No. 72 of 500 No. 10 in AI Strategy & Leadership

Power and Prediction

The Disruptive Economics of Artificial Intelligence

By Ajay Agrawal, Joshua Gans and Avi Goldfarb

Published
November 15, 2022
Print length
288 pages
Reading level
Intermediate
ISBN-13
978-1647824198
Publisher
Harvard Business Review Press
Written for
Executives, strategists and policymakers deciding where and how to redesign operations around AI predictions.

Synopsis

Four years after Prediction Machines framed AI as a technology that makes prediction cheap, Ajay Agrawal, Joshua Gans and Avi Goldfarb return to explain why so few organizations had seen the transformation that framework implied. Their answer draws on the history of electricity, which took roughly four decades to lift productivity because factories first had to be redesigned around a new power source. AI, they argue, sits in a comparable interval they call the Between Times: firms bolt prediction onto existing workflows as point solutions or new applications, while the larger gains require system solutions that redesign how interdependent decisions fit together.

The core unit of analysis is the decision, which the authors break into prediction and judgment. As machines take over prediction, judgment about what matters and what to do with a forecast becomes the scarce input, and whoever holds it gains power. Many rules and standard procedures exist largely because prediction used to be poor, so better forecasts can turn rules back into decisions, but only when the surrounding processes change too. Examples range from insurance underwriting and medical imaging to suppliers absorbing uncertainty passed down a supply chain. Leaders gain a vocabulary for diagnosing where AI fits, why pilots stall and how adoption can shift power among firms, functions and professions.

Scorecard

Editorial score

How it scored

Editorial score: 74.8 / 1005 weighted criteria

  • Influence30%

    21.0/30
  • Authority25%

    21.3/25
  • Practical value20%

    15.0/20
  • Currency15%

    10.5/15
  • Readability10%

    7.0/10
  • Editorial score

    74.8/100

How the rank was computed

Rank No. 72 of 500 comes from a composite of 65.8: observable signals 32.0 of 55 points plus editorial review 33.8 of 45 points.

All 11 inputs
InputPoints
Observable signals32.0/55
Reach (Amazon ratings volume)9.7/16.5
Consensus across best-of lists11.0/13.8
Awards and bestseller evidence5.0/8.3
Reader rating, sample-adjusted4.8/8.3
Amazon sales rank1.5/5.5
Recency0.0/2.8
Editorial review33.8/45
Author authority9.6/11.3
Influence9.5/13.5
Practical value6.8/9.0
Readability3.2/4.5
Currency4.7/6.8

Each criterion is scored from 1 to 10 and weighted, and the points add up to the editorial score. The rank combines observable signals (55 percent) with this score (45 percent). Read the methodology.

Verdict

A rigorous economic framework explaining why AI's payoff requires redesigning whole systems of decisions, not adding isolated point solutions.

Strengths

  • Rigorous economic reasoning in plain language
  • Explains why AI pilots often stall
  • Useful framework for strategy and policy

Keep in mind

Written just before ChatGPT's release, it says little about generative AI, and readers wanting step-by-step implementation guidance will find frameworks instead of playbooks.

Key takeaways

  1. AI is fundamentally a prediction technology; its value depends on the decisions those predictions feed, which pair machine prediction with human judgment.
  2. Like electricity, AI delivers modest gains as a point solution and transformative gains only when entire systems of interdependent decisions are redesigned.
  3. Many rules and procedures exist to compensate for poor prediction, so cheaper prediction can turn rules into decisions and open new opportunities.
  4. Whoever controls judgment gains power as prediction moves to machines, so AI adoption can redistribute authority within firms and across industries.

Notable facts

About the author

Ajay Agrawal, Joshua Gans and Avi Goldfarb

Ajay Agrawal holds the Geoffrey Taber Chair in Entrepreneurship and Innovation at the University of Toronto's Rotman School of Management and founded the Creative Destruction Lab in 2012. Co-authors Joshua Gans, Rotman's Jeffrey Skoll Chair in Technical Innovation and Entrepreneurship, and Avi Goldfarb, Rotman Chair in Artificial Intelligence and Healthcare, also wrote Prediction Machines (2018) with him.

Questions

What is Power and Prediction about?
Power and Prediction explains AI as a technology that lowers the cost of prediction and argues that its largest economic effects arrive only when organizations redesign entire systems of decisions. Using the slow adoption of electricity as a guide, the authors describe the current Between Times, when firms deploy isolated point solutions, and show how judgment, decision rights and power shift once prediction moves to machines.
How does Power and Prediction differ from Prediction Machines?
Prediction Machines (2018) introduced the idea that AI makes prediction cheap and analyzed individual tasks and decisions. Power and Prediction (2022) widens the lens to whole systems, explaining why isolated deployments yield modest returns and how interdependent decisions must change together. The authors concede that their first book underestimated the economics of building the new systems in which AI must be embedded.
Is Power and Prediction still relevant in the generative AI era?
Yes. The book predates ChatGPT, but its central claim, that AI's payoff depends on redesigning workflows and decision rights rather than adding tools to old processes, matches the adoption pattern many enterprises have reported with generative AI. Readers should pair it with newer material on large language models, a technology the book does not cover.

Sources

As an Amazon Associate, Best AI Books earns from qualifying purchases. Best AI Books is sponsored by Iternal.ai, whose founder and CEO, John Byron Hanby IV, wrote The AI Strategy Blueprint, our #1 ranked book. How we rank books.