Microsoft’s recent deal to reopen the Three Mile Island nuclear plant marks a turning point in the relationship between artificial intelligence (AI) and energy infrastructure. The decision to power its AI projects with the plant’s entire output highlights the immense energy demands of advanced computing. It also shows the tech industry’s need for innovative, sustainable solutions. While nuclear power has long stirred debate, AI’s rapid growth demands an energy sector capable of evolving alongside it.
Three Mile Island’s scheduled reopening by 2028 isn’t just a logistical energy fix for Microsoft. It reflects a broader trend where tech companies face soaring power consumption as AI systems become embedded in daily life. From healthcare diagnostics to financial markets, AI requires immense computing power and, therefore, massive amounts of electricity.
Microsoft’s Bold Strategy: A Glimpse of the Future
Microsoft’s decision to turn to nuclear energy may seem bold, but it’s a practical solution. Known for forward-thinking strategies in technology and sustainability, this decision aligns with the company’s goals. Data centres, particularly on the scale required by global tech companies, consume vast amounts of energy. AI models are becoming more complex, driving their power requirements even higher.
The partnership between Microsoft and Constellation Energy signals the industry’s need for reliable, zero-emission power sources. AI data centres, which operate continuously, place immense pressure on existing grids. While tech companies commit to reducing their carbon footprints, traditional energy sources like fossil fuels, and even some renewables, struggle to meet this demand. Nuclear power stands out by providing consistent, high-output energy without the greenhouse gas emissions that undermine climate goals.
Microsoft’s decision likely sets a trend. Other companies, including Amazon, have pursued similar agreements. AI-driven technologies continue to consume energy at unprecedented levels, and companies that fail to secure stable energy sources risk falling behind.
AI and Energy: An Unbreakable Connection
AI’s energy demands are rapidly becoming a central issue. Technologies powered by AI now drive almost every sector—from logistics to entertainment, healthcare, and finance. These systems run continuously, processing vast datasets and performing real-time analysis, and that requires an enormous amount of electricity.
The world’s current energy infrastructure wasn’t built to accommodate AI’s needs. Many grids still rely heavily on fossil fuels, conflicting with zero-emissions goals. Renewables, while part of the solution, are often intermittent. They depend on local conditions, such as sunlight or wind, and their output fluctuates accordingly. This creates gaps that nuclear energy, with its steady and reliable production, can fill.
Without a major energy transition, current systems will buckle under the growing demand. Microsoft’s Three Mile Island deal shows that tech companies are actively seeking long-term solutions. The power generated from the plant, enough to supply 800,000 homes, will feed into the PJM Interconnection grid, which serves 65 million people. By tapping into this large-scale solution, Microsoft can meet its AI needs while stabilising the broader energy grid.
The Role of Nuclear Energy in AI’s Future
Nuclear energy has faced significant scrutiny, but its potential for consistency and reliability remains unmatched. Unlike wind or solar, which depend on environmental factors, nuclear power provides constant energy output. This makes it ideal for powering the AI-driven future, where uninterrupted power is crucial for data centres and machine learning models.
Many tech companies have ambitious sustainability goals. For example, Microsoft aims to become carbon-negative by 2030. Nuclear energy, with its zero-emissions profile during operation, is a natural fit for these goals. It offers a reliable, scalable solution for meeting the immense power demands of AI while also reducing the company’s overall carbon footprint.
However, nuclear energy poses challenges. Nuclear plants are expensive to build and maintain, and waste management remains a critical issue. Reopening decommissioned plants like Three Mile Island also involves complex regulatory hurdles. Critics argue that public subsidies shouldn’t fund such projects, particularly when the energy produced benefits private companies rather than public utilities.
The Rewards Are Significant
Despite the risks, the potential rewards of nuclear energy for AI and broader industry needs are substantial. Reopening Three Mile Island is expected to create 3,400 jobs and contribute $3 billion in tax revenue. The plant will stabilise an overburdened grid while serving as a blueprint for future tech-industry energy partnerships. Improved safety measures and waste management strategies ensure that nuclear energy can play a key role in AI’s future.
Is This the Beginning of a Trend?
Microsoft’s nuclear deal is unlikely to remain an isolated case. As AI expands, other tech giants will likely follow suit. Many will enter long-term energy agreements or invest in alternatives like next-generation nuclear reactors and fusion energy. Bill Gates’ TerraPower project is one example of how small modular reactors could offer more localised energy solutions for tech companies.
AI’s future hinges not just on better algorithms but on the ability to power those algorithms sustainably. Without clean, reliable energy sources, AI cannot achieve its full potential. Microsoft’s deal with Three Mile Island is a critical step in the right direction, but it’s only the beginning of a global effort to ensure that AI’s growth is powered responsibly.
Follow Scott Dylan on X for more insights
