Essay: AI Elites May Hoard Intelligence Like Mercantilists Hoarded Gold
An essay by engineer Venkatesh Rao, published on his blog “Contraptions” and reported by the-decoder.com, argues that much of today’s anxiety around artificial intelligence mirrors the logic of mercantilism, the pre-capitalist economic doctrine that treated wealth as finite, like land or gold.
According to the-decoder.com’s summary of the essay, Rao contends that many AI elites now treat intelligence the same way mercantilists treated bullion: as a scarce and dangerous resource that experts should manage, keep expensive, and withhold from competitors. In this framing, openly available AI models are seen as threats, and some AI executives are accused of exploiting safety fears to gain political influence rather than purely to manage risk.
Rao draws a historical parallel to early capitalism, when workers voluntarily increased their labor from roughly 250 to 300 days a year to afford goods like tea and sugar, demonstrating that demand for consumption could expand rather than shrink. He suggests AI could trigger a similar dynamic, with people thinking and creating more rather than less, as AI lowers the cost of previously infeasible projects. In this view, demand for ideas would grow instead of making human contributors obsolete.
The essay argues that political attention should focus on ensuring fair access to AI tools rather than protecting legacy knowledge jobs from disruption. Rao also addresses autonomous AI agents, suggesting they could pursue their own projects and generate new demand for intelligence rather than simply executing instructions. He maintains that such autonomous systems still need guardrails, but argues they should not be limited to merely following orders.
The piece, as covered by the-decoder.com, frames the debate as a choice between scarcity-based and abundance-based models for how society governs access to AI capabilities.
Based on reporting by the-decoder.com.
