All over the world, the surge in Artificial Intelligence (AI) applications has sparked debates between the need for stricter regulation to address ethical and societal concerns, and the belief in its potential to boost economic prosperity and productivity. But amid all that discussion, we’re omitting a critical question: how do we align AI’s rapid growth with sustainability and efficiency standards?

The impact of AI on data center operations

One thing is certain: AI isn’t going anywhere, with adoption set to surge. Gartner forecasts suggests global spending on AI software will increase at a CAGR of 19.1 percent over the next six years. Much of this growth stems from enterprises’ hunger for High-Performance Computing (HPC), as they develop innovative applications and services for internal use and customer-related needs. Yet this growth also brings environmental concerns, particularly regarding increased energy demands and the need for advanced cooling systems in data centers to accommodate higher-density deployments.

Data centers are fundamental to AI adoption, providing the underpinning digital infrastructure for a range of innovative applications across numerous industries. However, they face unprecedented challenges in managing the complexity and intensity of AI workloads and balancing technological advancement with environmental sustainability.

Continuous improvement & innovation

Let’s be clear: AI hasn’t caught the industry by surprise. Dialogue around the growing demand for AI and data analytics, and the associated environmental implications, has been ongoing for nearly a decade.

Throughout that time, we’ve taken proactive measures to innovate and enhance our data center operations. We’ve led initiatives such as High-Density (HD) Colocation, where we not only optimize space for multiple customers' computing equipment but also significantly improve energy efficiency and reduce resource use. This approach, coupled with the adoption of advanced cooling systems like Direct Liquid Cooling (DLC), Air Assisted Liquid Cooling (AALC), and Rear Door Heat Exchangers (RDHx), optimizes the management of high-intensity AI workloads, ensuring our digital infrastructure handles their demands while reducing overall resource and carbon footprint.

Our upcoming AI data center in Portland, for example, is capable of accommodating the latest in cutting-edge technology, thanks to tailored infrastructure for AI deployments. Similarly, our new HPC data center in Paris is equipped with a DLC system that adapts to evolving infrastructure needs, optimizing energy efficiency and reducing costs.

Our modular infrastructure features “plug-and-play deployments,” adapting to workload fluctuations and accelerating go-to-market times for enterprises, while preventing overprovisioning and reducing energy consumption. These efforts underscore our commitment to simplifying AI adoption while promoting efficiency and sustainability within our data centers.

Construction plays a part in this equation too. We’re re-envisaging our data centers’ architecture to significantly reduce their environmental footprint, including designing buildings free from reliance on water and fossil fuels. Alongside energy reuse systems and district heating networks, which redistribute excess heat to nearby communities, these initiatives ensure our infrastructure not only meets customers’ technological needs but also integrates into and benefits the communities we serve.

AI’s role in building a sustainable future

While discussions about AI and sustainability often focus on their energy and water consumption, it's vital to recognize AI's potential in fostering environmental progress.

Even within data centers, AI can enhance efficiency and reduce environmental impact. Our in-house AI platform Apollo, for instance, has achieved significant energy savings across multiple sites, equivalent to powering over 1,600 US households annually, while streamlining operational processes, helping us meet targets outlined in the European Energy Directive (EED).

We’re also enhancing resilience and addressing capacity challenges in other ways. In London, our partnership with EkkoSense, sees us using its cutting-edge AI, immersive 3D visualizations, and thermal optimization analytics at our Cloud House data center to pinpoint and rectify airflow and cooling inefficiencies – delivering savings and greater efficiency. Through this, we’ve cut cooling energy consumption by 20 percent within a year.

Beyond data centers, AI is aiding climate change tracking and mitigation, for instance, by enhancing deforestation mapping through satellite imagery, combating forest loss, and cutting carbon emissions. In Brazil, AI drones expedite reforestation, achieving rates 100 times faster than traditional methods.

AI aids emission reduction by optimizing supply chains, improving demand prediction to combat overproduction, and enhancing transportation efficiency with data-driven models. In waste management, solutions like Greyparrot's AI boost recycling efficiency by analyzing waste processing facilities and reducing landfill waste through material recovery.

These diverse applications underscore AI's pivotal role in driving sustainability and tackling climate change, offering innovative solutions to global environmental challenges.

A greener AI future

AI is here to stay, and as industry leaders, we must recognize our responsibility to innovate and drive sustainability within the data center industry. While technology can pose sustainability challenges, it equally offers solutions to environmental and societal issues.

Therefore, as we advance technologically, prioritizing efficient, sustainable infrastructure is crucial for a future where technology and environmental stewardship coexist harmoniously to drive a greener future for everyone.

Aaron Binkley is Vice president of sustainability, at Digital Realty