Startups & Business News

The AI Boom: Chasing More Data, Chips, Power, and Water

futureTEKnow
July 17, 2025

As artificial intelligence (AI) accelerates into every corner of technology, its demand for raw resources is reshaping the world behind the scenes. It’s not just about smarter software; it’s an era defined by a scramble for data, specialized chips, electrical power, and a surprisingly critical resource—water.

More Data and More Chips

At the heart of the AI surge is the insatiable need for massive datasets and high-performance chips. Training advanced models like large language models or image generators doesn’t just require clever algorithms—it needs access to oceans of information and cutting-edge semiconductors built specifically for AI workloads. These are not your ordinary computer chips; AI chips can consume up to 100 times more power than traditional servers, driving a monumental shift in data center design and energy use.

The Power Problem

With AI’s skyrocketing adoption, the energy footprint of data centers has become a major issue. In 2023, AI’s global power consumption was estimated at 4.5 gigawatts, accounting for about 8% of the total energy used by data centers that year. This substantial footprint is set to grow, with projections suggesting that by 2028, AI could require between 14 and 18.7 gigawatts of power, which would make up as much as 20% of total data center consumption globally.

The surge is fueled by the widespread adoption of power-intensive generative AI applications, such as large language models (LLMs), and the construction of new infrastructure to support them. This growing demand presents significant challenges for the energy sector, as well as for policymakers and tech companies, who must balance innovation with the need to manage and mitigate the environmental and grid impacts associated with AI’s rapid expansion.

Industry leaders are exploring liquid cooling and ultra-efficient chips to keep this growth in check, but the challenge is far from solved. Even with such innovations, the basic reality is that AI workloads demand orders of magnitude more energy than their traditional predecessors.

The Thirst for Water

What often goes unnoticed is how AI data centers consume water—especially in regions already suffering water stress. An average 100-megawatt data center can use around 2 million liters of water per day, equivalent to thousands of households’ needs. Most of this water is lost through evaporative cooling, a critical process for keeping power-hungry hardware from overheating. As a result, global water consumption by data centers could climb from 560 billion liters annually to over 1,200 billion liters by 2030.

The transparency around this water footprint remains elusive, with many companies and operators reluctant to share fine-grained usage data. Nevertheless, the impact is real and growing, impacting local communities and raising larger sustainability questions.

Not Enough to Go Around?

There’s a mounting concern not just about energy and water, but also supply chain strains, especially for AI-specific chips. Forecasts show that, to satisfy projected growth, the U.S. would need more than 90% of the world’s new AI chip supply from 2025 onward—a level of dominance that seems unlikely to materialize without significant changes in production or policy.

Smarter Solutions on the Horizon

Faced with these challenges, tech giants and researchers are pushing for smarter, more sustainable innovations:

Optimizing electricity usage through AI-powered sensors and advanced algorithms.
Adopting liquid cooling to reduce overall water and energy demands.
Accelerating development of energy-efficient AI hardware.
Seeking alternative energy sources to offset environmental impact.

The Road Ahead

The future will depend on how quickly the industry can pivot to more resource-aware designs—balancing the amazing potential of AI with its very real costs. As the world gets smarter, the physical and environmental realities behind the curtain become impossible to ignore.

AI might be digital, but its story is as much about the material world as it is about algorithms and code. The next wave of progress will require not just better models—but an all-new blueprint for the infrastructure powering the intelligent future.

futureTEKnow is a leading source for Technology, Startups, and Business News, spotlighting the most innovative companies and breakthrough trends in emerging tech sectors like Artificial Intelligence (AI), Robotics, and the Space Industry.

Become a contributor

Find out how

Discover the companies and startups shaping tomorrow — explore the future of technology today.