About The Durability Curve: Where value migrates when intelligence becomes cheap.

Who makes it

Where to start

You do not need to read everything. Pick the door that fits.

1. 01

   You are new here

   Start here by the problem you have: one thing to run on your own work, one piece to read first, and a ledger of checked claims on the same problem.

2. 02

   You would rather run something

   The 7 free browser tools run on your own stack, metric or decision.

3. 03

   You have a specific problem

   Each path below is a short run of pieces in the order that builds the answer.

• Path 01 · 5 pieces · 54 min

  Deciding whether to trust an agent

  You are about to put an autonomous system somewhere it can do damage. Five pieces, in the order that builds the check.

  Starts with How Reliable Is Your AI Agent?

• Path 02 · 5 pieces · 47 min

  Why your benchmark is lying to you

  Every number on your dashboard was produced by something. This is what each one is actually measuring.

  Starts with Ten Lines of Code Scored 100%. One Agent Broke Eight Benchmarks.

• Path 03 · 5 pieces · 47 min

  Building an agent that gets better

  The model is the part you cannot change. Five pieces on the parts you can.

  Starts with Same Model, Different Product: The Case for Harness Engineering

• Path 04 · 5 pieces · 58 min

  Where the money actually lands

  Everyone can see the buildout. Far fewer can say which layer of it keeps the return.

  Starts with How Long Until Your AI Edge Stops Paying?

• Path 05 · 5 pieces · 58 min

  What the machine is doing to you

  The capability arrived. Five pieces on what it quietly took on the way in.

  Starts with The Difficulty You're Escaping Was Making You

• Path 06 · 5 pieces · 36 min

  Learning the lens itself

  Three pieces that set out the terms this publication runs on, then two that put them to work.

  Starts with The Lens Lexicon

How to check it

Every analysis is written to be checked. The bar has three parts, and a piece that clears it shows the work on the page.

1. i · Declares

   It names which of the laws below, or which principle derived from them, it is testing, up front, in the byline.

2. ii · Falsifies

   It states the evidence that would prove it wrong.

3. iii · Compares

   It compares two real things with real numbers, so you can check the claim yourself.

Not every piece clears it. Where one does not, tell me.

The five laws

Five claims do most of the work here. Each of the five has been amended since it was first stated: Laws II, III and V when tests found cases they did not fit, Law I when its first wording (upward only) ran against its own best applications, and Law IV when physics and mathematics showed where it stops. Each law’s page carries what would prove it wrong and every amendment. They were first set out on 5 April 2026, in I Read 3,000 Papers Across 12 Fields.

1. 01

   Bottleneck Migration

   Value moves to the most stubborn adjacent layer. Usually that is upward, as the layer below commoditises. When the scarcity is physical or institutional (compute, power, fabs, regulatory access), it moves down instead.

   Binds on: Near · 1–5y Amended 21 Jun 2026: Bottleneck Migration
2. 02

   Difficulty Is Load-Bearing

   The hard part is what produces the value, when it gives you a second, independent way to check whether the answer is right. Effort that only repeats the same route can go. This covers knowing whether you are right; the effort that builds capacity, like training or practice, sits outside it and is untested.

   Binds on: Structural Amended 18 Aug 2026: Difficulty Is Load-Bearing
3. 03

   Architecture Outlives Content

   Where a system runs a process, the structure persists while the content turns over, so the structure is the moat. Where a system preserves something, it inverts: the content is what matters, and the structure can be rebuilt from it.

   Binds on: Long · 5–20y Amended 18 Aug 2026: Architecture Outlives Content
4. 04

   Instruments Over Theory

   Hidden structure stays hidden until you build the instrument that measures it. That holds wherever we build things first and understand them later. In fundamental physics and pure maths, theory usually leads.

   Binds on: Structural Amended 21 Jun 2026: Instruments Over Theory
5. 05

   The Targeting Problem

   Misaimed capability always wastes the effort, but it only harms the system when the miss is hard to spot or correct. When a target looks wrong, treat that as a hypothesis and test it before you re-aim.

   Binds on: Structural Amended 2 Sep 2026: The Targeting Problem

Questions