Artificial intelligence is driving the largest surge in U.S. electricity demand in decades, but most companies deploying the technology are giving little weight to its environmental impact\u2014a gap that could raise costs and operational risk for wholesale distributors, according to new research from The Conference Board.<\/p>\n
Only 13% of corporate sustainability leaders surveyed said environmental impact is a major consideration in their companies\u2019 responsible AI strategies, while 42% said it is a minor concern or not a priority. The findings come as AI workloads fuel rapid expansion of data centers, intensifying demand for electricity and water in regions that also serve as major distribution and logistics hubs.<\/p>\n
For distributors, the implications are increasingly tangible. Energy-intensive warehouses, automated fulfillment systems, AI-driven pricing tools, and cloud-based procurement platforms all rely on data-center infrastructure whose power consumption is climbing sharply. Rising utility costs, grid constraints, and emissions-related regulations are beginning to flow through to operating expenses.<\/p>\n
The Conference Board estimates that data centers supporting AI now represent a growing share of U.S. electricity demand. Three of the largest U.S. cloud providers doubled their electricity consumption between 2021 and 2024, with their combined usage in 2024 equivalent to about 2% of all electricity generated in the U.S. More than 4,250 data centers operate nationwide, with one-third concentrated in California, Texas, and Virginia\u2014states that also host dense networks of distribution centers.<\/p>\n
Energy demand ranked as the top environmental concern among sustainability leaders surveyed. About 63% cited data-center electricity demand as a concern, 58% pointed to overall energy consumption and 56% cited greenhouse-gas emissions from power use. Water consumption also emerged as a growing issue.<\/p>\n
Despite those pressures, most companies are using AI for sustainability in limited, compliance-focused ways. About 34% of respondents said they apply AI primarily to sustainability disclosure and reporting, while far fewer are using it for operational improvements such as energy optimization, logistics efficiency, or water management.<\/p>\n
\u201cAs AI investment continues at record pace, its environmental footprint is becoming impossible to ignore,\u201d said Andrew Jones, principal researcher at The Conference Board and the report\u2019s author. He said companies that succeed in 2026 will be those that manage AI\u2019s resource demands while also using the technology to improve operational efficiency.<\/p>\n
For distributors, that operational gap is significant. AI tools can be used to reduce electricity consumption in distribution centers, optimize fleet routing to cut fuel use and improve inventory placement to lower transportation intensity. Yet those applications remain underutilized, even as margins tighten and energy costs rise.<\/p>\n
The report also highlights upstream pressures that may affect distributors\u2019 cost structures. AI hardware relies on resource-intensive components such as graphics processing units, advanced memory and specialized metals including cobalt, nickel, and rare earth elements. Faster hardware refresh cycles are increasing electronic waste, while recycling capacity for advanced AI components remains limited.<\/p>\n
\u201cAI\u2019s environmental story is not only about its footprint\u2014it is also a promising toolkit for sustainability performance,\u201d said Brian Campbell, leader of The Conference Board Governance & Sustainability Center. He said the highest-value opportunities lie in operational uses that can drive measurable reductions in energy and emissions.<\/p>\n
The survey was based on responses from more than 60 sustainability leaders at large U.S. and multinational companies. For distributors facing higher utility costs, expanding environmental, social, and governance (ESG) disclosure requirements and growing customer scrutiny, the findings underscore a widening gap between AI adoption and the operational discipline needed to manage its environmental and financial impact.<\/p>\n