Digi Power X (DGXX) Q2 2026 Earnings Call Transcript

Digi Power X reported Q2 revenue of $6.6M (with AI-driven GPU bare-metal revenue of $1.1M, about five weeks of operations) and positive adjusted EBITDA of $3.3M, up $3.2M YoY, but posted a $14.4M net loss. The company secured a 10-year contracted AI revenue stream worth $1.1B (expansion option to $2.5B) and is targeting Phase 1 AI campus service in Dec. 2026 (15MW) and Phase 2 by March 2027 (25MW). Management highlighted stronger liquidity ($150M cash as of Aug. 14, 2026 vs $1.7M a year earlier) to mitigate dilution via ATM-to-cash build and ongoing debt financing discussions for Alabama data center expansion, while citing NY’s data-center moratorium as limiting power footprint expansion beyond the existing 78MW.

Analysis

The important read-through is not the AI revenue itself; it is whether the market starts capitalizing this as a project-finance story instead of a perpetual-dilution story. If lenders accept GPU residual value and the company can lock asset-backed debt, equity value can re-rate because incremental growth would stop being funded at the stock level. That is a real positive for financing-heavy intermediaries and GPU suppliers, but the benefit is second-order; the first-order winner is still management if it can finally stop using equity as bridge capital.

The near-term risk is that the operating narrative outruns the balance sheet reality. A five- to seven-week revenue sample on the new fleet is too small to prove utilization, and any slippage in equipment delivery or debt syndication quickly converts this back into a cash-burn-and-dilution story. The key falsifier over the next 1-3 months is not enthusiasm; it is whether financing terms are signed and whether Phase 1 stays on the December 2026 path without another equity raise.

Contrarian view: consensus is likely underweighting how binary this is. The market may treat this as an AI infrastructure compounder, but the moat is not software, customer stickiness, or scale economics yet; it is access to capital and power, both of which can disappear fast if lenders balk or GPU collateral marks down. Over 6-18 months, the upside case is real if debt closes and the campus ramps, but absent that, the stock is still mostly an optionality vehicle with heavy dilution overhang.

Key Decisions for Investors

• DGXX: do not chase the initial optimism; wait for a signed debt facility and firm Phase 1 delivery evidence. If those are confirmed, treat it as a high-beta momentum long for 3-6 months; if not, fade any post-call strength because dilution risk remains the dominant factor.
• If borrow/liquidity allow, express a bearish event trade in DGXX via short stock or put spreads into strength over the next 2-8 weeks. The thesis breaks if management announces non-dilutive financing and on-time energization with no incremental equity issuance.
• Relative value: long GS/BLK/KKR basket versus short DGXX only as a financing-quality vs execution-risk pair. The upside here is from capital-market fee pools and asset-backed lending optionality, while DGXX remains the balance-sheet risk bearer.
• Watch DUK and other Southeast utility/load names as a 6-18 month secondary beneficiary if Alabama and North Carolina campuses convert into real MW demand. This is an alert, not a trade, until permitting and interconnection milestones become independently verifiable.