Nuclear energy demand rises as AI data centers drive infrastructure boom

The rapid expansion of artificial intelligence is driving a massive surge in energy demand, positioning nuclear power as a critical solution for data center infrastructure. Goldman Sachs reports that data center power demand is expected to accelerate by 175% by 2030, a growth equivalent to a new top-10 power-consuming nation coming online.

Small modular reactors (SMRs) are emerging as a key technological focus due to their lower upfront costs, improved safety, and smaller footprints compared to traditional plants. Companies like NuScale Power and Oklo Inc. are leading this sector; NuScale targets utility operators for grid connections, while Oklo aims to co-locate reactors directly with AI data centers to bypass traditional utility intermediaries.

However, the AI infrastructure build-out faces financial and logistical hurdles. AI-related debt issuance reached nearly $500 billion by August 2026, with projections from Morgan Stanley and JPMorgan suggesting total AI infrastructure spending could reach between $3 trillion and $4.1 trillion by 2030. Investors are increasingly favoring safer projects backed by major cloud providers as credit environments tighten. Additionally, companies like Oracle have faced challenges, including power and permitting issues that have complicated data center development.