The energy cost of crypto and AI

Cryptocurrency mining, the process of validating transactions on blockchain networks, is notoriously energy-intensive. On the other side, AI is also a significant consumer of electricity. AI models like ChatGPT also rely on vast amounts of data for training and operation, which is stored and processed in energy-intensive data centres.

In the ever-evolving digital landscape, two of the most groundbreaking technologies—crypto assets and artificial intelligence (AI)—are reshaping industries. However, beyond their revolutionary potential lies a less glamorous reality: both technologies are voracious consumers of electricity. As the world becomes increasingly digitised, the energy demands of crypto and AI are soaring, posing significant environmental challenges.

Cryptocurrency mining, the process of validating transactions on blockchain networks, is notoriously energy-intensive. It requires high-performance computers to solve complex mathematical puzzles, ensuring network security and integrity. But this computational power comes at a steep cost. A single Bitcoin transaction, for example, can consume as much electricity as an average person in Ghana or Pakistan uses over three years. The scale of energy consumption in crypto mining is staggering, raising serious concerns about the sustainability of these digital currencies in their current form.

On the other side, AI is also a significant consumer of electricity. AI models like ChatGPT rely on vast amounts of data for training and operation, which is stored and processed in energy-intensive data centres around the world. These centres often require massive power to cool and maintain servers running these complex algorithms. A single query to an AI model like ChatGPT can consume up to ten times more electricity than a typical Google search, highlighting the immense energy demands associated with AI.

The combined energy demands of crypto mining and AI data centres are contributing to a significant increase in global electricity consumption. In 2022, these two sectors accounted for approximately 2% of global electricity demand—a figure expected to rise to 3.5% by 2026. To put this in perspective, this consumption is equivalent to the current electricity usage of Japan, the world’s fifth-largest consumer of electricity. The rapid growth in energy demand from these technologies is unsustainable and poses serious threats to global efforts to combat climate change.

The environmental impact of this energy consumption is profound. A recent working paper by the International Monetary Fund (IMF) estimates that crypto mining alone could contribute 0.7% of global carbon dioxide emissions by 2027. When the emissions from data centres are factored in, the total projected carbon emissions could reach 450 million tons by 2027, accounting for 1.2% of global emissions. This is a significant contribution to global warming, and it underscores the urgent need for regulatory intervention to curb the environmental damage caused by these technologies.

To address the environmental impact of crypto mining and data centres, targeted taxation is a necessary step. The IMF suggests that a direct tax of $0.047 per kilowatt-hour on crypto mining could incentivise the industry to reduce its carbon emissions in line with global climate goals. Similarly, a tax of $0.032 per kilowatt-hour on data centres could encourage the adoption of more energy-efficient practices.

However, the current situation is paradoxical. Many data centres and crypto miners enjoy generous tax exemptions and incentives, which only exacerbate the environmental impact by making it cheaper to consume large amounts of electricity. Policymakers need to re-evaluate these tax regimes, taking into account the environmental costs, the limited job creation in these sectors, and the strain they place on electrical grids.

At the same time, policy incentives should be designed to encourage the development of AI applications that deliver positive societal benefits while minimising environmental harm. Measures such as carbon pricing, credits for zero-emission agreements, and renewable energy certificates can help reduce the carbon footprint of these technologies and promote the use of renewable energy sources.

As the world races to mitigate the effects of climate change, it is clear that the energy demands of crypto and AI must be addressed. These technologies, while transformative, have an insatiable appetite for electricity that, if left unchecked, could undermine global efforts to reduce carbon emissions.

By implementing targeted taxation, promoting renewable energy use, and incentivising energy-efficient practices, we can balance harnessing the power of these technologies with protecting the environment. The goal is not to stifle innovation but to ensure it progresses sustainably and responsibly. In doing so, we can continue to benefit from crypto and AI while minimising their environmental footprint, securing a more sustainable future for all.

(The writer is a data privacy and technology lawyer)