Purpose & Review Justification

The gap this paper closes

Today’s AI strategy content splits in two:

• Too technical — lab blogs, papers, model cards. Useful, but readers without an ML background can’t translate them into capital allocation.
• Too consumer-facing — op-eds, hype takes, vendor marketing. Easy to read, useless for decisions.

Nothing is written for capital-rich, conceptually-light operators. Many of them already control the exact assets that will matter most in the next phase of AI — proprietary data, distribution, regulated-vertical incumbency, energy, real estate — and yet are watching from the sidelines because no one has handed them a usable map.

This paper exists to give them that map.

Why secondary research

• Primary sources already exist and are public: lab releases, earnings calls, capex disclosures, analyst reports, founder essays, regulatory texts.
• Original empirical work would take 6+ months. Synthesis can ship in 4–6 weeks.
• The contribution is the frame, not new data. The frame is what is missing in the market.

Honesty about stance

The paper is opinionated by construction. The working hypotheses — H1_L0-L7-ladder, H2_u-curve-of-value, H3_orthogonal-axes-under-priced — are starting scaffolding. The research may reshape them, and if it does, the paper says so. Pretending to neutrality would be dishonest given the kickoff intent.

What would make this paper fail

• If it reads as a defense of a pre-baked thesis rather than an inquiry.
• If it skips the bear case and skeptic voices.
• If readers come away with vibes instead of moves they can execute.