As generative artificial intelligence training and inference demands drive up global data center electricity consumption, the supply-side question of "who will meet this demand" has become a pressing reality.

U.S. nuclear technology company X-energy announced in late August a strategic partnership with Amazon (AWS), Korea Hydro & Nuclear Power Company (KHNP), and Doosan Enerbility to deploy over 5GW of Small Modular Reactors (SMRs) in the U.S. market, with AWS serving as the anchor customer.

This collaboration, mobilizing up to $50 billion in capital, aims not only to power AI and data centers but also to reshape a new value pathway across capital markets, supply chains, and the energy landscape.

**Nuclear Power and AI: Demand-Driven Logic**

Financial markets have long explained data center electricity growth through renewable energy, natural gas, and storage solutions.

However, in recent years, power consumption growth from AI training has exceeded expectations.

According to the International Energy Agency's (IEA) "Energy and AI" report released on April 10th this year, global data center electricity consumption could nearly double to 945TWh by 2030, with the United States being the primary load growth driver.

In this context, single sources like wind or solar cannot provide round-the-clock, low-carbon, predictable power, leading capital to reassess nuclear energy.

Amazon's choice carries clear financial logic: rather than merely binding power plants through Power Purchase Agreements (PPAs), it's directly entering upstream partnerships with reactor developers and equipment manufacturers to jointly build projects.

This means capital investment extends beyond electricity procurement costs to equipment manufacturing, fuel supply chains, and long-term financial arrangements, forming a "vertical integration" architecture.

If this model succeeds, it could transform uncontrollable power procurement risks into calculable investment returns.

The real challenge with Small Modular Reactors isn't technical principles but financing discipline.

Past harsh lessons from U.S. nuclear projects—Georgia's Vogtle third-generation reactors' delays and cost overruns, NuScale-UAMPS small reactor cancellation due to cost issues—tell capital markets that without replicable financing and delivery models, projects easily fall into "increasingly expensive" traps.

The X-energy and AWS collaboration is attempting to establish new order across two dimensions: financing and project execution.

**Financing Structure:** The up to $50 billion capital mobilization framework aims to support multiple projects in batches rather than financing a single demonstration plant. Capital deployment through phased investment and scaled procurement reduces project uncertainty.

**Project Execution:** Based on 80MW single modules, initially building 320MW-level plants with rolling expansion capability. This model reduces initial cash flow pressure while allowing investors to reassess midway.

From a financial perspective, this approaches portfolio investment logic: a group of projects launching in batches with gradual expansion, spreading returns and risks across different time periods.

Capital markets' primary uncertainty concern lies in supply chains.

SMRs generally require High-Assay Low-Enriched Uranium (HALEU) fuel, but current U.S. mass production capacity is limited.

While America's first commercial HALEU fuel manufacturing facility TRISO-X has received regulatory approval, production capacity ramp-up still requires several years. This directly impacts capital's discount rates for projects: if fuel supply gaps exist, even under-construction projects may face delayed commercial operations.

Amazon's introduction of Korean partners addresses this concern.

KHNP and Doosan possess mature experience and cost control capabilities in nuclear manufacturing, providing replicable manufacturing capacity for core components like pressure vessels, graphite components, and heat exchange systems.

For investors, this essentially uses supply chain stability to hedge project risks, making financing structures' interest rates and capital costs more controllable.

This approach resembles "risk sharing" in infrastructure investment: manufacturing and construction segments are secured by experienced international manufacturers, while the fuel segment relies on domestic capacity expansion from the Department of Energy (DOE) and TRISO-X (X-energy's wholly-owned subsidiary specializing in commercial TRISO fuel development).

Different risk segments are clearly divided and allocated to capable entities, forming a financing model acceptable to capital markets.

**IDC and Financial Market Anchoring Effects**

In traditional energy financing, power plants typically rely on utilities or power companies as electricity purchasers. However, in AI-driven energy landscapes, data center (IDC) companies directly become anchor customers. This shift dramatically changes capital structures.

As one of the world's largest cloud service providers, AWS's electricity demand is long-term and rigid. Capital markets favor demand sources with high credit backing. Financial investors are more willing to finance a group of power plants directly tied to AWS due to predictable cash flows and lower default risks.

This represents a structural change observed by capital markets: future nuclear financing may increasingly depend on large electricity consumers' credit rather than purely relying on electricity price markets or government subsidies. This makes nuclear power more like a "corporate energy debt instrument": generation project guarantees come from stable large customers rather than policy orientation.

Amazon and X-energy have already implemented a demonstration project in Washington State, initially four modules (320MW) with expansion reserved to 960MW.

Such demonstration plants' significance extends beyond power supply to forming financially replicable templates: once projects can deliver on schedule and budget, capital markets can use actual experience to calibrate risk models for future 5GW deployments.

Simultaneously, the Texas industrial thermal energy project with Dow Chemical provides another financial perspective: nuclear energy can power data centers while forming long-term thermal-electric bundled contracts with chemical and manufacturing industries. This diversified scenario makes financing less dependent on single revenue sources, improving debt and equity capital acceptance.

Regulatory pace directly affects capital confidence.

The Nuclear Regulatory Commission (NRC) has announced an 18-month construction license review timeline, significantly faster than before.

This sends positive signals to financial markets regarding reduced policy risks; however, community acceptance, waste management, and cooling water resource hidden costs still need incorporation into financial models.

When evaluating projects, capital markets typically hedge these uncertainties by adding risk premiums, which may also influence financing scale and interest rate variables.

**Conclusion**

Amazon and X-energy's collaboration isn't a single energy project but a capital model experiment.

Whether this experiment can power AI depends not only on technological maturity but also on financing discipline, supply chain resilience, customer credit, and policy coordination.

If the first plants in Washington State and Texas can achieve schedule, cost, and fuel security by around 2030, capital markets will rapidly replicate this model. The 5GW deployment plan would then transform from a slogan into a realistic pathway for structured financing projects.

In other words, this isn't purely an energy transition but a financial restructuring driven by AI demand: between power supply, capital operations, and customer credit, nuclear power is being redefined as "long-term asset allocation" for data centers.