WASHINGTON (TNND) — As artificial intelligence continues to transform daily life and work, concerns are mounting over its environmental impact. The Government Accountability Office (GAO) reported that data centers, which power AI technologies, accounted for approximately 4% of U.S. electricity consumption in 2022. This figure could rise to 6% by 2026, though the specific impact of AI remains unclear due to a lack of transparency from companies.

The GAO identified five major risks associated with AI: lack of accountability, data privacy, cybersecurity, unsafe systems, and bias. However, measuring these risks is challenging as AI evolves rapidly and many tech companies remain tight-lipped about their operations.

Experts recommend improved data collection and transparency, urging developers to disclose energy use, carbon emissions, and water consumption. Despite this, many companies consider such information proprietary. There is also a push for innovation, with suggestions for governments and researchers to collaborate on creating more efficient models, hardware, and infrastructure. However, industry action may be slow without a clearer understanding of costs.

To mitigate risks to people, experts advocate for using existing AI accountability frameworks to enhance transparency and establish shared best practices and standards across industries. Achieving consensus between public and private sectors, however, is expected to take time.

The rapid growth of generative AI underscores the urgency of understanding its full impact on both people and the planet, as data to measure these effects is still catching up.