TeraWulf stands out among bitcoin miners pivoting to AI data centers because its massive 20-year Anthropic deal gives it real legitimacy, and some capacity is already energized, moving it beyond pure hype. Plenty of execution and buildout risk remains before that partnership produces, and the chart is losing its uptrend line, suggesting more short-term chop. Long term, it should do well if the buildout goes well and the AI revolution is real. x.com 

TeraWulf's pivot from bitcoin mining to AI compute is drawing financing and analyst support after a reported $19B AI data-center lease with Anthropic. Per Bloomberg, citing CFO Patrick Fleury, it plans about $3.5B of leveraged loans and high-yield bonds, run by Morgan Stanley, to build the Kentucky campus under a 20-year Anthropic lease. Rosenblatt and Bernstein raised targets, with Bernstein seeing up to 70% upside. It must hold above $24.50 to keep its reclaim alive. x.com  x.com 

TeraWulf's reported $20 billion lease with Anthropic and roughly 3 gigawatts of compute capacity look like barely hidden fraud for a formerly bankrupt company now competing with Meta on capacity. x.com  x.com 

TeraWulf sits at AI's next bottleneck, power, having shifted its crypto mining facilities into leasing power for AI. The grid cannot deliver enough electricity to run the chips companies have bought at full capacity, and nuclear will take a long time to approve and regulate. TeraWulf has now announced a $19 billion deal with Anthropic. x.com 

TeraWulf should break all-time highs now on its Anthropic news, after neocloud stocks fell 30-50%. x.com 

TeraWulf deserved to rise on its Anthropic deal, and Sharon AI on a possible deal using its existing Australian data center footprint, unlike IREN, whose move is pure speculation with no physical footprint in Australia. Holding through the market's short-term flushes is the hard part of owning them long term. x.com 

Electricity is a clear bottleneck for AI, favoring AI-focused power and data-center infrastructure names over residential utilities. The US could be short 107 GW to power data centers through 2029, while China adds capacity year after year. Medium-term preferences are Cipher, Hut 8, Riot, TeraWulf, Galaxy, Equinix, Bloom Energy, Solaris, Vistra, GE Vernova, INIO, Williams, Liberty Energy, EROC and Cummins. x.com  x.com 

Alphabet's and Amazon's decades-long nuclear power deals show the AI buildout is not going backwards, since hyperscaler growth depends on securing electricity. Google committed to 20 years of nuclear power from Constellation Energy, supporting 890 MW of new capacity at 11 existing units, plus a separate 15-year deal for 2,700 MW, 3.59 GW combined. Amazon signed a 20-year deal supporting the Calvert Cliffs expansion. The AI opportunity extends well beyond chip designers and model builders. x.com 

AI infrastructure need not repay the entire late-decade buildout by 2030 because buildings and GPUs have years of earning life left. On an accrual basis, estimated revenue needed through 2030 is $4.3-5.1T, not Gartner's $8.5T, versus its $3.6T ceiling for AI revenue tied to data centers. Oracle's and CoreWeave's booked contracts also challenge Gartner's market sizing. For Nebius and TeraWulf, capital remains a risk; Nebius also faces renewal pricing into the 2027-28 supply wave. x.com 

AI infrastructure operators such as Core Scientific, CoreWeave, Galaxy Digital and TeraWulf benefit if out-year gigawatt projections are not met. Constraints on power generation, grid transmission, labor and wafer capacity limit how many debt-financed data centers and chips get built, which is healthier longer term if the build-out is funded from operating cash flow. @LEJCapital counters that this assumes demand keeps growing exponentially and stays far above supply. x.com  x.com