---
type: archive
source: "John Loeber (@johnloeber)"
url: https://essays.johnloeber.com/p/32-contra-citrini7-repost
date: 2026-02-23
tags: [rio, ai-macro, labor-displacement, rebuttal, scenario-analysis]
linked_set: ai-intelligence-crisis-divergence-feb2026
---

# Contra Citrini7 — John Loeber

Rebuttal to Citrini's "2028 Global Intelligence Crisis." Originally published as X thread, republished on Substack. Argues the bear case underestimates institutional momentum, software demand elasticity, and re-industrialization capacity.

## Core Arguments

### 1. Institutional Momentum
- "Every time, existing institutions with momentum have proven themselves far more durable than onlookers thought"
- Real estate broker example: people have called for their end for 20 years, but regulatory capture and market inertia make them resilient
- The "iron rule": everything is always more complicated and takes much longer than you think, even if you already know about the iron rule
- Change will be more gradual, giving time to respond and adjust

### 2. Software Has Infinite Demand for Labor
- "Virtually all current software is garbage"
- Current SaaS products "fucking suck" — they're being repriced because AI enables competition, not because software demand is falling
- Even with a Software Singularity, demand for labor is "practically infinite"
- Every software product could scale up complexity and features by ~100x before saturating demand
- Jevons Paradox: efficiency gains increase total demand, not decrease it
- Software engineering isn't forever-resilient, but saturation will be a slow process

### 3. Re-Industrialization
- US has "virtually limitless capacity and need for re-industrialization"
- Physical infrastructure: batteries, motors, semiconductors, ammonia (China makes 90% of world supply)
- Employment megaprojects as political path of least resistance
- Subject to physical-world friction, not AI singularity speed
- "People will find it gratifying to see the fruits of their labor in the real world"

### 4. Path to Abundance
- Industrial megaprojects → independence → large-scale low-cost production → material abundance
- AI taking margins to zero makes consumer products equivalently cheap
- Different parts of the economy "take off" at varying speeds — virtually all slower than Citrini suggests
- Government showed during Covid it's willing to be proactive and aggressive with stimulus
- "The point is material prosperity for people in the course of their lives... not satisfying the accounting metrics or economic norms of the past"

## Key Tension with Citrini
- Agrees disruption is real, disagrees on speed and severity
- Loeber's framework: gradual displacement + institutional inertia + policy response = manageable transition
- Citrini's framework: self-funding feedback loop + no natural brake = unmanageable acceleration
- The mechanism disagreement is about whether AI displacement has a natural speed limit imposed by real-world friction

## Connections to Knowledge Base
- Jevons Paradox argument maps to [[internet finance generates 50 to 100 basis points of additional annual GDP growth]] — expanded access creates new demand
- Re-industrialization thesis is orthogonal to internet finance — physical economy absorbing displaced digital workers
- Institutional momentum argument challenges the speed assumptions in [[what matters in industry transitions is the slope not the trigger]]