Disclaimer: This analysis is for informational and educational purposes only and does not constitute investment advice. All investments carry risk, including the risk of loss. Past performance does not guarantee future results. Please consult with a qualified financial advisor before making investment decisions.
Position Disclosure: The author holds positions in the following symbols: AAOI, AAPL, ACHR, ALAB, AMD, AMPX, AMZN, APLD, APP, ASTS, AUR, AVAV, BBAI, BFLY, BKSY, BRK.B, CEG, CLS, COHR, CRDO, CRSP, CRWV, ESTC, ETH, FN, GOOG, GOOGL, HURA, HUT, IBIT, INOD, IONQ, IREN, JOBY, KTOS, LITE, LMND, MBLY, META, MSFT, NBIS, OKLO, OUST, PDYN, PL, PLTR, POET, QBTS, QUBT, RCAT, RGTI, RKLB, RLAY, RR, SDGR, SERV, SMR, TEM, USMV, VGT, VICR, VRT.
Every week, we score ~120 public companies on one question: as AI changes how businesses think, build, and compete, who benefits most? These five ranked highest this week — a sample of the full analysis available to members.
How We Picked Them: Selected from our full coverage based on growth potential, AI advantage, and how well-positioned each company is as of April 28, 2026. We diversified across sectors so you see a range of opportunities, not just one hot corner of the market.
Also available as a PDF download.
IREN Limited (IREN)
ai
cloud
energy
crypto
hardware
Synopsis
This is a real AI infrastructure asset story, not just a miner
rerate. The upside comes from turning scarce powered campuses into contracted compute revenue faster than
dilution, pricing resets and build delays erode per-share value.
Thesis
IREN is becoming a power-backed AI infrastructure developer rather than just a miner; if it keeps converting scarce secured power into contracted compute and campus revenue faster than
dilution rises, revenue can plausibly scale from 750 to 6000 by 2031 and support a low-to-mid 3x enterprise-value outcome.
Last Economy Alignment (0.7/1.0)
IREN sells a hard AI bottleneck: powered campuses, delivered compute capacity and deployment speed. Software commoditization and agent bypass risk are low, but capital intensity and pricing resets can
cap value capture.
Critique
Scarce power is valuable, but IREN still monetizes contracted capacity; if big clouds internalize demand,
GPU pricing resets, or financing keeps swelling the share count, strong asset growth may not become strong per-share value.
Tempus AI, Inc. (TEM)
healthcare
ai
biotech
software
enterprise
Synopsis
The core bull case is a shift from selling tests toward owning more of the oncology decision workflow and research evidence stack. That can support strong value creation, but only if revenue mix and operating leverage improve fast enough to justify a platform valuation.
Thesis
Tempus can grow from a fast-growing precision-medicine company into a higher-value oncology operating layer if it keeps converting testing volume and governed
data rights into larger workflow, evidence, and research contracts faster than regulation and capital needs compress its multiple.
Last Economy Alignment (0.7/1.0)
Tempus benefits as cheap AI increases the value of hard-to-copy clinical data, trusted healthcare workflows, and verified decision support; the main
cap is regulation, not model scarcity.
Critique
If
Hub,
Lens, and care-pathway tools remain useful add-ons rather than must-have workflow gates, Tempus may keep growing revenue but still be valued like a good diagnostics lab instead of a durable healthcare operating layer.
Applied Digital Corporation (APLD)
ai
cloud
energy
hardware
Synopsis
Scarce powered campuses and long leases create a credible AI infrastructure platform. The equity outcome now depends on converting announced
megawatts into live rent without letting power and financing bottlenecks absorb the upside.
Thesis
Applied Digital has a real shot to compound equity by turning scarce, power-linked AI campuses into long-duration contracted rent, but the upside will be governed less by demand discovery than by financing depth,
energization, and whether management can recycle capital faster than it dilutes shareholders.
Last Economy Alignment (0.7/1.0)
It sells scarce powered capacity that becomes more valuable as AI demand rises; the main
cap is that power delivery, financeability, and
hyperscaler self-build can limit how much of that value common equity keeps.
Critique
If outsourced AI campuses prove mostly interchangeable financed shells, pricing will drift toward commodity
megawatt economics and
hyperscalers plus lower-cost-capital landlords will keep the durable value while Applied Digital keeps the build risk.
Oklo Inc. (OKLO)
nuclear
energy
ai
Synopsis
The setup is attractive because reliable clean power is becoming more valuable in an AI-heavy world and this company controls scarce nuclear pathways. The catch is that the equity case still depends on turning approvals, fuel, and customer intent into financed, operating
megawatts by 2031.
Thesis
Oklo is a scarce AI-power option on permissioned nuclear capacity: if it converts DOE progress, fuel access, and anchor-customer demand into one operating Idaho asset plus an initial Ohio phase, the market can still value it as an early clean-power platform rather than a speculative reactor concept.
Last Economy Alignment (0.7/1.0)
AI makes reliable, clean, permissioned power more valuable, and Oklo’s control points are siting, fuel access, and contracted capacity rather than software seats. The risk is execution through licensing and fuel, not software commoditization.
Critique
The scarcity story may already be capitalized: if financed, operating
megawatts are still limited by 2031, contracted-capacity pricing power will not matter and the stock can de-rate well before fleet economics are proven.
CoreWeave, Inc. (CRWV)
cloud
ai
software
enterprise
Synopsis
This is a real AI infrastructure growth story, but the key question is not demand. The investment case hinges on whether scarce contracted capacity can be turned into live revenue fast enough to outrun leverage and eventual multiple normalization.
Thesis
A rare pure-play AI infrastructure company with real demand visibility: if it keeps converting
contracted power,
backlog, and financing into live clusters while adding higher-value software and trust layers, revenue can scale several-fold by 2031 even as the
valuation multiple matures.
Last Economy Alignment (0.8/1.0)
CoreWeave sells one of the scarcest AI-era inputs: powered
GPU capacity. Its contract-backed financing and deployment flywheel fits the
Last Economy well, but the score stops short of elite because
hyperscalers and large customers can
self-build and returns remain sensitive to power delivery and capital costs.
Critique
The clean bear case is that CoreWeave becomes a levered AI utility: demand stays real, but pricing normalizes, big customers internalize more spend, and software attach never thickens enough to protect equity returns.