Artificial intelligence is developing faster than renewable energy. So, while individuals are asked to turn down the heating, GAFAM is relaunching nuclear power. The International Energy Agency is convening a global congress on artificial intelligence and energy on December 4-5, 2024.
By Thomas Le Goff, Télécom Paris – Institut Mines-Télécom
The American nuclear power plant “Three Mile Island” is infamous for having been the scene of one of the most terrible nuclear accidents in history in 1979, and she will soon return to service to power Microsoft’s artificial intelligence (AI) systems. This announcement, made in September 2024 and which concerns a reactor independent of the one which caused the 1979 accident, is part of a more global trend of massive investments by digital giants in nuclear energy.
Thus, Google also announced the signature of an agreement with the start-up Kairos Energyspecialized in the construction of small modular nuclear reactors (said “SMR”), to finance its development and reserve part of its production capacity for 2030. Amazon, following suit, concluded a similar partnership with the start-up X-energy.
The reason for these investments is simple: the exponential development of generative AI requires significant computing capabilitieslocated in particularly energy-intensive data centers.
The most recent studies show that AI represents between 10 and 20% of the electricity consumed by all data centers in the world, which increases by 20 to 40% each year according to the International Energy Agency (IEA). In some countries, such as Ireland, consumption linked to data centers has even exceeded the amount of electricity consumed by households.
The disproportion of these figures raises the question, moreover in a context where the climate emergency is on everyone’s minds and while citizens are asked to limit their heating to 19°C, is this race for computing capacity truly sustainable and desirable? Should we really seek by all means to build new electricity production capacities to keep up with the pace of development of data centers?
The solutions to this crisis are not obvious as there are so many divergent interests and factors to take into consideration. However, avenues for limiting the energy consumption of AI and the explosion in the number of data centers, such as taxation or regulation, are beginning to emerge in international discussions.
Why does AI need so much energy?
Every time we ask our favorite generative AI system a question, the request is sent over the Internet to be processed in a data center that may be located in different regions around the world. The latter consumes electricity to power the computer components it houses and its cooling system, not to mention the energy required to build the center and the electronic components themselves.
In recent years, the main AI models have increased in complexity and require ever greater computing capacities to operate, from 4 to 5 times more each year since 2010 according to the most recent studies. At the same time, the number of users continues to increase, with more than 200 million users every week just on ChatGPT.
These trends explain why AI providers need more and more power, are investing heavily in renewable energy to power their systems, and are projecting construction of new infrastructure all over the world.
Why is the proliferation of data centers a problem for the planet?
The acceleration in demand for computing capacity linked to the generative AI trend is accompanied by significant negative effects on the environment.
First, the production of electricity consumed by data centers generates greenhouse gas emissions depending on the source used. These emissions already represent 1 to 3% of global emissions according to the IEA and are likely to increase if the number of centers increases.
Then, data centers being particularly energy intensive, they can affect the stability of the network on a local scale. In an electrical network, the quantity of electricity produced must always be equal to the quantity of electricity consumed otherwise it is blackout (the breakdown). Adding infrastructure consuming a lot of electricity in geographic areas where the production-consumption balance is already fragile aggravates the risk of blackoutparticularly when the energy mix is largely based on renewable energies, which are intermittent in nature.
Finally, the pace of development of AI completely exceeds that of electricity production capacities from renewable energies such as photovoltaic panels or wind turbines. To meet their needs, digital giants will likely resort to carbon-based energy sources such as coal or gas, which are available more quickly. This leads them to catastrophically move away from their carbon neutrality objectives, Microsoft having posted a 29% increase in its emissions compared to 2020 and Google by 48% compared to 2019. At the same time, they communicate intensively on their investments in renewable energies in order to to forget their poor environmental performance.
What solutions to face the AI energy crisis?
The solution is not necessarily to ban the construction of new data centers, for three reasons.
Indeed, the new data centers built by digital giants are overall more efficient than old infrastructures. The construction of new centers also responds to other challenges since they contribute to the economic development of territories (in creating jobs and activity locally) but also to the establishment of sovereign computing power (for example in Europe), less subject to the potential effects of geopolitical disputes on an international scale.
Furthermore, unless there is a global moratorium on the construction of new infrastructure, prohibiting local implementation projects will only lead to their relocation, potentially to countries where the energy mix is even more carbon-intensive, which is not not desirable from an ecological point of view…
How can we better regulate the environmental impact of digital technology? Source: Telecom Paris.
The urgency of international reflection on the regulation of data centers
Like the European directive on energy efficiency and of European code of conduct applicable to data centersit is essential to ensure that each new project uses the best available technologies in terms of energy efficiency, but also to prevent consumption from increasing due torebound effectand is powered by low-carbon electricity. The more standards are harmonized at the global level, the less risk of relocation to countries with more flexible standards, but potentially less virtuous from an environmental point of view, will be significant.
Regulation of the number of data centers on a global scale could also be envisaged, via a global organization, on the model ofInternational Telecommunications Unionwhich manages the allocation of radio frequencies.
A reflection on the taxation of data center operators is also necessary in order to determine whether it can be used to promote the supply of green energy and adopting more sustainable practicesvia tax reductions or the establishment of a specific tax for the least virtuous operators. For example, the avenue was mentioned in the Senate’s information mission on the environmental footprint of digital technology in 2020, which led to the conditioning of a reduced tax for data centers respecting energy performance criteria, only in France.
Finally, it is also possible to act on the uses of AI. Raising public awareness of the environmental issues of AI would make it possible to direct uses towards a more virtuous use of the technology by limiting recreational uses, for example.
Very often, in debates on the environmental footprint of AI, the need to balance the negative externalities linked to its development, such as those mentioned in this article, with the potential positive effects that AI can bring in different sectors, especially economic (wealth creation) or environmental (reduction of emissions through optimization of the energy efficiency of other activities).
If the argument is attractive and seems rational, hypothetical positive effects in the long term cannot justify an unreasonable development of AI in the short term, causing irreversible damage to the environment and risking compromising our ability to bequeath to the future generations a healthy environment.
Thomas Le GoffLecturer in digital law and regulation, Télécom Paris – Institut Mines-Télécom
This article is republished from The Conversation under Creative Commons license. Read theoriginal article.