Sangeet Paul Choudary argues that technology disruptions are misread because analysts ask how AI improves the existing game, not what becomes scarce once the old bottleneck disappears — and traces three competing hypotheses for where scarcity migrates in knowledge work.
Technological disruptions are commonly misread because analysts focus on how a new technology improves the existing game rather than on what becomes newly scarce once the old bottleneck disappears; applying this lens to AI reveals three competing hypotheses about where scarcity in knowledge work migrates.
Hollywood read streaming as an access channel, but its real effect was removing the scarcity of the prime-time programming slot, which restructured the industry's economics, talent power, and creative range.
Cited as an actor whose career was transformed once streaming allowed platforms to make more bets and tolerate a longer tail of outcomes than broadcast scarcity permitted.
Cited as a series that became globally important without being designed as global television, illustrating streaming's removal of national-market scarcity.
Cited alongside Bobby Deol as an actor whose career was transformed by streaming's tolerance for a longer tail of outcomes.
Streaming platform cited as the example that removed prime-time programming-slot scarcity, enabling portfolio experimentation over editorial capital allocation.
Every generation of strategy is organized around managing what is scarce; Reshuffle argues AI is now removing scarcities around the production of usable knowledge work.
Knowledge work is too broad a category to analyze directly; the specific scarcity AI removes is the high cost of exploration that once forced organizations to ration which questions deserved investigation.
Three candidate hypotheses, each placing the new bottleneck at a different point in the knowledge-work-production system.
The three directions are not contradictory; none of framing, generation, or judgment is independently the new source of advantage. The scarce capability is the architecture that connects them.
Author of Platform Revolution and Reshuffle; Senior Fellow at UC Berkeley; writes on platform strategy and AI's impact on the knowledge economy.
Cited as an actor whose career was transformed once streaming allowed platforms to make more bets and tolerate a longer tail of outcomes than broadcast scarcity permitted.
Cited alongside Bobby Deol as an actor whose career was transformed by streaming's tolerance for a longer tail of outcomes.
Streaming platform cited as the example that removed prime-time programming-slot scarcity, enabling portfolio experimentation over editorial capital allocation.
Technological disruptions are misread because people ask how the new technology improves the existing game rather than asking what becomes scarce once the old bottleneck disappears; strategy is organized around scarcity, so a technology that changes what is scarce changes where power and value accumulate.
Hollywood read streaming merely as a new distribution channel for the same old product, missing that it removed the scarcity of the broadcast schedule itself, which changed the game.
Not the scarcity of content access, but the scarcity of the prime-time programming slot - the fixed grid of a handful of scarce broadcast slots each day.
It changed which projects were economically viable (narrower genres, serialized storytelling), which actors had bargaining power, how portfolios were managed (portfolio experimentation over ex ante editorial allocation), and how talent's power relative to platforms shifted.
Reshuffle: Who wins when AI restacks the knowledge economy argues that AI is removing scarcities around the production of usable knowledge work, and examines who gains advantage as the knowledge economy restructures.
The high cost of exploration - the scarce expert time that once forced organizations to ration which questions deserved investigation before they knew which would prove valuable.
Judgment becomes more valuable because selection becomes the bottleneck: if AI generates many plausible strategies, the strategist's value shifts to deciding which ones deserve belief and commitment.
Judgment becomes less important because experimentation replaces prediction: as exploration becomes cheap, organizations can postpone judgment, test many options, and let survival data do the selecting.
The real scarcity becomes framing: because abundance can expand the search space faster than the ability to navigate it, the scarce capability moves upstream to constructing the framework within which answers are generated.
None of framing, generation, or judgment is independently the new source of advantage; the scarce capability is the learning architecture that connects them - spanning possibility to evidence to commitment.
A deliberately narrower phrase than 'intelligence' - the article treats raw intelligence as a distraction unless it delivers usable knowledge work that organizations can actually act on.
Many companies treat AI as a productivity layer for the existing firm - the same mistake Hollywood made with streaming - while missing that the real game shifts elsewhere.
Because if AI can generate alternatives, it can also rank, criticize, run adversarial checks on, and learn from them - so a bet on permanent human judgment as the bottleneck may just be a temporary inefficiency arbitrage.
Frame the right search space, generate many possibilities within it, cheaply eliminate weak ones, test the promising ones against reality, determine where irreversible commitment is justified, and feed results back into the next round of search.
The economic condition of limited availability relative to demand; the article treats every era of strategy as organized around managing whatever resource is currently scarce.
Knowledge work that is actually actionable and deliverable, as distinct from raw 'intelligence' - the article's deliberately narrower unit of analysis for what AI makes abundant.
The system spanning possibility to evidence to commitment - framing a search space, generating options, eliminating weak ones, testing survivors, and committing selectively - identified as the true scarce capability.
A scarce broadcast television scheduling position that forced programmers to exercise judgment under severe capacity constraints before streaming removed the fixed grid.
The broadcast-era practice of deciding ex ante which few programs deserved scarce airtime, before a platform could observe actual audience behavior.
The streaming-era practice of commissioning a much larger content portfolio and learning from actual audience behavior rather than picking winners in advance.
Betting on a capability gap that is actually temporary rather than structural - the article compares organizations banking on permanent human judgment to prompt engineers in 2024.
An organization built to frame search spaces, generate possibilities, eliminate weak options cheaply, test survivors against reality, and commit selectively as a repeating cycle.
The set of possibilities considered worth generating and evaluating; the article argues abundance can expand this space faster than an organization's ability to navigate it.
The process by which the bottleneck that defines competitive advantage relocates to a different point in a system once a prior constraint is removed by new technology.
The capacity to define which problem is being solved, which variables belong in the model, and which outcomes are being optimized for - identified in Direction 3 as a candidate new bottleneck.
The expense of investigating a hypothesis or course of action; historically high enough that organizations had to ration which questions deserved investigation before knowing which would prove valuable.
Interactive graph visualization derived from the companion RDF. Click nodes to resolve, drag to explore. Graph data embedded from companion RDF at generation time.
Query the companion RDF data using SPARQL. Use the default endpoint or copy queries to your own SPARQL client.
Explore Knowledge Graph using SPARQL