StartupsEventsJobsNewsTV
Founders·Investors·Ecosystem·
DutchStartup.ai
EventsJobsNewsTV
All articles

News

Can data centers be cooled with urine? A beer brand asked the question out loud

21 August 2026·3 min read

Can data centers be cooled with urine? A beer brand asked the question out loud

Beverage brands Liquid Death and Garage Beer launched a satirical advertisement on 18 August 2026 in which former NFL player Jason Kelce urinates into a jar and sings about "endless gallons of pee" as an alternative to drinking water for cooling data centers. It is a marketing stunt, but the water appetite of AI infrastructure is a real and growing issue.

Kelce is a co-owner of Garage Beer, along with his brother Travis Kelce, who became the company's largest investor and co-operator in June 2024. The campaign was conceived by comedian Sabrina Jalees and executed by LD Death Machine, Liquid Death's in-house advertising agency. As a by-product, the campaign released a Mason jar-style cup for eighteen dollars, labelled "Data Center Coolant". It sold out quickly. The product page carries a disclaimer: it is illegal to send urine by post unless it is a medical specimen meeting specific packaging requirements.

The underlying message taps into widespread public discontent over the arrival of large data centers. A Gallup survey from May 2026 found that seven in ten Americans are opposed to the construction of AI data centers in their neighbourhood.

How much water do data centers consume

The water consumption of data centers is substantial. A mid-sized American data center uses approximately 300,000 gallons of water per day (around 1.1 million litres) for cooling. Large hyperscale or AI-focused facilities can consume up to five million gallons per day, equivalent to nearly nineteen million litres. Amazon used 2.5 billion gallons of water globally for data center cooling in 2025.

Water is used in data centers as a coolant because it dissipates heat efficiently. As AI models become more computationally intensive and workloads increase, so does the demand for cooling capacity. In regions facing water scarcity, this is fuelling resistance among both residents and policymakers.

What would technically go wrong with urine as a coolant

Human urine consists of 91 to 95 percent water, which makes the comparison with drinking water seem less absurd at first glance than it sounds. However, experts and Google's own AI overview tool indicate that using urine in cooling systems would cause serious damage to equipment. The remaining five to nine percent consists of minerals, salts and organic compounds. These precipitate in pipes and heat exchangers, causing corrosion. In addition, the bacteria present in urine form what is known as biosludge, a biological deposit that clogs cooling pipes and significantly reduces thermal efficiency.

In other words: the joke has a kernel of truth regarding the chemical composition of urine, but the practical consequences for infrastructure make it no serious option.

What the campaign does reveal

The advertisement is primarily a brand activation for two beverage companies seeking positioning with a broad audience that is sceptical of Big Tech. That said, the campaign does ignite a debate that is growing louder in Europe and the Netherlands as well.

Dutch municipalities and provinces are regularly under pressure from hyperscalers requiring large amounts of energy and cooling water. In the Amsterdam metropolitan region, local authorities have already introduced restrictive measures for new data centers on several occasions, partly due to pressure on the electricity grid and spatial planning. The water dimension receives less attention in the Dutch debate, but is also relevant here during dry summers.

For investors and founders active in AI infrastructure or related cooling technology, the widespread public discontent presents opportunities: alternatives to water cooling, such as air cooling, immersion cooling or the reuse of residual heat, are high on the agenda of both data center operators and policymakers. The marketing stunt by Liquid Death and Garage Beer is a joke, but the search for more sustainable cooling solutions is not.

Relevant from our ecosystem

Weather Solutions NL B.V.Weather Solutions NL B.V.StartupNauwkeurige weerdata en -voorspellingen op maat voor energie, mobiliteit en landbouwaqa.earthaqa.earthStartupRealtime monitoring en voorspelling van bodem- en waterkwaliteitSpectralSpectralStartupSoftware voor energiebeheer en verduurzaming van gebouwen en assets

Relevant from our ecosystem

Weather Solutions NL B.V.Weather Solutions NL B.V.StartupNauwkeurige weerdata en -voorspellingen op maat voor energie, mobiliteit en landbouwaqa.earthaqa.earthStartupRealtime monitoring en voorspelling van bodem- en waterkwaliteitSpectralSpectralStartupSoftware voor energiebeheer en verduurzaming van gebouwen en assets
PreviousSimile raises $200 million and reaches $2 billion valuation with AI that predicts human behaviourNextTWINport360 raises €1 million for autonomous quality control in the bakery industry

Sources

This article draws in part on the following sources.

  • newsweek.com
  • si.com
  • ign.com
  • phillyvoice.com
  • pcmag.com
  • wersm.com
  • marketingdive.com
  • reddit.com

Related articles

EclecticIQ builds AI, but sells trust
dutchstartupyesterday

EclecticIQ builds AI, but sells trust

Almost every cybersecurity company is now adding an AI assistant to its platform, but Amsterdam-based EclecticIQ is looking for its competitive edge elsewhere. The question is not whether the AI works, but where this company's real differentiator actually lies.

PitchPitchClemberClemberRhiteRhite
A wave of new AI agent tools is emerging around early September 2026
aiyesterday

A wave of new AI agent tools is emerging around early September 2026

Around 4 September 2026, several new AI agent frameworks and developer tools were launched or updated, ranging from GitHub's multi-model orchestration preview HydraFusion to specialised tools for location data, security and shared terminal environments.

OpenAIOpenAIY CombinatorY CombinatorSequoia CapitalSequoia Capital
Pixyle AI CEO discusses product data ownership as an organisational problem in retail
dutchstartupyesterday

Pixyle AI CEO discusses product data ownership as an organisational problem in retail

Svetlana Kordumova, CEO of Amsterdam-based Pixyle AI, took part in a LinkedIn Live discussion on 23 June 2026 about fragmented product data and unclear accountability in the retail sector. The session was part of the "AI & eCommerce Talks" series.

Svetlana KordumovaSvetlana KordumovaRockstartRockstartSouth Central VenturesSouth Central Ventures

Watch next

How Famou Agent is Advancing AI for Science3:23
researchBaidu Inc.

How Famou Agent is Advancing AI for Science

Toename datacenters in AI-race | Met Clarice! | NPO Radio 11:32
policyNPO Radio 1

Toename datacenters in AI-race | Met Clarice! | NPO Radio 1

Microsoft datacenter gebruikt 1 procent van alle stroom in Nederland11:15
policyBNR

Microsoft datacenter gebruikt 1 procent van alle stroom in Nederland

This Shipping Container Powers 20,000 AI Chips13:41
researchSiraj Raval

This Shipping Container Powers 20,000 AI Chips

How international standards can make AI more energy efficient | ISO & IEC1:29:27
policyAI for Good

How international standards can make AI more energy efficient | ISO & IEC

DutchStartup.ai

The platform for the Dutch AI scene.

Add your startup

Discover

  • DS TV
  • Dutch-language videos
About·Contact·Privacy·Terms