The Dead Economy Theory

2026-05-31 · source · source ↗ · raw capture · wiki page

Owen McGrann introduces the "dead economy theory" as a successor to the "dead internet theory" — while the internet has become "a performance staged by machines," the economy risks following suit: a system where productive capacity operates without needing human participation.

Part I: The Numbers Problem

AI companies' valuations (OpenAI at ~$830B, Anthropic at ~$380B) require an addressable market the size of the global labor market. The "gentler language ('copilot,' 'assistant,' 'augmentation') is marketing." The financial model requires "elimination of human cost centers at civilizational scale."

Part II: The Three Turns (The AI Layoff Trap)

  1. Company replaces workers → costs drop, margins expand, stock surges (e.g., Jack Dorsey cut ~half of Block's workforce; stock rose 25%)
  2. Replaced workers stop earning → they cut spending → businesses they patronized see revenue decline → those businesses also adopt AI
  3. The company discovers its customers were other companies' workers → "Revenue growth stalls" → the efficiency investment becomes self-destructive

Cites the Wharton paper "The AI Layoff Trap" (Falk & Tsoukalas), which describes a prisoners' dilemma where "every firm rationally automates beyond the socially optimal level."

Part III: Political Crisis

Democratic governance rests on a bargain — governors need labor, tax revenue, military service, consumer spending from the governed. "Remove labor from that equation and watch what happens." The author warns of "the erosion of democratic leverage" as value is generated by AI systems owned by a handful of corporations.

The public funded foundational AI research (DARPA, universities, national labs) but "private companies captured the reward."

Part IV: Solutions as "Ahistorical Bullshit"

UBI proposals, retraining programs, and the "leisure economy" are dismissed. Invokes Case & Deaton's "deaths of despair" research — the mechanism "isn't just poverty" but loss of purpose, status, and perceived future. Piketty argues UBI fails to address structural problems. David Shor's polling found a federal jobs guarantee has more popular support than UBI: "People don't want a check. They want work. They want purpose."

Anthropic's own research showed "active deskilling" — junior engineers using AI coding agents completed tasks no faster and "understood their work less."

Part V: Intellectual Poverty of Silicon Valley

Attacks the philosophical foundations of AI acceleration: Nietzschean Übermensch misreadings, Effective Altruism's naive utilitarianism ("utilitarianism reinvented by people who have apparently never encountered Bernard Williams"), Longtermism as "warmed-over Parfit without the rigor," and rationalist rediscoveries of Bayesian epistemology without knowledge of Kuhn, Lakatos, or Feyerabend.

Key contradiction: OpenAI published a white paper calling for a 32-hour workweek and a public wealth fund, while its president "helped fund a super PAC that spent over two million dollars on ads against" a candidate who proposed taxing AI. The company also removed a profit cap.

Part VI: Camus vs. Sartre

Frames the entire tension through the Camus-Sartre break — whether present people can be "acceptable casualties in the pursuit of a better future." Sartre and Marxists said yes. Camus said no. McGrann endorses the Camusian position: "The person who exists now… is the unit of account."

Part VII: The Closing Window

The interventions that could work are known — public ownership stakes, antitrust enforcement, taxing automated labor, wider capital distribution — but "all of them require functioning democratic institutions with the will to challenge the richest companies in human history."

The dead economy defined: "one where plenty happens and none of it requires you."

Key Sources Referenced