Analysts rank NuScale, Oklo, and GE Vernova nuclear stocks by backlog depth rather than hype, validating backlog as the credible metric for datacentre-power project viability.

Nuclear energy stocks have attracted growing investor attention as artificial intelligence providers invest aggressively in data-center construction. This represents one of the largest building booms in recent history. Data centers are energy-intensive, and grid infrastructure in many regions has stagnated in recent years, leaving it ill-equipped to support the multiyear demand surge now underway.

Nuclear power offers an ideal solution—it can deliver substantial, reliable, low-carbon electricity. According to Goldman Sachs research: "Nuclear power will be a key part of a suite of new energy infrastructure built to meet surging data-center power demand driven by artificial intelligence. In the U.S. alone, big tech companies have signed new contracts for more than 10 GW of possible new nuclear capacity in the last year, and Goldman Sachs Research sees potential for three plants to be brought on line by 2030."

The challenge is structural: nuclear plants typically cost billions of dollars and take a decade or more to construct. AI companies need new power sources quickly and cannot afford to divert capital from computing infrastructure. This constraint has refocused attention on small modular reactors, or SMRs. Bank of America notes: "[N]ew advancements in technology may now make the tipping point in sight for small modular reactors to reshape nuclear energy supply chains over the next decade." The bank identifies five major advantages—lower upfront costs, improved safety, lower carbon emissions, smaller physical footprints, and modularity.

While dozens of companies pursue SMR technology globally, most remain privately held or embedded within larger, diversified conglomerates. Three publicly traded companies offer meaningful SMR exposure.

**Oklo** (OKLO) operates on the premise that AI providers will want to control their energy production directly. The company finances and constructs plants adjacent to data centers, selling power to AI companies through long-term purchase agreements. Oklo's project backlog stands at 18.1 gigawatts, though 12 gigawatts come from a single customer whose commitment is largely nonbinding. Many other deals are similarly nonbinding, though some include milestone prepayments. The company is backed by OpenAI founder Sam Altman, offering investors direct exposure to AI energy investment trends.

**NuScale Power** (SMR) pursues a more traditional model, selling SMR systems to utilities rather than directly to AI companies. Power then flows to data centers through standard grid mechanisms. NuScale's backlog totals between 6 and 7 gigawatts, concentrated almost entirely with a single eastern U.S. utility. If constructed, that system would become the world's largest SMR installation. Like Oklo, NuScale's backlog is largely nonbinding and carries cancellation risk. However, for investors preferring not to restructure energy delivery, NuScale is the stronger choice.

**GE Vernova** (GEV) is less focused on SMRs, which currently represent less than 3% of company sales, though this share could expand rapidly as adoption accelerates. The company's backlog quality, however, is the strongest here. Construction has already begun on its 1.2-gigawatt SMR system in Darlington, Ontario. Its 300-megawatt Tennessee project received construction clearance this week. GE Vernova offers the highest-quality backlog but the weakest SMR exposure, requiring investors to weigh growth potential against immediate market traction.