Prediction Machines

The Simple Economics of Artificial Intelligence

Ajay Agrawal, Joshua Gans, Avi Goldfarb

• Publisher: Harvard Business Review Press
• Year: 2018
• Print ISBN:
  9780262514972
  9780226206844
• eBook ISBN:
  9781633695689
• Topic: Computer Science
• Subtopic: Business Strategy
• Index: Computer Science

A strategic guide to employing the most-talked-about--and potentially most disruptive--technology of the decade.

• The authors' simple model is easy to understand, and clarifying for busy executives who have heard so much about AI but haven't wrapped their heads around it
• Yet it's hard to put into practice and apply if you're not one of them, so there's good reason to buy the book still
• Despite many books on robots, the book is ahead of the curve in presenting something useful

Audience: This book is targeted at decision?makers who want to understand how artificial intelligence will affect their business: business leaders, especially those tasked with thinking through strategy and capital allocation decisions.

Announced first printing: 20,000
Laydown goal: 3,500

1

Introduction

Machine Intelligence

If the following scenario doesn’t already sound familiar, then it will soon. A kid is doing homework alone in another room. You hear a question, “What’s the capital of Delaware?” The parent starts thinking: Baltimore … too obvious … Wilmington … not a capital. But before the thought is complete, a machine called Alexa says the correct answer: “The capital of Delaware is Dover.” Alexa is Amazon’s artificial intelligence, or AI, that interprets natural language and provides answers to questions at lightning speed. Alexa has replaced the parent as the all-knowing source of information in the eyes of a child.

AI is everywhere. It’s in our phones, cars, shopping experiences, romantic matchmaking, hospitals, banks, and all over the media. No wonder corporate directors, CEOs, vice presidents, managers, team leaders, entrepreneurs, investors, coaches, and policy makers are anxiously racing to learn about AI: they all realize it is about to fundamentally change their businesses.

The three of us have observed the advances in AI from a distinctive vantage point. We are economists who built our careers studying the last great technology revolution: the internet. During years of research, we learned how to cut through the hype to focus on what technology means for decision makers.

We also built the Creative Destruction Lab (CDL), a seed-stage program that increases the probability of success for science-based startups. Initially, the CDL was open to all kinds of startups, but by 2015, many of the most exciting ventures were AI-enabled companies. As of September 2017, the CDL had, for the third year in a row, the greatest concentration of AI startups of any program on earth.

As a result, many leaders in the field regularly traveled to Toronto to participate in the CDL. For example, one of the primary inventors of the AI engine that powers Amazon’s Alexa, William Tunstall-Pedoe, flew to Toronto every eight weeks from Cambridge, England, to join us throughout the duration of the program. So did San Francisco–based Barney Pell, who previously led an eighty-five-person team at NASA that flew the first AI in deep space.

The CDL’s dominance in this domain resulted partly from our location in Toronto, where many of the core inventions—in a field called “machine learning”—that drove the recent interest in AI were seeded and nurtured. Experts who were previously based in the computer science department at the University of Toronto today head several of the world’s leading industrial AI teams, including those at Facebook, Apple, and Elon Musk’s Open AI.

Being so close to so many applications of AI forced us to focus on how this technology affects business strategy. As we’ll explain, AI is a prediction technology, predictions are inputs to decision making, and economics provides a perfect framework for understanding the trade-offs underlying any decision. So, by dint of luck and some design, we found ourselves at the right place at the right time to form a bridge between the technologist and the business practitioner. The result is this book.

Our first key insight is that the new wave of artificial intelligence does not actually bring us intelligence but instead a critical component of intelligence—prediction. What Alexa was doing when the child asked a question was taking the sounds it heard and predicting the words the child spoke and then predicting what information the words were looking for. Alexa doesn’t “know” the capital of Delaware. But Alexa is able to predict that, when people ask such a question, they are looking for a specific response: “Dover.”

Each startup in our lab is predicated on exploiting the benefits of better prediction. Deep Genomics improves the practice of medicine by predicting what will happen in a cell when DNA is altered. Chisel improves the practice of law by predicting which parts of a document to redact. Validere improves the efficiency of oil custody transfer by predicting the water content of incoming crude. These applications are a microcosm of what most businesses will be doing in the near future.

If you’re lost in the fog trying to figure out what AI means for you, then we can help you understand the implications of AI and navigate through the advances in this technology, even if you’ve never programmed a convolutional neural network or studied Bayesian statistics.

If you are a business leader, we provide you with an understanding of AI’s impact on management and decisions. If you are a student or recent graduate, we give you a framework for thinking about the evolution of jobs and the careers of the future. If you are a financial analyst or venture capitalist, we offer a structure around which you can develop your investment theses. If you are a policy maker, we give you guidelines for understanding how AI is likely to change society and how policy might shape those changes for the better.

