B.Riley reiterates New Era Energy stock Buy rating on power deal

B.Riley reiterated its Buy rating and $10 price target for New Era Energy & Digital after its subsidiary signed a 20-year PPA with Vistra affiliate Luminant for 200MW to 207MW for Phase 1 of its Texas data-center project, with power targeted for Q3 2027. The agreement resolves what B.Riley identified as the company’s largest outstanding project risk, though NUAI must post a $116 million letter of credit and will grant Vistra a 5% non-voting Phase 1 interest. The positive project milestone offsets weak Q2 2026 results, including a $0.21-per-share loss and just $36,500 of revenue; the company ended the quarter with $84.8 million of cash, cash equivalents, and restricted cash.

Analysis

NUAI has exchanged its largest development uncertainty for a financing and execution problem. The required $116M letter of credit exceeds reported quarter-end cash and restricted cash, so the relevant question is not the current ratio but the cost, collateral terms, and dilution embedded in any facility; a heavily collateralized LC or equity-linked financing could materially reduce residual equity value before first power delivery. With commercial operation still roughly 18 months away, the equity is likely to trade on tenant announcements, financing terms, and construction milestones rather than conventional earnings.

VST's economic exposure to a roughly 200MW commitment is unlikely to move consolidated estimates, but the strategic rights are more valuable than the initial energy sale: they give VST preferential access to incremental data-center load without requiring it to fund speculative campus development today. This is a modest positive for VST's Texas load-growth narrative and could support power-price optionality if ERCOT capacity tightens, while NUAI remains exposed to the opposite risk—its project returns compress if gas-fired power pricing or tenant power requirements outrun assumptions.

Consensus may treat a signed power arrangement as equivalent to a financeable data-center asset. It is not: absent disclosed power pricing, pass-through provisions, tenant pre-leases, project capex, and the LC funding source, there is no basis to underwrite project-level EBITDA or equity dilution. Near term, a promotional move in NUAI can persist on AI/data-center scarcity rhetoric, but a financing announcement that requires discounted equity, warrants, or restrictive security would be the key falsifier of the bullish thesis; conversely, an investment-grade tenant commitment with meaningful deposits would validate it.

The asymmetry favors treating NUAI as an event-driven watch rather than a core long. The company has ceded future-project optionality through preferential rights while retaining substantial construction, financing, and customer-acquisition risk, which can cap upside relative to the apparent headline de-risking.

Key Decisions for Investors

• Do not initiate a fundamental NUAI long before LC terms are public; set an alert for financing that limits new share issuance/warrant coverage to less than 15% and preserves at least 12 months of construction liquidity. A tenant pre-lease with deposits is the preferred entry catalyst over the power agreement alone.
• If NUAI rallies more than 25-30% without a disclosed funding package or contracted tenant, consider a small tactical short or put structure only where borrow/options liquidity permits; cover on a fully funded LC plus binding tenant announcement, since momentum can dominate micro-cap valuation.
• Maintain VST as the cleaner expression of ERCOT data-center demand, but do not add solely on this development. Add on weakness if management confirms incremental contracted load or reprices its Texas capacity outlook; the initial project is too small to change earnings independently.
• Monitor ERCOT forward power prices, Odessa-area interconnection/congestion signals, and natural-gas basis over the next 1-3 months. A sustained rise in delivered power costs without contractual pass-through would impair NUAI's tenant economics and strengthen the case for VST relative outperformance.