Hyperscale Data Establishes $750 Million Minimum Sale Threshold for Michigan AI Data Center; Management Believes Value Could Reach or Exceed $1.25 Billion

Hyperscale Data said it would seriously consider selling its Michigan AI data-center campus only at valuations starting at $750 million, with management estimating a current value range of $750 million to $1.25 billion. An executed 10-year MSA for an initial 20 MW deployment could produce more than $1.2 billion over a maximum 20-year term; a customer option for an additional 32 MW could lift total contract revenue above $3.0 billion for 52 MW. The company is evaluating a campus sale, a minority IPO of subsidiary Sentinum to fund expansion, or continued development, though no transaction has been decided and further buildout remains contingent on financing, approvals and demand.

Analysis

The market should discount management's valuation framing heavily until the MSA economics are independently reconciled to capex, power-delivery timing, customer credit quality, and financing terms. The implied contracted-revenue figure is not equivalent to enterprise value: a 20 MW deployment requires substantial upfront electrical and GPU-adjacent infrastructure investment, while revenue concentration in one undisclosed neocloud counterparty creates renewal, utilization, and collectability risk. GPUS also carries conglomerate complexity and a future separation process that can absorb management attention and constrain clean valuation comparables.

Near term, the announcement can drive a retail-led re-rating because the stated asset-value range may be large relative to GPUS's current equity value; liquidity and borrow availability make the move potentially discontinuous rather than fundamentally efficient. The 1-3 month catalyst path is narrower: a named customer, committed power interconnect, construction milestones, project financing, or a formal process with third-party bids would validate the narrative. Absent those disclosures, the likely outcome is valuation leakage as investors capitalize only the operating 20 MW at a risk-adjusted multiple and assign little value to the uncontracted expansion option.

Contrarian view: the strategic-review language may be less a sale catalyst than a financing signal. A genuine buyer will underwrite delivered power, customer assignment rights, construction completion guarantees, and asset-level leverage—not management's comparison to AI-infrastructure public multiples. The thesis is falsified positively by binding financing or a signed LOI at or above the stated floor; negatively by equity issuance, related-party financing, delayed energization, or customer non-exercise of incremental capacity within the first two years.

Six-to-eighteen months, scarce powered capacity can justify substantial option value if the campus proves expandable without punitive dilution. But this is a binary execution story, not a clean AI-data-center multiple-arbitrage: small-cap capital structure, the legacy businesses, and Bitcoin exposure can dominate reported equity returns even if the facility itself performs.

Key Decisions for Investors

• Do not underwrite a core GPUS long solely on the announced valuation range. Treat as an event-driven watch position until SEC filings identify customer concentration, minimum revenue commitments, capex per MW, energization schedule, and committed funding.
• For high-risk event capital, consider a small GPUS long only after sustained volume confirms liquidity and the company discloses non-dilutive project financing or a credible third-party strategic-process milestone; target a 1-3 month catalyst window and size for total-loss risk typical of micro-cap execution stories.
• Avoid shorting the initial momentum without confirmed borrow: low-float, retail-sensitive AI narratives can trade materially above underwriting value. Reassess a tactical short only if the stock rerates on the announcement but subsequent filings show weak contract enforceability, unfinanced capex, or equity-funded construction.
• Set alerts for: binding asset-sale LOI, Sentinum IPO registration, customer identity/credit disclosure, utility interconnection agreement, construction completion dates, and any ATM/preferred/debt issuance. These datapoints determine whether value accrues to GPUS common or is transferred to project financiers.