OPINION: Generative AI Users have given up on the climate

By Brandon Harad ‘27

Courtesy of MIT News.

Generative artificial intelligence (gen-AI) is a social fixture at the Catlin Gabel School (CGS). I encounter gen-AI every day, from students “joking” about using it to complete their math homework, to teachers including AI-generated images on Canvas. Whether it sneaks into my day through a suspiciously bullet point-laden email or a strangely phrased explanation from a classmate in Chemistry class, gen-AI always manages to appear.

Gen-AI’s popularity significantly damages the environment. MIT Technology Review reports that gen-AI infrastructure consumes “4.4% of all the energy in the US” at “48% higher [carbon intensity] than the US average.” CGS community members know this: last year, an open letter circulated around the Upper School highlighted the environmental harms of gen-AI. CGS community members understand the severity of runaway climate change. Nevertheless, the CGS community uses gen-AI all the time.

CGS is not alone in this decision. Stanford reports that gen-AI adoption outpaces the initial adoption rates of the personal computer or Internet, even though Pew Research found that 80% of surveyed adults are willing to make substantial lifestyle changes to reduce climate change. Gen-AI users have to reconcile their usage with their impact on the climate. Conveniently, they have a Get Out of Jail Free Card: climate doomism.

The BBC describes climate doomism as “the idea that we are past the point of being able to do anything at all about global warming - and that mankind is highly likely to become extinct.” An AP poll found that climate doomism is spreading: each year, more people believe that their actions have no effect on climate change.

Doomism drives climate inaction. The AP elaborates that people who feel helpless in the face of climate change do not pursue eco-friendly choices. Worsening the climate crisis means “rising temperatures, sea level rise, drought, flooding, and more,” to quote the National Oceanic and Atmospheric Administration. While those outcomes terrify everyone, a climate doomist sees them as a reason to give up rather than act.

Another guise of climate doomism is the narrative that modern tools and research are insufficient to combat the climate crisis. We need a new approach, doomists argue, because they have already discarded existing solutions. Doomists are always on the lookout for new miracle solutions to climate change that scientists have not come up with yet.

Gen-AI companies prey on this idea by positioning gen-AI as that miracle solution. Google representatives writing in Fortune Magazine claimed that “AI could mitigate 5 to 10% of global greenhouse gas emissions by 2030” by optimizing ineffective emission reduction schemes.

Gen-AI users accommodate the harmful effects of gen-AI by ignoring them. Maybe gen-AI will save the world or maybe nothing can be done. Either way, users can stop thinking about it. Without hope to fuel climate action, nothing gets done.

Meanwhile, the climate harms of gen-AI are spiraling out of control. MIT Sloan reports that gen-AI greenhouse emissions will be “on the order of a tenth of a degree Celsius by 2100.” Even significant reductions in gen-AI emissions without reduction in usage “would lead to a negligible reduction in global warming,” the article elaborates. For reference, the United Nations reports that a reduction of even half a degree Celsius of warming would significantly diminish the effects of climate change.

On the other hand, gen-AI climate benefits are hard to find. A comprehensive study organized by Stand.Earth reports that claims that gen-AI will be a net climate benefit are unsubstantiated across the board. In fact, their analysis “did not uncover a single example where consumer generative systems such as ChatGPT, Gemini, or Copilot were leading to a material, verifiable, and substantial level of emissions reductions.”

Climate doomism is not supported by science in any of its forms. The World Wildlife Fund reports that scientists “still see a path forward” to climate crisis mitigation, even if we overshoot warming targets. APS News adds that “Current policies alone likely keep warming below… the ‘worst-case’ scenarios.” 

Every year, scientists reiterate that climate change is a policy issue, not a scientific mystery. The Climate Change Project makes clear that in 2026 “there are a million ways to save the climate.” Doomists wishing for climate miracles are overlooking the solutions that have been in front of them the whole time.

Not all climate doomsday scenarios are equally awful. Better outcomes are still feasible, and we know how to achieve them: coordinated community-wide movements motivated by hope for the climate. Climate change is terrifying, and that is exactly why action is urgently needed.

The speed with which gen-AI has spread throughout the world makes it hard to remain hopeful for mitigating its effects on the climate. Each time I encounter gen-AI in my day-to-day life, a bit of my motivation to oppose its proliferation fades.

CGS does not have to accept gen-AI’s environmental harms or the false narratives that it will one day solve climate change. If CGS intentionally reduces its gen-AI usage, other communities will be motivated to follow suit. Hope catalyzes the changes we need to make to avoid the worst impacts of climate change.

We cannot afford to ignore climate change, but we do not need to be paralyzed by fear. Climate doomism makes us feel powerless. The truth is: we all have power. Together, the CGS community has the ability to impact climate change. All we need to do is acknowledge the harm gen-AI usage creates, and address that harm as a community.