Amazon has announced a landmark $500 million investment in X-energy to develop small modular nuclear reactors (SMRs) that will power its energy-intensive artificial intelligence operations. This move positions Amazon as the latest tech giant to embrace nuclear energy as a solution to the massive power demands created by AI development and generative AI services.
The investment, part of a Series C-1 financing round that includes prominent investors like Citadel’s Ken Griffin, Ares Management Corp., NGP, and the University of Michigan, will support the deployment of advanced modular nuclear reactors specifically designed to provide carbon-free power for Amazon Web Services (AWS) data centers. AWS provides cloud services to clients, including cutting-edge generative AI products that require enormous amounts of electricity to train and operate large language models.
Kevin Miller, Amazon’s vice president of global data centers, emphasized the dual challenge facing the company: meeting growing energy demands while addressing climate change. The partnership aims to develop more than 5 gigawatts of new power projects across the United States by 2039, with initial sites planned in Washington and Virginia through partnerships with Energy Northwest and Dominion Energy.
The Washington installation will begin with four 80-megawatt reactor units generating 320 megawatts of power, with the capability to scale up to 12 units delivering 960 megawatts. This modular approach offers flexibility compared to traditional coal-powered plants that generate 2,000 to 4,000 megawatts but produce significant carbon emissions.
Amazon’s nuclear investment follows a clear industry trend among Big Tech companies racing to secure clean energy for AI development. Microsoft recently helped advance plans to reopen the Three Mile Island nuclear plant, site of one of America’s worst nuclear disasters, while Google announced a partnership with Kairos Power for small-modular-reactor technology just days before Amazon’s announcement.
Earlier this year, Amazon purchased a $650 million data center in Pennsylvania with direct access to the Susquehanna Steam Electric Station nuclear plant. Additional proposals in New Jersey, Ohio, and Texas would build data centers adjacent to nuclear facilities to bypass traditional power grids entirely.
X-energy’s modular system distinguishes itself from traditional nuclear facilities by being “road-shippable” and installable where needed, offering unprecedented flexibility for tech companies seeking to power their AI infrastructure sustainably.
Key Quotes
We need smart solutions that can help us meet growing energy demands while also addressing climate change. X-energy’s technology will be integral in helping achieve this, and is an important step in Amazon’s work to achieve our Climate Pledge commitment to be net-zero by 2040.
Kevin Miller, Amazon’s vice president of global data centers, explained the strategic rationale behind the nuclear investment, highlighting the dual challenge of powering AI infrastructure while meeting environmental commitments.
Nuclear is an important source of clean and reliable power that our nation needs to meet the growing demand for energy. X-energy provides an impactful solution to a critical challenge — and the support Amazon, Dow, and other major corporations have provided underscores its potential and merit.
Ken Griffin, founder and CEO of Citadel and a major investor in the Series C-1 financing round, emphasized the broader significance of nuclear power for meeting national energy demands driven by AI and other technologies.
Our Take
The simultaneous nuclear investments by Amazon, Microsoft, and Google reveal an uncomfortable truth about AI: the technology’s energy appetite is fundamentally unsustainable with current infrastructure. This isn’t just about corporate responsibility—it’s about survival in the AI race. Companies that secure reliable, scalable energy sources will dominate the next decade of AI development. The embrace of nuclear power, including reopening Three Mile Island, shows how desperate the situation has become. What’s particularly notable is the shift to small modular reactors rather than traditional nuclear plants, suggesting tech companies want energy infrastructure they can control and deploy rapidly. This could fundamentally reshape both the energy and AI industries, creating a new competitive dimension where energy access becomes as important as computing talent or data. The 2039 timeline also indicates these companies are planning for AI’s energy demands to grow exponentially for at least another 15 years.
Why This Matters
This investment represents a critical inflection point in the AI industry’s energy crisis. As companies race to develop more powerful AI models and generative AI applications, the energy requirements have become staggering—training a single large language model can consume as much electricity as hundreds of homes use in a year. The convergence of Amazon, Microsoft, and Google all investing in nuclear power within weeks of each other signals that traditional energy sources cannot meet AI’s exponential growth.
The shift to nuclear power has profound implications for AI development timelines and sustainability. Without adequate clean energy sources, AI companies face a choice between slowing innovation or contributing massively to carbon emissions. Nuclear power offers a path forward that supports both rapid AI advancement and climate commitments. This also reshapes the competitive landscape—companies with secured energy sources will have significant advantages in training larger, more capable AI models. The investment further validates small modular reactor technology as commercially viable, potentially accelerating adoption across industries beyond tech.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources: