Former NFL center Jason Kelce just threw the tech industry a curveball nobody saw coming. During a recent podcast appearance, Kelce suggested data centers should use urine instead of drinking water for cooling - and the joke landed harder than expected. With AI data centers guzzling billions of gallons of freshwater annually, his offhand remark is sparking serious conversations about wastewater recycling in tech infrastructure. Turns out, the science isn't as crazy as it sounds.
Former Philadelphia Eagles center Jason Kelce just became an unlikely voice in the data center sustainability debate. His suggestion - delivered with characteristic humor - that tech companies should cool their massive AI facilities with human urine instead of drinking water has tech engineers doing something unexpected: taking him seriously.
The timing couldn't be more pointed. Microsoft, Google, and Amazon are racing to build AI infrastructure that's drinking up water supplies at an alarming rate. A single large language model training run can consume as much water as a small town uses in a week. According to recent estimates, data centers pulled roughly 1.7 billion gallons of water daily in 2024, with next-generation AI facilities pushing that number higher.
Kelce's comment, made during his podcast, initially seemed like pure comedy. But wastewater engineers and sustainability experts started chiming in on social media with a surprising take: the chemistry actually checks out. Human urine, after basic treatment, contains mostly water with dissolved salts - not fundamentally different from the non-potable water already used in some industrial cooling systems.
"The thermodynamic properties would work fine," one mechanical engineer noted on Twitter. "Evaporative cooling doesn't care if the water came from a reservoir or a bladder. You'd need filtration and some chemical treatment, but we're already doing that with greywater systems."
The real question isn't whether it's technically possible - it's whether the infrastructure makes sense. Collecting, treating, and transporting wastewater to data centers would require massive pipeline investments and regulatory approvals that could take years. Meta and other tech giants are already exploring greywater recycling and atmospheric water generation as alternatives to municipal drinking water.
Data center water usage has become a political flashpoint. In Arizona, where Google operates multiple facilities, local activists have protested the company's water consumption during ongoing drought conditions. Texas regulators are now requiring water impact assessments for new hyperscale data center permits. The pressure is mounting for tech companies to find alternatives to freshwater cooling.
Some facilities are already moving away from water-intensive evaporative cooling toward closed-loop liquid cooling systems that recirculate coolant without consuming water. Nvidia's latest GPU designs include direct-to-chip liquid cooling that can reduce water usage by up to 90% compared to traditional air conditioning.
But these systems cost significantly more upfront. As AI training demands explode - OpenAI's GPT-4 successor reportedly required facilities consuming 50 megawatts of power - companies are choosing speed over sustainability. Building a water-cooled facility remains faster and cheaper than implementing advanced liquid cooling, even if it means draining local water tables.
Kelce's joke lands because it highlights an absurdity: we're using drinking water to cool computers that generate chatbot responses while millions lack clean water access. Whether the solution is urine, greywater, or closed-loop systems, the status quo is becoming indefensible.
Some startups are already working on radical alternatives. One company is testing data centers cooled by treated sewage water in partnerships with municipal wastewater facilities. Another is exploring direct ocean water cooling for coastal data centers. The technology exists - it's the economics and regulations that lag behind.
The irony isn't lost on sustainability advocates. "We've spent decades telling people not to flush pharmaceuticals because they contaminate water supplies," one environmental engineer pointed out. "Now we're seriously discussing whether to pipe that same water to cool AI servers. The priorities are wild."
For now, tech giants are sticking with incremental improvements - more efficient chillers, better airflow management, and modest greywater pilots. But Kelce's comment has done something valuable: it's forced the industry to confront how absurd the current system looks from the outside. When a retired football player's bathroom joke contains more innovation than your sustainability roadmap, maybe it's time to rethink your approach.
What started as a podcast punchline has become a mirror reflecting tech's resource crisis. Jason Kelce's urine suggestion won't become industry standard anytime soon, but it's accomplishing something more important - forcing honest conversations about how sustainable our AI ambitions actually are. As data centers compete with households for drinking water in drought-stricken regions, the industry faces a choice: invest in genuinely innovative cooling solutions or keep defending the indefensible. Sometimes it takes an outsider's joke to expose what insiders have been avoiding. The tech giants can laugh it off, but they can't ignore the underlying question - if we're not willing to rethink something as basic as where our cooling water comes from, are we really serious about sustainability at all?
