SYNTHESIS NOTE
Topics›this note

Does AI actually commodify expertise or tokenize it?

The standard framing treats AI output like mass-produced commodities, but does AI's contextual, mutable nature fit better with token economics than commodity theory?

Synthesis note · 2026-04-14

The reflex framing for what AI does to expertise is commodification — Marx's category for what happens when previously bespoke things get standardized, mass-produced, and sold as identical interchangeable units. On this framing, AI is the latest stage of an old process: standardization of cognitive labor.

The framing fails on the central feature. Commodities are objects: identical in form, fixed once produced, possessable, stockpileable. AI output has none of these properties. It varies per prompt, per audience, per context. It is not stored as units; it is generated on demand and disappears unless captured. It is not interchangeable with itself, because the same prompt produces different output across runs and across contexts. The category-fit between "commodity" and "AI output" is poor.

A better category is the token — borrowed from monetary theory and from Giddens' symbolic-tokens analysis. Tokens are mediums of exchange whose value is in circulation, not in possession. They are mutable in form because their function is conversion: they convert intent into something the receiver can use. Money tokenizes labor; AI tokenizes intelligence. The structural analogy is exact: a fluid, contextual, infinitely reproducible exchange medium that converts user intent into expert-seeming output, valued not by what it IS but by what it DOES for the receiver.

This reframe matters because the diagnostic prescriptions diverge. The commodity frame predicts standardization and price collapse. The token frame predicts inflationary dynamics, currency-validator emergence, and a shift in the locus of value from the artifact to the act of receiving. The published Hyperinflation post and the Knowledge Custodian series both rest on the token frame; making the frame explicit lets the rest of the series cohere theoretically.

The strongest counterargument: tokenization is just commodification at a finer grain. The reply is that no commodity has the property of being generated anew per use — that property breaks the object/stock model commodities require.

Inquiring lines that read this note 35

This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.

Why does polished presentation create unearned authority in AI outputs? What happens to knowledge when intelligence becomes tokenized like a commodity? How should designers communicate what AI systems truly are and can do? Why do token-level mechanisms matter for learning to reason? Does AI assistance promote real skill development or substitute for independent learning? How does AI adoption across firms reshape employment and inequality? When should work require human-AI partnership versus full automation? When do multi-agent systems provide sufficient quality returns on token investment?

Related concepts in this collection 3

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
15 direct connections · 95 in 2-hop network ·medium cluster Open in graph ↗

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Original note title

AI tokenizes intelligence rather than commodifying it — flows replace stocks contextual mutability replaces identical mass-production