Skip to main content
September 2026
← All 28 pairs

Pair 2×4 · structured conjecture

Ownership concentration × Deployable task coverage

Coverage sets the stakes; ownership decides who collects — and the topology gets poured before the stakes arrive, cheaply and unopposed, while coverage still looks too unimpressive to fight over. Recursive leverage — AI improving AI, code, cyber, persuasion — is the one-way ratchet that pulls the whole grid toward concentration as coverage climbs.

The full 2×2. Click to enlarge.

The four scenarios

Two questions: who owns the compute, models, and distribution rails, and how much of the real world the AI can actually do end-to-end. The naive reading is 'concentration times power' — but coverage is what turns ownership into leverage. A broadly-owned narrow tool is nearly inert; a feudally-owned broad-dependable one is the maximal-leverage case.

What do these axes mean? ▸

Each axis is a spectrum. The card takes the two poles of each and reads off what the corner where they meet produces downstream.

Axis 2 · Ownership concentration

Low
Broad stakeholding rails and rents widely owned
High
Feudal capture a few own the rails and rents

Who owns the underlying rails — compute, models, distribution — and captures the rents they generate.

Axis 4 · Deployable task coverage

Low
Narrow / brittle patchy or pilot-scale
High
Broad / dependable whole workflows, production-grade

How much real work the system can do reliably — a few narrow pockets, or most cognitive-professional tasks end-to-end.

Broad Stakeholding · Broad/Dependable

Everyone Owns the Engine; Who Owns the Steering?

The abundance dream's catch: the binding constraint stops being capital or capability and becomes coordination — deciding what to point the engine at. That decision layer can quietly re-grow a feudal structure one level up even while the compute stays commons-owned, and status competition intensifies exactly as material scarcity relaxes. It isn't a resting state, either — recursive leverage keeps pulling ownership rightward, so holding it broad is a permanent, active tax, not a default.

Feudal Capture · Broad/Dependable

The Gilded Bypass

The deepest basin on the map: owners of a rail that can do nearly everything need the population less as workers, customers, or voters and more as dependents to pacify — the social contract loses its mechanical basis, and no strike or boycott bites the way it used to. Recursive coverage makes the moat self-sealing, since the same capability that out-improves challengers also manufactures consent. The twist: near-free capability could fund a lavish material floor as a rounding error — comfortable, gilded, and managed at once.

Broad Stakeholding · Narrow/Brittle

The Commons That Got Bored

The danger isn't capture, it's boredom: broadly-owned rails promised a windfall and delivered a utility, so the public coalition that funded the commons — cooperatives, open weights, sovereign compute — loses patience and sells out right before coverage broadens, handing the eventual upside to whoever stayed. What fills the gap is real, though: guilds regain status as the trusted interface to unreliable tools, and a whole audit-and-certification economy blooms around verifying what the AI can't be trusted to do alone.

Feudal Capture · Narrow/Brittle

The Toll Booth on a Small Bridge

The quiet one: a few owners control a tool that doesn't do much yet, so they extract rent from dependency and gatekeeping rather than capability — and it's in their interest to keep it narrow-but-essential, since brittleness itself becomes the moat (the worse the edge cases, the more customers need handholding, indemnity, proprietary eval suites). The real danger is timing: ownership lock-in hardens here, cheaply and unopposed, precisely because the stakes still look too small to fight over.

Read the vignette: The Toll Booth on a Small Bridge

How this whole reading could be wrong

What would undo the pair's thesis — not any single corner.

The lead mechanism — that ownership topology gets poured cheaply during the narrow-coverage phase and turns irreversible before anyone thinks the stakes justify fighting over it — is single-family (warm Claude) conjecture partially checked against one cross-family cold pass. That cold pass also complicated two of its central claims: it found Q1 far denser (institutional, not thin) than the warm read claimed, and found Q4's 'decoupled bypass' texture overstated, arguing owners still need the masses as dependents and courted patronage rather than bypassing them outright. If metered deployment turns out to be the norm — owners deliberately holding capability back from full deployment — the ramp's top-right basin may be far less self-sealing than the pair claims.

Related pairs

Other cards that share one of these variables.

More pairs with Ownership concentration

More pairs with Deployable task coverage