NuScale Power Stock Broke Out in August. Is It a Buy?

NuScale Power rose 10.1% in August after announcing nuclear-focused AI tools that could cut engineering information-search time by up to 80%, but Q2 revenue collapsed 99% year over year to $75,000 after a Romania engineering project concluded. The company ended Q2 with $1.9 billion of cash and investments, up $900 million sequentially, and filed to sell up to $750 million of additional shares, creating further dilution risk. Despite NRC approval for its SMR design and potential AI-data-center power demand, NuScale's first reactor remains years from operation and its investment case depends on converting non-binding agreements into signed PPAs and factory orders.

Analysis

SMR’s valuation remains an option on contract conversion rather than a discounted operating business. The cash balance reduces near-term insolvency risk but does not solve the core problem: each quarter without a binding, financeable offtake agreement raises the probability that equity issuance funds overhead rather than value-accretive construction. The authorized equity capacity creates a persistent supply overhang, particularly after AI-related headlines attract momentum capital; absent a PPA, rallies are likely opportunities for the company to de-risk its balance sheet at shareholders’ expense.

The more investable AI-power beneficiaries are likely owners of existing dispatchable generation and credible development pipelines, not pre-revenue reactor designers. Constellation Energy (CEG) and Vistra (VST) can monetize tightening power markets through contracted capacity and power-price repricing now; Cameco (CCJ) and Centrus (LEU) offer nuclear-build optionality with less dependence on one project sponsor. SMR’s regulatory position has strategic value, but its commercial bottleneck is customer financing, construction-risk allocation, and interconnection—not document-search productivity.

Over the next 1-3 months, a signed PPA with creditworthy counterparty, defined capacity, construction timetable, and financing terms is the only catalyst that can support durable multiple expansion. Over 6-18 months, the decisive proof points are EPC/factory orders, fixed-price versus cost-plus risk allocation, and dilution-adjusted cash runway. The bearish thesis is falsified if SMR secures a bankable PPA plus committed project capital without materially discounted equity issuance; until then, consensus may still underappreciate how quickly a long-dated development story can be diluted.

Key Decisions for Investors

• Avoid initiating directional SMR longs into AI/nuclear momentum; treat any move unsupported by a binding PPA as a potential short-sale or put-spread setup after confirming elevated borrow availability and post-rally volume exhaustion. Thesis horizon: days to 3 months.
• Express data-center power demand through long CEG or VST versus short SMR in equal-dollar exposure for 3-6 months. The pair isolates near-term cash-generating power scarcity from SMR’s execution and financing risk; reassess if SMR announces a fully financed PPA.
• For nuclear fuel-cycle exposure, favor CCJ and selectively LEU over SMR on a 6-18 month horizon: new-build enthusiasm can tighten fuel/enrichment economics before any SMR deployment produces revenue. Key risk is uranium price retracement or reactor-build delays.
• Set an event alert—not a buy trigger—for SMR: require disclosure of counterparty credit quality, contracted MW, pricing/escalators, project financing, EPC guarantees, and expected equity contribution. A framework agreement or AI-product announcement without these terms should not change underwriting.
• Risk limit for any SMR short: cover on a binding PPA accompanied by committed non-recourse financing or a strategic investment that materially extends runway without discounted dilution; these events could drive a sharp speculative re-rating.