The convergence of environmental crises, geopolitical tensions, and technological disruption by AI creates a landscape of profound uncertainty, one that is shaped by these mega-trends. As we race towards an AI-driven future, the question of how to secure sufficient energy to sustain this progress remains critical, pushing governments and companies to strategize to achieve the goal of leading in this area. This blog explores the approaches of key stakeholders and provides an overview of the current situation.
Racing Towards an AI-Driven Future
Uncertainty has become a prevalent aspect of today's global activities, especially as we witness interconnected, fundamental changes and disruptions. While we can assert that uncertainty is the most certain element we currently face, there is also an undeniable and unstoppable force: the advancement of AI.
Tech giants have declared AI as their top priority, and a diverse array of companies across all industries are eager to catch the wave with new AI advancements. According to Statista, in 2023, worldwide spending on AI-centric systems was estimated at USD 154 billion across all industries, with the banking sector’s investments at the top. At a country level, the US leads with a private investment of USD 67.2 billion, followed by China with USD 7.7 billion in 2023 (Source: HAI – Stanford University Human-Centered Artificial Intelligence).
The potential of AI is huge. The applications that individuals and companies are finding for it are increasing every day, and concerns about adoption are diminishing as enhancements make it more stable and accurate. At the same time, fears about job displacement are not unfounded, as a McKinsey report predicts that 400 million workers could be displaced by AI between 2016 and 2030. Upskilling in AI could be the solution for many workers to address the impact of AI.
At this stage of the AI race, we are all amazed and excited to explore new applications and push our limits into the unknown. However, a pertinent question remains: where will we obtain sufficient energy to sustain this progress?
The Challenge of Powering AI
Global energy consumption has been growing steadily, with electricity accounting for 20% in 2024, and it is expected to grow to 26% by 2035 (Source: World Energy Outlook 2024 – IEA). Electricity use has increased at twice the pace of overall energy consumption, driven by light industrial consumption, electric mobility, cooling, data centers, and AI. There are approximately 11,000 data centers worldwide, often spatially concentrated, which pose a significant impact on local electricity markets. In the US, a report from the US Department of Energy (DOE) states that data centers represented 4.4% of electricity consumption in 2023, and this figure is expected to reach as much as 12% by 2028. Although this is considered a small share, it still poses a challenge to energy systems when considering the concentration of their locations.
It is evident that the world needs to find more and new energy sources to sustain progress in AI and support economic growth. Fossil fuels still dominate global energy consumption, accounting for approximately 80% and generating a significant amount of CO2 emissions. Renewables represent 15%, and nuclear power accounts for 5% (Source: Energy Institute). This leads to the question: is it possible to procure a clean energy source for the growing demand of AI and new technologies?
The Nuclear Bet: A New Wave of Investments
When comparing nuclear energy with renewable sources, the latter not only lacks the potential and capacity to generate the required energy levels for the coming years but also produce waste and disrupt the ecosystems in which they operate. Therefore, they are not entirely harmless. Perhaps we are demonizing nuclear energy due to past accidents, which seem distant to the generations who were born after the Chernobyl disaster (1986) and are too young to remember Fukushima (2011).
Big tech companies have invested heavily in renewable energy but are also making big bets in nuclear energy. The year 2024 was decisive for new investments with Google, Amazon, and Meta announced partnerships with nuclear reactor developers. Microsoft followed suit by entering into a 20-year power purchase agreement with Constellation Energy to support the restart of the Three Mile Island Unit 1 nuclear plant, now renamed the Crane Clean Energy Center. Through this agreement, Microsoft will purchase the entire electric generating capacity of the plant. Additionally, the US Department of Energy (DOE) has been actively supporting nuclear energy initiatives, including a recent payment of USD 57 million in March 2025, as part of a USD 1.52 billion loan to reopen Holtec's Palisades reactor in Michigan. This initiative was part of the Biden administration's efforts to support nuclear energy as a clean energy option for the US. It underscores that nuclear energy is not only a private initiative but also a public one, that has continued with support from the Trump administration to increase domestic energy production. In Trump's case, this support does not come exclusively at the expense of clean sources, as his administration rolled back numerous environmental regulations to facilitate oil and gas drilling and declared a national energy emergency to further promote oil and gas extraction.
