TeraWulf Powers On With New Tenant, Anthropic, Despite Capex/Debt Risks

Anthropic’s 20Y deal is being cited as evidence of durable AI compute demand, supporting WULF’s build-first-and-lease-later model. The acquisition of Morgantown Generating Station positions WULF as a power-centric digital infrastructure provider to monetize the AI super cycle. WULF’s reported $27B backlog underpins near-term cash flow, but the aggressive capex ramp implies balance sheet and earnings headwinds.

Analysis

This is less a pure AI-demand read-through than a validation of who can actually monetize that demand. For WULF, the critical variable is not backlog size but whether each incremental MW can be funded and brought online without equity dilution or step-up financing costs; in that sense, the stock trades like a credit story wrapped in an AI narrative. The near-term setup is positive for sentiment, but the medium-term equation still hinges on cost of capital and execution cadence, not just contract duration.

The second-order beneficiaries are the picks-and-shovels names that sell power, interconnect, cooling, and electrical gear, plus operators with already-funded capacity. Relative losers are other high-beta compute hosts that lack cheap power or balance-sheet flexibility, because this kind of announcement pushes the market to reward contracted power access over generic AI exposure. If WULF can repeatedly show non-dilutive project finance, the multiple can expand; if not, backlog becomes a longer-dated promise rather than a near-term earnings engine.

The contrarian view is that the market may be over-assigning present value to long-dated contracts while underpricing ramp risk. Any slip in commissioning, higher-than-expected capex, or a new equity raise would quickly flip the stock from AI winner to financing overhang, likely over days to weeks. Falsifiers to the bullish thesis are simple: delayed MW delivery, widening project spreads, or guidance that implies backlog converts slower than the current narrative assumes.

Key Decisions for Investors

• Do not chase WULF after the headline; wait for either a 10-15% pullback or evidence of non-dilutive project-level financing before adding exposure. The risk/reward is poor if the market immediately prices backlog as cash flow.
• If maintaining an AI infra basket, rotate toward names with cleaner funding profiles and already-committed capacity; use WULF only as a smaller, higher-beta sleeve. This preserves upside to the AI power theme while reducing dilution risk.
• Consider a relative-value pair: long WULF / short a less power-secured AI-hosting peer such as CORZ or IREN only after WULF shows funding discipline. Target a 1-3 month spread move; exit if WULF announces equity issuance or project delays.
• For directional upside, prefer a 6-12 month call spread instead of stock to capture optionality from backlog conversion while capping financing overhang. Best used only if investors believe the next catalyst is project finance, not just more contract signings.