AI’s Water Footprint: ChatGPT Is Draining More Than Your Battery

Every time you ask ChatGPT a question, a precious resource silently trickles away. It’s not just the energy powering servers or the carbon emissions climbing into the atmosphere—it’s water. Fresh, irreplaceable water, vanishing into the air.

If that makes you uneasy, it should.

Generative AI, hailed as the future of human progress, is guzzling resources at an alarming rate. Each interaction with ChatGPT, the friendly AI assistant millions now rely on, wastes about half a litre of water. That might seem trivial, but consider the billions of daily interactions—and suddenly, you’re looking at rivers of water evaporating into thin air.

Why? Because the data centres that power ChatGPT need constant cooling to keep servers from overheating. This cooling process relies heavily on fresh water, which is pumped in, cooled, and then vaporised. In an age where droughts are becoming a grim reality for millions worldwide, the idea of throwing away water for chatbots feels almost dystopian.

AI's Thirst for Water and Energy

The problem doesn’t end with water. AI is an insatiable energy hog. Generative AI systems like ChatGPT are already consuming as much power as a small country. In fact, AI’s energy demands are on track to rival Japan’s—one of the world’s largest economies—by next year. This is 10 times the energy of a typical web search, tripling the tech industry’s energy use in just two years.

The climate crisis demands urgent action, yet we’re pouring precious resources into systems that actively worsen the problem. It’s a vicious cycle: AI drives climate change, and climate change heightens the demand for AI-powered solutions.

The irony? The very technologies marketed as tools to solve global problems are exacerbating the biggest one of all—our planet’s survival.

What’s the Real Cost of Convenience?

It’s easy to forget that every time you use generative AI, you're contributing to this resource drain. For you, it’s just a click. For the planet, it’s a cost we can’t afford.

The question we need to ask ourselves is: Is the convenience of instant answers worth drying up our water reserves? If every interaction evaporates half a litre of water, how many casual queries add up to a devastating environmental impact? Not to mention they tell us half-truths or outright lie to us most the time anyway!

This isn’t just about blaming consumers, either. The companies developing these systems must bear the brunt of responsibility. Generative AI may seem magical, but there’s nothing magical about the environmental toll it takes.

Is There a Better Way?

Experts like Professor Kate Crawford are calling for a complete rethinking of how we build and use AI systems. She suggests prioritising sustainability, renewable energy, and recycled water. AI should be designed with efficiency in mind, not just capability.

The technology is already here to make AI less harmful. Cooling systems that use seawater instead of fresh water, renewable energy-powered data centres, and low-energy AI models are all possibilities. But tech giants have been more focused on winning the AI race than on reducing their environmental footprint.

We’re at a critical juncture. The AI boom of the last two years brought shock and awe—but now it’s time for reflection. What kind of world do we want AI to build? One where innovation comes at the expense of the planet? Or one where progress is in harmony with sustainability?

Australia’s Role and Global Responsibility

Countries like Australia, already grappling with the harsh realities of climate change, have a unique opportunity to lead. By setting policies that mandate sustainable AI development, Australia could become a global role model.

The federal government has already begun discussions about managing AI infrastructure growth sustainably. But policies are only as effective as their enforcement. Tech giants must be held accountable for their environmental impact, and that includes transparency about water and energy usage.

Reimagining AI as a Climate Ally

The silver lining? AI has the potential to help combat the climate crisis if we use it wisely. Imagine AI forecasting weather patterns to optimise renewable energy, monitoring supply chains to reduce waste, or even creating low-energy systems that are self-sustaining.

We need to shift the narrative. AI shouldn’t just be about what it can do but about what it should do. Its purpose must align with preserving our planet, not exploiting it.

Final Thoughts

As generative AI becomes an integral part of daily life, it’s time to consider the true cost of our digital habits. The next time you type a question, remember the invisible toll: the water wasted, the energy consumed, and the planet pushed closer to its limits.

We don’t need to abandon AI, but we do need to demand better solutions—ones that don’t come at the environment’s expense. The future of AI doesn’t have to be a choice between progress and sustainability. It can be both, but only if we act now.

The question is, are we ready to pay attention? Or will we keep pouring water—and our future—down the drain?