John Morgan and Cold Coffee are enduring the beginning of summer in Las Vegas as they ponder the future of society and AI’s impact on the world’s freshwater drinking supply. No, really. Plus there’s some MMA talk. Enjoy.
Hey if you don’t believe me ask ChatGPT or whichever you like: “Should we be worried about the fresh water supply as ai usage becomes more and more a reality and necessity of life?” The response I got is below….
‘Yes, concern about freshwater usage related to AI is legitimate—especially as AI becomes more integral to everyday life and operations. Here’s a concise breakdown of why and what to consider:
🔍 Why AI Impacts Freshwater Usage
- Data Center Cooling:
- AI runs on large data centers that consume immense electricity and produce a lot of heat.
- Many of these facilities use water-based cooling systems (like evaporative cooling), which consume millions of gallons of water annually.
- Training AI Models:
- Training a large language model (like GPT) can use hundreds of megawatt-hours of energy, indirectly translating to significant water consumption for electricity generation and cooling.
- Water-Energy Nexus:
- Electricity generation (especially in thermal power plants) requires water.
- More compute = more energy = more water use unless the grid is fully renewable and water-independent.
🌍 Real-World Implications
- Local Water Stress: If a data center is in an area already facing drought or water scarcity, AI expansion can exacerbate local water issues.
- Global Scaling: As AI applications scale across sectors (healthcare, education, logistics, etc.), the demand for infrastructure—and thus water—scales with it.
- Opaque Reporting: Many cloud and AI service providers don’t fully disclose water usage, making it hard to assess the true environmental impact.
🧩 What Can Be Done?
- Shift to Liquid-Immersion or Air Cooling: Reduces or eliminates water use entirely.
- Location Matters: Build data centers in water-rich, cool climates to reduce freshwater draw.
- Renewable + Low-Water Energy Sources: Transition away from coal or nuclear, which use more water, to solar and wind.
- Transparency: More tech companies are being pushed to report water usage metrics, not just carbon emissions.
🧠 Bottom Line
Yes, AI contributes to freshwater stress, especially when scaled massively. While it’s not the single largest water-consuming industry, its rapid growth and hidden footprint should make it part of the broader conversation on sustainable tech—particularly in water-stressed regions.
If you’re thinking long-term—about infrastructure planning, sustainability, or ethical AI—this is absolutely a dimension to monitor and advocate solutions for.’