Generative AI’s Rapid Growth Spurs Environmental and Societal Concerns

Generative artificial intelligence (AI) is expected to dramatically reshape workflows and productivity across numerous sectors, but new findings highlight critical risks that accompany its expansion. A report released April 22 by the U.S. Government Accountability Office (GAO) identifies significant concerns about the environmental footprint and societal impacts of these technologies, underscoring the need for clearer data and robust policy responses.

Generative AI, which powers tools ranging from content creation engines to customer service bots, demands substantial computing power. This leads to high consumption of electricity and water, especially within data centers. However, companies have largely avoided disclosing detailed information about these resource demands. Although some data exists on energy use and carbon emissions from training AI models, the water consumption required for cooling remains poorly documented. As generative AI continues to drive demand for data infrastructure, its precise contribution to rising energy usage—projected by the International Energy Agency to reach 6% of total U.S. electricity demand by 2026—remains uncertain.

Beyond the environmental impact, GAO outlines five key risks to humans and society that arise from generative AI. These include the potential for systems to produce misleading information, undesirable content, or to be exploited for malicious purposes. The rapid pace of AI development and a lack of transparency from developers make it difficult to fully assess or mitigate these risks. Additionally, generative AI may lead to workforce displacement and exacerbate national security vulnerabilities.

To address these challenges, GAO proposed several policy options. These include encouraging industry reporting on environmental impact, fostering innovation in more resource-efficient technologies, and promoting the adoption of risk management frameworks. Policymakers may also consider creating industry standards or sharing best practices to improve oversight and accountability. Still, implementation could face resistance due to proprietary concerns, costs, and the evolving nature of AI technologies.

Article by multiple RFHC contributors, based upon information from the U.S. Government Accountability Office (GAO), report GAO-25-107172.

---