As AI Drives Electricity Demand, Sustainability Remains a Low Priority

AI is driving the largest surge in US electricity demand in decades. Yet only 13% of surveyed sustainability leaders say environmental impact is a major consideration in their companies’ responsible AI strategy. Another 31% say it ranks behind ethics, bias, and security. That’s according to a new report by The Conference Board.

The study also examines how companies are using AI to advance sustainability goals. While about a third of surveyed leaders (34%) are using it for sustainability-related disclosure and reporting, far fewer are applying it to higher-impact operational improvements, such as energy optimization.

“As AI investment continues at record pace, its environmental footprint is becoming impossible to ignore. Data centers already account for a growing share of US electricity demand, and water use is rising as AI workloads scale. Yet the same technology is also unlocking new tools for decarbonization, grid optimization, and operational efficiency. In 2026, the leading companies will be those that take a dual lens—managing AI’s resource demands while harnessing AI to accelerate sustainability outcomes,” said Andrew Jones, author of the report and Principal Researcher at The Conference Board.

The survey draws on responses from more than 60 corporate sustainability leaders at large US and multinational companies. The report also offers a broader analysis of AI’s environmental impact and sustainability applications.

 1—Sustainability Leaders: AI Concerns & Uses

For most surveyed companies, environmental sustainability isn’t a major consideration in their responsible AI strategy—even as AI drives the largest surge in US electricity demand in decades.

• 4% of surveyed leaders say environmental sustainability is core to all responsible AI efforts.

• 9% say it’s a major consideration, but not the primary focus.
• 31% say it’s one of several priorities, but secondary to ethics, bias, and safety.
• 42% say it’s a minor consideration or not a priority.

Energy demand, consumption, and emissions are sustainability leaders’ top three concerns about AI’s environmental footprint.

• 63% of surveyed leaders cited data center energy demand as an environmental concern.
• 58% cited energy consumption.
• 56% cited GHG emissions from electricity use.
• 37% cited water use.
• Only 5% aren’t concerned about AI’s environmental footprint.

More than 60% of sustainability leaders are using AI for environmental objectives, most commonly for disclosure and reporting.

• 34% of surveyed leaders say they’re currently using AI for disclosure and reporting.
• 22% are using it for carbon accounting and emissions tracking.
• 15% are using it for climate risk modeling and scenario analysis.
• 12% are using it for circularity, waste, and water management.

“AI’s environmental story is not only about its footprint—it is also a promising toolkit for sustainability performance. While early applications center on reporting and disclosure, the highest-value opportunities lie in emerging operational uses that can drive far greater environmental impact,” said Brian Campbell, Leader of The Conference Board Governance & Sustainability Center.

2—AI’s Impact on the Environment

Data centers are expanding quickly to meet AI demand, creating significant environmental impacts.

• Hyperscale data centers: These large facilities, built to support AI workloads and cloud services, have doubled in the past five years—over half of which are in the US.
• Regional clusters: The US has over 4,250 data centers, 33% of them in California, Texas, or Virginia.
• Energy use: Three of the top US cloud providers doubled their electricity use between 2021 and 2024, and their total consumption in 2024 was equivalent to 2% of all electricity generated in the US.

Beyond energy and water, AI drives wider environmental pressures—often outside companies’ control.

• Resource-intensive hardware: Producing graphics processing units, high-bandwidth memory, and advanced packaging is carbon- and water-intensive.
• Construction and land use: Building AI data centers demands emissions-heavy materials (e.g., steel).
• Electronic waste: Faster hardware refresh cycles are increasing e-waste, while few facilities can recycle advanced AI components or recover critical minerals at scale.
• Critical-mineral dependence: AI hardware relies on cobalt, nickel, copper, rare earths, gallium, and germanium—all linked to significant land, water, and biodiversity impacts.

3—AI Applications for Sustainability

AI can pose environmental challenges but can also help advance sustainable outcomes.

• Early use cases: Streamlining data collection, analyzing emissions, and drafting reports and disclosures.

• Integrating data: Some firms are using AI-assisted carbon accounting tools to get more detailed data and integrate emissions with financial and operational systems.

AI’s operational promise remains largely untapped. Few companies are using AI for high-impact operational improvements such as energy and logistics optimization.

• Energy usage: AI can cut energy use by optimizing operations and detecting abnormal consumption.
• Logistics and fleet routing: AI-driven routing systems can reduce mileage and fuel use.
• Water and climate resilience: Tools like Google’s Flood Hub use machine learning and remote sensing to deliver local flood forecasts in dozens of countries, improving early warnings.
• Circularity and recycling: AI-guided computer vision and robotics are enhancing waste sorting.