An essay titled “The Dead Economy Theory” set out a warning about artificial intelligence that focused less on model capabilities than on the economic consequences of replacing paid workers. Its central argument was that companies could improve margins by automating labor while simultaneously weakening the consumer demand required to sustain those gains.
The writer began from two premises presented as features of the current AI market: leading developers and technology companies were committing hundreds of billions of dollars to infrastructure, and valuations implied expectations of a very large addressable market. In the essay’s interpretation, the global labor market was the only opportunity large enough to support those expectations. Claims that agents could perform the work of multiple analysts were therefore read as promises of labor substitution, not merely assistance.
The proposed mechanism unfolded in stages. A company adopting agents could cut payroll, lower costs and receive a favorable response from shareholders. Displaced employees would then reduce spending. Businesses serving those former workers could lose revenue and dismiss additional staff, extending the contraction. Eventually, the first company could discover that its customers had also depended on wages elsewhere in the economy.
That sequence was described as a prisoner’s dilemma. Each business would have an incentive to automate before competitors because it retained the immediate savings, while the loss of demand was spread across the wider market. Collectively, however, those decisions could reduce the total market available to every participant.
The essay addressed the common counterargument that earlier technologies created new forms of work. It cited MIT economist David Autor’s finding that roughly 60% of present-day jobs did not exist in 1940. But the writer emphasized the speed and hardship of past transitions, pointing to a 140-year decline in agricultural employment and Carl Benedikt Frey’s account of workers waiting 70 years for wages and employment to recover during industrialization.
A more rapid AI transition was presented as the crucial risk. The writer also invoked comments from Bharat Ramamurti, who previously served as the National Economic Council’s deputy director. Ramamurti said displacement might unfold in two years because model developers faced pressure to generate revenue from heavy investment. It also contrasted task-specific tools such as the power loom or spreadsheet with general-purpose systems aimed at cognitive work across industries.
This was a theoretical commentary, not a documented forecast. Its investment figures, historical comparisons and cited assessments were supplied through the author’s argument and were not independently corroborated in the packet. The essay’s contribution was a proposed feedback loop: automation might succeed for individual firms yet become destabilizing if wages, employment and purchasing power contract faster than new work emerges.
