Listen to This Article Enhance your reading experience by listening to this article.

The United Nations has raised concern that the advancement in technology will further complicate efforts to avert adverse effects of climate change globally if early interventions are not formulated.

United Nations Environment Programme (UNEP) notes that despite the environmental impact AI poses, the world is still adamant about addressing escalating challenges.

The agency has highlighted mining, emissions, water consumption, generation of E-waste and energy consumption as the major environmental impacts posed by the use of AI.   

“AI’s energy consumption and greenhouse gas emissions are significant, with more sophisticated models, such as large language models (LLMs), driving a significant increase in energy use. One study suggests that a single LLM query requires 2.9 watt-hours of electricity, compared with 0.3 watt-hours for a regular internet search,” a policy document calling for comprehensive assessment of AI’s impact on the environment published by UNEP said.

Global demand

The document also estimated that the global demand for water resulting from AI may reach 4.2–6.6 billion cubic metres in 2027.

“Semiconductor production requires large amounts of pure water, while data centres use water indirectly or for electricity generation and directly for cooling. The growing demand for data centres in warmer water-scarce regions adds to water management challenges, leading to increased tension over water use between data centres and human need,” the document reads in part.

Additionally, UNEP said that as AI use increases, it will generate higher volumes of electronic waste which is becoming one of the fastest-growing waste streams in the world.

Demand for chips

The report stated that the increased popularity of AI tools is driving a massive growth in demand for chips made from minerals and metals with the number of chips sold between 2001 to 2022 having quadrupled.

“Key minerals and metals needed for digitalization are nearly the same as those required for transitioning to a low-carbon economy. While there is no specific data on the demand for minerals in the AI sector, multiple studies suggest minerals and rare earth elements (REE) mining has significant environmental impacts, such as water and air contamination, biodiversity degradation and greenhouse gas emissions,” the report said.

“About 52 per cent of copper mines are located in high water stress areas [18]. The recycling rates for minerals remain rather low, varying between 46 per cent for copper and 1 per cent for REE.”

UNEP Executive Director Inger Andersen said that the environmental dimension of technologies such as AI is a ticking boom that the world cannot treat as an afterthought.

“I hope that the world has learned from our climate failures and climate procrastination, that prevention is not only better, but a lot easier, than the cure. AI produces significant greenhouse gas emissions. Data centres needed for AI are drawing more power, at a time when almost 800 million people lack access to electricity,” Andersen added.

She went on: “There are also serious impacts from the extraction of critical minerals and metals vital to AI chips and, of course, the energy transition. Water and air contamination. The degradation of indigenous people’s lands. Loss or displacement of biodiversity, including species such as Great Apes, our closest living relatives.”

Calling for solid policies on environmental stewardship, Andersen revealed that only one per cent of rare earth elements are recycled while for e-waste in general, only 22 per cent is recycled and disposed of in an environmentally sound manner.

UNEP said that AI’s rapid pace of innovation, exponential growth and demand for resources have raised significant concerns about its potential negative impact on the environment.

Additionally, the widespread use of AI has exacerbated existing inequalities, particularly affecting women, minorities and individuals from low-income backgrounds.

Consequently, UNEP has called on member states to establish standardized methods and metrics for measuring AI’s environmental impacts.

“The immediate priority should focus on the most concerning direct effects, and those for which data is more accessible, namely the consumption of energy, water and mineral resources, and the production of emissions and e-waste,” the agency said.

The revelations come at a time when world leaders have gathered in New York, US for the 79th session of the United Nations General Assembly (UNGA) where climate action has been among the key agenda that will be canvassed.