37c8c6dc19
Archive schema migration: 49 source files standardized with status + claims_extracted. schemas/source.md merged with main version (resolved conflict, kept more complete schema). Reviewed by Rio.
100 lines
6.2 KiB
Markdown
100 lines
6.2 KiB
Markdown
---
|
|
type: archive
|
|
source: "Michael Bloch (@michaelxbloch)"
|
|
url: https://michaelxbloch.substack.com/p/the-2028-global-intelligence-boom
|
|
date: 2026-02-22
|
|
tags: [rio, ai-macro, deflation, labor-displacement, scenario-analysis]
|
|
linked_set: ai-intelligence-crisis-divergence-feb2026
|
|
status: processed
|
|
claims_extracted:
|
|
- "AI labor displacement operates as a self-funding feedback loop (co-source, challenges)"
|
|
- "Technology-driven deflation is categorically different from demand-driven deflation because falling production costs expand purchasing power and unlock new demand while falling demand creates contraction spirals"
|
|
---
|
|
|
|
# THE 2028 GLOBAL INTELLIGENCE BOOM — Michael Bloch
|
|
|
|
Bull scenario counterpart to Citrini's crisis memo. Also written from June 2028 perspective. Argues technology-driven deflation expands purchasing power, and the same AI that destroys jobs creates replacements faster than any prior technology cycle.
|
|
|
|
## Core Thesis
|
|
|
|
AI is "the most powerful deflationary force in human history." Technology-driven deflation (costs fall because production costs collapsed) is categorically different from demand-driven deflation (costs fall because nobody's buying). The former has produced prosperity every time it's been tested over 200 years.
|
|
|
|
## Key Mechanisms
|
|
|
|
### Technology-Driven Deflation ≠ Demand-Driven Deflation
|
|
- When prices fall because cost of production collapsed → living standard boom
|
|
- Historical precedent: automobiles, televisions, air travel, computing, mobile phones
|
|
- Each time: deflation coincided with MORE economic activity because affordability unlocked new demand
|
|
- AI did this to the entire services economy simultaneously (70% of consumer spending)
|
|
|
|
### The Purchasing Power Reframe
|
|
- Bears focused on wages. What matters is purchasing power = wages AND prices
|
|
- Household earning $100K in 2025 only needs $85K in 2027 for same standard of living
|
|
- AI-driven services deflation running 8-12% annualized
|
|
- Average household spending $8-12K/year on services whose value proposition was navigating complexity (tax prep, insurance, financial advice, real estate commissions)
|
|
- AI agents compressed these costs 40-70% — equivalent to $4-7K annual raise, tax-free
|
|
- "The intelligence tax did" unwind — not the intelligence premium
|
|
|
|
### Intermediation Repricing (Not Collapse)
|
|
- DoorDash take rate collapsed → restaurants kept more, consumers paid less, drivers earned more per delivery
|
|
- Real estate commissions compressed from 2.5-3% to under 1% → $42B/year flowing to homebuyers instead of intermediaries
|
|
- Mastercard: per-transaction interchange compressed but total volume accelerated — people buy MORE things at better prices
|
|
- "The intermediation economy didn't collapse. It got competed down to its actual value and the surplus went to everyone else."
|
|
|
|
### Labor Market Recovery Through New Business Formation
|
|
- Unemployment peaked at 5.8% (Feb 2027) — genuinely concerning but short-lived (~9 months)
|
|
- Same AI tools that eliminated roles made it dramatically cheaper to START things
|
|
- Cost of launching a business fell 70-80% in 18 months
|
|
- Census Bureau: 7.2M new business applications in 2027, shattering 5.5M record from 2021
|
|
- "Minimum viable ambition" dropped to nearly zero — laptop + credit card + domain expertise
|
|
- "AI-assisted" prefix for every professional services category — substantive roles, not "prompt engineer" memes
|
|
- "AI didn't just destroy jobs faster; it created the replacement jobs faster too"
|
|
|
|
### SaaS Repricing as Feature
|
|
- Software spending is an INPUT, not output
|
|
- When cost of input drops, businesses deploy more toward expansion, R&D, new hires
|
|
- Long tail of SaaS (Monday, Asana, Zapier) decimated, but total economic activity INCREASED
|
|
- By Q3 2027, total enterprise tech spending recovered but composition unrecognizable
|
|
|
|
### Private Credit: Contained
|
|
- Zendesk default was real, but concentrated in narrow vintage (2021-23 LBOs) in specific sector (horizontal SaaS)
|
|
- Total exposure ~$80-100B against $2.5T private credit AUM = 3-4% loss rate
|
|
- Broader portfolio (real estate, infrastructure, asset-backed) performing fine or better due to AI productivity
|
|
- Insurance regulatory response proportionate — concentration limits, not forced deleveraging
|
|
- No forced selling mechanism → no contagion
|
|
|
|
### Mortgage Market: Held
|
|
- White-collar income disruption was transitional (9 months), not structural
|
|
- Household with 10% income drop but 20% non-housing expense drop is BETTER positioned for mortgage payments
|
|
- 30-day prime delinquency peaked at 2.1% (vs 5%+ for systemic distress)
|
|
- National home price index positive; only expensive coastal metros softened modestly
|
|
|
|
## Key Data Points (fictional, scenario-based)
|
|
- S&P 500: crossed 12,000; Nasdaq above 40,000
|
|
- Unemployment: peaked 5.8%, recovered by Q3 2027
|
|
- Real median household purchasing power: up 18% since 2025
|
|
- New business applications: 7.2M (2027 record)
|
|
- Services deflation: 8-12% annualized
|
|
- Consumer confidence: rebounded to pre-2020 levels by Q3 2027
|
|
|
|
## What Bears Got Right (per Bloch)
|
|
- Transition was painful
|
|
- SaaS was overvalued
|
|
- Intermediation businesses built on friction were in trouble
|
|
- PE-backed software was a ticking time bomb
|
|
- Labor market went through genuine disruption
|
|
|
|
## Where Bears Went Wrong (per Bloch)
|
|
- Assumed companies would uniformly fire rather than redeploy
|
|
- Assumed displaced workers would stay displaced rather than adapt
|
|
- Assumed reduced spending in one category = reduced spending overall
|
|
- Assumed deflation is always contractionary
|
|
- Treated economy as closed system where AI is zero-sum substitution
|
|
- "The deepest error was in treating the economy as a closed system"
|
|
|
|
## Connections to Knowledge Base
|
|
- Purchasing power reframe directly challenges Citrini's Ghost GDP mechanism
|
|
- New business formation thesis validates [[cryptos primary use case is capital formation not payments or store of value]] — but through traditional business, not tokens
|
|
- Deflation thesis supports [[internet finance generates 50 to 100 basis points of additional annual GDP growth]] — abundance creates more economic activity
|
|
- Intermediation repricing validates Belief #5 (legacy intermediation is rent-extraction) AND shows it can be bullish
|
|
- "Intelligence tax" framing connects to [[giving away the intelligence layer to capture value on capital flow]]
|