We use Google Analytics to count anonymous page views and understand which content gets read. No ads, no profiles. Decline keeps you on cookieless mode. Details.
TeraWulf Inc.
Information Technology · IT Services
Power-backed AI compute through long-term hyperscale data center leases. WULF holds ~770 MW of contracted/permitted capacity across Lake Mariner + Cayuga with zero-carbon energy (nuclear PPA at Cayuga, hydro/grid at Lake Mariner) - the scarce input in the AI buildout.
Core42 (G42 subsidiary) anchors 200+ MW at Lake Mariner; Fluidstack lease + $GOOGL backstop adds ~200 MW more. Mining cash flow funds the AI conversion without dilution drag of pure-play neoclouds.
(1) Power is the bottleneck - WULF has interconnect + permits that take 4-7yr to replicate; (2) Core42 + Fluidstack/$GOOGL leases convert BTC volatility into fixed-tenor lease cash flow at data-center multiples; (3) Cayuga nuclear adjacency is rare among miners - fits hyperscaler net-zero PPAs; (4) Mining segment still generates EBITDA to self-fund capex; (5) Re-rate optionality from miner multiple (~8x) to DC landlord multiple ($DLR/$EQIX ~20x).
(1) Execution risk - liquid-cooled GPU buildout is new to WULF; (2) Hyperscaler tenant concentration (Core42 + Fluidstack ~ majority of AI capacity); (3) BTC cash flow halves with each halving cycle, narrowing the funding bridge; (4) Capex per MW for AI-ready buildout is 3-5x mining shells; (5) $IREN, $CIFR, $CORZ, $APLD all pivoting the same direction - power-PPA scarcity is the only moat.
The thesis, the numbers, the bear case and the level map for WULF, in one read. The deep version of what you just read.
Each weekday: the scan's ENTER setups with entry, stop and target, every new ticker teardown, and the bubble shifts that move names like this one. Free. Education, not advice.
No spam. One email per day max. Pro adds Telegram trade alerts and higher AI-assistant limits.
Auto-computed retracement of the trailing 52w low-high leg. Recomputed on each refresh; not a curated level.
Auto-computed retracement of the trailing 52w low-high leg. Recomputed on each refresh; not a curated level.
Auto-computed retracement of the trailing 52w low-high leg. Recomputed on each refresh; not a curated level.
Auto-computed retracement of the trailing 52w low-high leg. Recomputed on each refresh; not a curated level.