Geopolitical Strategies for Dominating AI Development
In addition to the energy required for AI development, there is concern about the supply of minerals used in the production of AI chips and the elements needed for nuclear reactors. China produces 60% of the world’s rare earth elements, which are vital for high-tech industries, and processes nearly 90% of them (including its own production and imports from other countries). This gives China a dominant market position and places the US in a complex situation regarding its ambitions to be a technology leader.
As the four major tech companies—Meta, Microsoft, Google, and Amazon—are based in the US, the country concentrates AI investments at a global level. It was no coincidence that the CEOs of these companies, along with executives from TikTok, Uber, OpenAI, Apple, and Tesla, attended Trump’s inauguration ceremony, demonstrating the priorities of this government. In this context, it is not surprising that President Trump is interfering in the peace negotiations between Ukraine and Russia while simultaneously securing access for US companies to exploit critical minerals in Ukraine. Similarly, there is interest in Greenland and the Democratic Republic of the Congo is offering the US exclusive access to critical minerals and infrastructure projects in exchange for security assistance as it battles a rebellion backed by Rwanda.
Essentially, the tech and AI race involves having a strategy to secure the supply chain to prevent a bottleneck. Regarding the impacts, there are significant environmental and social challenges associated with AI technology. On one hand, the mining of rare earth elements is considered highly polluting due to the processes used and is very demanding of water and energy. On the other hand, the use of energy from nuclear reactors poses risks that must be managed.
AI Regulations: Balancing Innovation and Sustainability
As is often the case with rapidly developing technologies like AI, regulations tend to lag. In the US, the number of AI regulations increased significantly with the Biden administration. However, this trend may have been interrupted, as in January this year, President Trump issued an executive order titled “Removing Barriers to American Leadership in Artificial Intelligence,” marking a shift towards deregulation. This represents a turnaround from the previous administration, which focused on oversight and risk mitigation to ensure the safe and responsible development of AI.
In the European Union, the AI Act entered into force in August last year, with a gradual implementation between this year and 2027. Compliance measures fall into two categories: risk management and data governance, and penalties can be as high as €35 million or 7% of global annual turnover for violations.
Regarding nuclear power, it seems that most countries agree on its development. In the US, 25 states have taken pro-nuclear actions, including passing laws that support the development and deployment of advanced nuclear technologies. Additionally, at the federal level, several initiatives have been implemented, such as funding for the research and development of Small Modular Reactors (SMRs) and other advanced nuclear technologies. In the EU, France is leading nuclear developments as part of the country’s energy strategy. Similarly, China, India, Russia, and South Korea are in a phase of expansion.
Conclusion
AI and nuclear power are here to stay, and although history may teach some valuable lessons on the cost of nuclear energy, the future also has much to say about fossil fuels and greenhouse gas emissions. The economy depends on energy, and we are far from decoupling this relationship; we expect energy demand to soar, along with the demand for critical minerals and rare earth elements used in nuclear power and AI. The US has made it evident that it wants to lead in AI, but just like with regulations, it remains to be seen how these pieces will fit into the puzzle to find the right balance between boosting AI and innovation, while simultaneously protecting people and the environment from unethical practices and pollution. As uncertainty is the most certain factor we face today, businesses and companies must adapt to navigate this ever-changing future, shaped by mega forces like AI, which is particularly relevant and exciting.
Evalueserve’s AI-enhanced managed services revolutionize business intelligence, fast-tracking decision-making and ensuring client competitiveness in a data-centric world.
Discover here how Evalueserve can add value to your business strategy.
