AI’s environmental impact remains secondary despite growing power and water demands, study finds

A new industry report highlights that while AI drives record electricity and water use, most firms prioritise ethics over environmental concerns, risking missed opportunities for climate benefits.

Artificial intelligence is reshaping corporate priorities as its appetite for electricity and cooling grows, yet environmental consequences remain secondary for many firms rolling out AI, according to a new industry study. The Conference Board report finds that while AI is driving the largest surge in U.S. electricity demand in decades, only a small minority of surveyed sustainability leaders treat environmental impact as a primary consideration in their companies’ responsible AI strategies. (Survey respondents were drawn from more than 60 sustainability executives at large U.S. and multinational companies.) Sources show firms are more likely to flag ethics, bias and security ahead of environmental concerns.
According to The Conference Board, roughly a third of sustainability leaders are deploying AI for disclosure and reporting on environmental performance, but far fewer are using the technology for operational changes that would materially cut emissions or energy use. That gap leaves the technology’s most promising climate applications, such as real‑time energy optimisation, predictive maintenance and logistics efficiencies, largely untapped.
Industry analysts and researchers warn that the physical infrastructure behind AI is imposing growing strains on power and water systems. Reporting on energy markets indicates households and utilities have already seen measurable increases in electricity prices as data centre demand rises, with some states recording sharp spikes in power costs as grids stretch to accommodate new loads. Meanwhile, scientific analysis highlights substantial water consumption tied to server cooling and electricity production for data centres, raising concerns about regional environmental stress as AI workloads scale.
The Conference Board’s principal researcher, Andrew Jones, framed the issue as a dual challenge and opportunity: “As AI investment continues at record pace, its environmental footprint is becoming impossible to ignore,” he said, noting that data centres account for a rising share of U.S. electricity demand and that AI’s water use is growing. He urged companies to manage AI’s resource demands while exploiting the technology to accelerate decarbonisation and grid optimisation.
The report and corroborating sources underline that the environmental effects of AI extend beyond firms’ direct control. Power generation mix, local water availability, and grid capacity shape the net footprint of AI services, meaning corporate decisions about cloud providers, data centre locations and workload scheduling can have outsized impacts. At the same time, experts note that unlocking AI’s environmental potential will require moving beyond analytics and disclosure toward deployment of operational solutions at scale.
For companies aiming to reconcile rapid AI expansion with sustainability goals, the path forward will likely involve tighter coordination between sustainability, IT and procurement teams, investment in energy‑and water‑efficient infrastructure, and greater use of AI to optimise resource use across operations. The Conference Board suggests that in 2026, market leaders will be those that both curb AI’s resource demands and apply AI to accelerate measurable sustainability outcomes.

Source Reference Map

Inspired by headline at: ^[1]

Sources by paragraph:

Source: Noah Wire Fuse Wire ServicesServices