Meta contaminates water in Wyoming: the hidden price of AI

What happened?

According to The Next Web, a contractor working on Meta's new artificial intelligence campus in Cheyenne, Wyoming, accidentally released a rare and potentially lethal bacteria into the city's wastewater system. The incident, which occurred during the construction of the data center that Meta plans to use to train cutting-edge AI models, was detected by local authorities, leading to the immediate suspension of all industrial discharges from data centers in the city. Although the specific identity of the bacteria has not been disclosed, sources close to the case indicate it is an opportunistic pathogen that can cause serious infections in immunocompromised individuals. These types of microorganisms are commonly found in soil and stagnant water, but their presence in an urban wastewater system poses a significant public health risk.

Why is it important?

This incident is a turning point in the debate over the environmental impact of artificial intelligence. Data centers consume enormous amounts of water for cooling, and demand has skyrocketed with the rise of generative AI. According to a 2024 report by the International Energy Agency (IEA), global water consumption by data centers could double by 2030, reaching 4.5 billion cubic meters annually, equivalent to the water consumption of a country like France. Meta, like Google, Microsoft, and Amazon, has multiplied its investments in AI infrastructure, often in water-stressed regions. Wyoming, an arid state with average annual precipitation of only 250 mm, is an example of how technological expansion can clash with natural resources and public health. Meta's Cheyenne campus, with an estimated investment of $800 million, will require approximately 5 million liters of water per day for cooling, according to estimates from the University of Wyoming.

Immediate consequences

• Suspension of discharges: Cheyenne has temporarily banned data centers from discharging industrial wastewater, which could delay the construction and operation of Meta's campus. The emergency municipal ordinance sets fines of up to $10,000 per day for non-compliance.
• Health risk: The released bacteria is potentially deadly, although no infections have been reported. Wyoming health authorities have initiated weekly monitoring of water quality in the sewer system and receiving water bodies, such as Crow Creek.
• Regulatory scrutiny: The incident could accelerate the implementation of stricter regulations on water use by data centers across the United States. The Environmental Protection Agency (EPA) has already announced it will review wastewater management protocols at high-tech facilities.

Broader context

This is not an isolated case. In 2024, a report by The Guardian revealed that Google's data centers in the Netherlands consumed as much water as a city of 100,000 inhabitants, and that in 2023, Google's water consumption grew 20% year-over-year, reaching 5.6 billion liters. In Chile, local communities protested against Amazon data center projects due to water use in a region with chronic drought, leading the company to commit to recycling 80% of the water used by 2027. In Spain, a Microsoft data center in Aragon was criticized for consuming 1.5 million liters of water per day in an agricultural area. AI not only consumes electricity but also water, and competition for this resource is intensifying. According to a study by the University of California, Riverside, training a large language model like GPT-3 can consume up to 700,000 liters of water, equivalent to a person's consumption over 10 years.

“AI has an insatiable thirst, and Wyoming is just the latest battleground,” said an analyst at TheVortiq.

What should readers know?

AI data centers require cooling systems that evaporate large volumes of water. Traditional cooling towers can lose up to 30% of water through evaporation, and the remainder is discharged as effluent with high concentrations of salts and chemicals. Companies like Meta are investing in more efficient cooling technologies, such as direct liquid cooling, which reduces water consumption by up to 50%, but the transition is slow. Meta has announced that by 2030, it plans for 50% of its new data centers to use liquid cooling, but currently only 10% of its facilities employ it. Meanwhile, local communities pay the price. In Cheyenne, residents have expressed concern at public meetings, demanding independent audits of Meta's water consumption. Investors and regulators must demand transparency in water use and mitigation plans, such as gray water reuse or rainwater harvesting.

Looking ahead

The Cheyenne case could set a legal and regulatory precedent. If more cities follow suit, tech companies will be forced to relocate their data centers or adopt water recycling technologies. In 2025, the city of Amsterdam already imposed restrictions on new data center construction due to water impact, and in Singapore, the government requires data centers to recycle at least 70% of the water used. Public pressure is also mounting: environmental organizations like Greenpeace have launched campaigns to highlight AI's water footprint, and a Change.org petition against Meta's Cheyenne campus has gathered over 50,000 signatures. In the long term, the industry may be forced to innovate: from underwater data centers, like Microsoft's Natick project, to air-cooling systems in cold regions. What is clear is that the Cheyenne incident is not an isolated accident but a wake-up call about the hidden costs of the AI revolution.