting the burden of climate mitigation onto a new source of pollution? And how do we weigh the benefits AI brings to healthcare, accessibility and disaster response against the long-term ecological toll of powering it?
AI might offer tools to help us, but if it’s being powered by unsustainable systems, then the solution becomes part of the problem. 

This all makes me ask myself a tough question: Should I, a student promoting sustainability, be using AI at all? It’s something I have thought about more and more. On one hand, AI helps me be more efficient — I’ve used it to brainstorm ideas for campus events or even draft educational content. On the other hand, by using these tools, I’m contributing to the environmental costs I’m trying to reduce.

It feels hypocritical at times. But maybe the answer isn’t to completely avoid AI, but to use it more mindfully. If I’m aware of the trade-offs and push for more responsible practices — both in how I use it and how it’s built — maybe I can push towards a solution rather than be a silent contributor to the problem. It’s an ongoing ethical balance, and I think a lot of students like me are starting to see it, too. 

It’s easy to feel helpless, but I genuinely believe we have power — especially within campus communities where change can start small and ripple outward. Here are a few things I’m trying to keep in mind and bring into my work: 

• Push for Transparency: If universities are using AI tools or hosting data centers, they should be transparent about their environmental costs. 
• Advocate for Green AI: Support research and projects that focus on making AI more energy- and data-efficient. 
• Encourage Responsible Use: Just because we can use AI doesn’t mean we always should. Ultimately we, including myself, must be more intentional about where and how AI is being used. 
• Choose Renewable-Powered Tools: When possible, promote platforms and companies that power their AI with renewable energy.

Being in the sustainability field means living in tension between innovation and preservation. AI is one of the most powerful tools of our time, but if we want to use it responsibly, we need to hold it to the same sustainability standards we demand in other industries. I don’t have all the answers but I do know that if we’re serious about protecting the planet, we can’t ignore the environmental cost of the tools we use to save it. And maybe that starts with conversations like this — on campus, in classrooms and in communities where we’re shaping the future.