How AI Emails Are Draining More Than Just Your Time

Every time someone uses AI to draft an email, the environmental cost might be higher than expected. Research shows that even a single AI request can drink up half a liter of water—about as much as a small water bottle. That might not sound like much, but when you consider how many emails get sent daily, the numbers add up fast. The issue isn’t just about typing out messages; it’s about the hidden water footprint behind every AI interaction. Data centers, the powerhouses running AI tools like ChatGPT, are thirsty giants. Some facilities use billions of gallons of water just to keep their servers cool. One Google data center alone slurped up over 6 billion gallons in a single year. And with companies pouring trillions into expanding these centers, the problem is only getting worse. By 2030, AI’s water use could rival what entire countries consume in a year. Yet, most people don’t even realize their digital habits have a physical cost.The problem isn’t just about the water used directly for cooling. AI models need electricity, and producing that power often requires water too. Training a single AI model can guzzle hundreds of thousands of gallons. That’s enough to fill thousands of bathtubs. Even if future AI systems become more efficient, demand will likely keep rising, canceling out any savings. So, while companies race to build more data centers, the water crisis looms larger. Surprisingly, many new data centers are popping up in drought-stricken areas like Texas. By 2040, AI could gobble up nearly 9% of the state’s water supply. That’s like putting a giant straw into an already shrinking resource. Researchers suggest smarter solutions, like tracking water use better or building centers in cooler, wetter regions. But so far, little has changed. For individuals, cutting back on AI emails won’t fix the problem. Over a billion people use AI daily, so one person’s choice won’t make a dent. Still, the bigger picture matters. While plastic straws get blamed for waste, AI’s environmental impact flies under the radar. Maybe it’s time to ask: what’s really worth the hidden cost?