Here’s a jaw-dropping fact: the energy required to power artificial intelligence is sparking a heated debate, and OpenAI’s CEO, Sam Altman, has just thrown a curveball into the conversation. But here’s where it gets controversial: Altman argues that the energy consumption of AI isn’t as alarming as it seems when compared to the resources needed to ‘train a human.’ Yes, you read that right—he’s drawing a parallel between raising a child and developing an AI model.
During a recent appearance at the AI Impact Summit in India, Altman told the Indian Express, ‘People often criticize the energy it takes to train an AI model, but let’s not forget—it takes a lot of energy to train a human, too. It’s a 20-year process, plus all the food and resources consumed along the way, before someone becomes truly capable.’ And this is the part most people miss: While he acknowledges the public’s concerns about AI’s energy use as ‘fair,’ he insists that the focus should shift to adopting sustainable energy sources like nuclear, wind, and solar—fast.
Altman’s comments come at a time when the environmental impact of AI is under intense scrutiny. Datacenters, the backbone of AI models, accounted for roughly 1.5% of global electricity consumption in 2024, according to the International Energy Agency. Worse, this demand is projected to grow by 15% annually until 2030—four times faster than other sectors. Noman Bashir, a fellow at MIT’s climate and sustainability consortium, warns, ‘The pace of datacenter construction is unsustainable. Most of the electricity powering these facilities still comes from fossil fuels.’ In December, over 230 environmental groups called for a moratorium on new datacenters in the U.S., citing threats to economic, environmental, and water security.
Here’s another bold claim: Altman dismisses concerns about water usage in datacenters as ‘totally fake.’ He argues that modern facilities no longer rely on water-intensive cooling methods, debunking viral claims that a single ChatGPT query consumes 17 gallons of water. CNBC supports this, reporting that newer datacenters are increasingly water-free. Yet, skeptics like Mike Weinstein, director of sustainability at Southern New Hampshire University, question whether AI’s resource demands are justified. He points to a 2025 OpenAI report revealing that 70% of ChatGPT usage isn’t work-related, raising doubts about its societal value.
Altman’s remarks have ignited a firestorm online. Matt Stoller, research director at the American Economic Liberties Project, slammed the comparison as ‘dystopian,’ tweeting, ‘Equating a baby with a spreadsheet? That’s morally bankrupt.’ Sports commentator Jeff Johnson drew parallels to Black Mirror, warning, ‘This tech-speak is dehumanizing. We’re letting these ‘geeks’ destroy the planet.’
Now, here’s the question for you: Is Altman’s defense of AI’s energy toll a fair comparison, or does it trivialize the unique value of human life? And more importantly, can we afford to ignore the environmental costs of AI’s rapid growth? Let’s debate—the comments are open.