Economics provides a well-established foundation for understanding uncertainty and what it means for decision making. As better prediction reduces uncertainty, we use economics to tell you what AI means for the decisions you make in the course of your business. This, in turn, provides insight into which AI tools are likely to deliver the highest return on investment for the work flows inside your business. This then leads to a framework for designing business strategies, such as how you might rethink the scale and scope of your business to exploit the new economic realities predicated on cheap prediction. Finally, we lay out the major trade-offs associated with AI on jobs, on the concentration of corporate power, on privacy, and on geopolitics.

What predictions are important for your business? How will further advances in AI change the predictions you rely on? How will your industry redesign jobs in response to advances in prediction technology just as industries reconfigured jobs with the rise of the personal computer and then of the internet? AI is new and still poorly understood, but the economics toolkit for evaluating the implications of a drop in the cost of prediction is rock solid; although the examples we use will surely become dated, the framework in this book will not. The insights will continue to apply as the technology improves and predictions become more accurate and complex.

Prediction Machines is not a recipe for success in the AI economy. Instead, we emphasize trade-offs. More data means less privacy. More speed means less accuracy. More autonomy means less control. We don’t prescribe the best strategy for your business. That’s your job. The best strategy for your company or career or country will depend on how you weigh each side of every trade-off. This book gives you a structure for identifying the key trade-offs and how to evaluate the pros and cons in order to reach the best decision for you. Of course, even with our framework in hand, you will find that things are changing rapidly. You will need to make decisions without full information, but doing so will often be better than inaction.

KEY POINTS

---

• The current wave of advances in artificial intelligence doesn’t actually bring us intelligence but instead a critical component of intelligence: prediction.
• Prediction is a central input into decision-making. Economics has a well-developed framework for understanding decision-making. The new and poorly understood implications of advances in prediction technology can be combined with the old and well-understood logic of decision theory from economics to deliver a series of insights to help navigate your organization’s approach to AI.
• There is often no single right answer to the question of which is the best AI strategy or the best set of AI tools, because AIs involve trade-offs: more speed, less accuracy; more autonomy, less control; more data, less privacy. We provide you with a method for identifying the trade-offs associated with each AI-related decision so that you can evaluate both sides of every trade in light of your organization’s mission and objectives and then make the decision that is best for you.

2

Cheap Changes Everything

Everyone has had or will soon have an AI moment. We are accustomed to a media saturated with stories of new technologies that will change our lives. While some of us are technophiles and celebrate the possibilities of the future, and others are technophobes who mourn the passing of the good ole days, almost all of us are so used to the constant drumbeat of technology news that we numbly recite that the only thing immune to change is change itself. Until we have our AI moment. Then we realize that this technology is different.

Some computer scientists experienced their AI moment in 2012 when a student team from the University of Toronto delivered such an impressive win in the visual object recognition competition ImageNet that the following year all top finalists used the then-novel “deep learning” approach to compete. Object recognition is more than just a game; it enables machines to “see.”

Some technology CEOs experienced their AI moment when they read the headline in January 2014 that Google had just paid more than $600 million to acquire UK-based DeepMind, even though the startup had generated negligible revenue relative to the purchase price but had demonstrated that its AI had learned—on its own, without being programmed—to play certain Atari video games with superhuman performance.

Some regular citizens experienced their AI moment later that year when renowned physicist Stephen Hawking emphatically explained, “[E]verything that civilisation has to offer is a product of human intelligence … [S]uccess in creating AI would be the biggest event in human history.”¹

Others experienced their AI moment the first time they took their hands off the wheel of a speeding Tesla, navigating traffic using Autopilot AI.

The Chinese government experienced its AI moment when it witnessed DeepMind’s AI, AlphaGo, beating Lee Se-dol, a South Korean master of the board game Go, and then later that year beating the world’s top-ranked player, Ke Jie of China. The New York Times described this game as China’s “Sputnik moment.”² Just as massive American investment in science followed the Soviet Union’s launch of Sputnik, China responded to this event with a national strategy to dominate the AI world by 2030 and a financial commitment to make that claim plausible.

Our own AI moment came in 2012 when a trickle and then a surge in the number of early-stage AI companies employing state-of-the-art machine-learning techniques applied to the CDL. The applications spanned industries—drug discovery, customer service, manufacturing, quality assurance, retail, medical devices. The technology was both powerful and general purpose, creating significant value across a wide range of applications. We set to work understanding what it meant in economics terms. We knew that AI would be subject to the same economics as any other technology.

The technology itself is, simply put, amazing. Early on, famed venture capitalist Steve Jurvetson quipped: “Just about any product that you experience in the next five years that seems like magic will almost certainly be built by these algorithms.”³ Jurvetson’s characterization of AI as “magical” resonated with the popular narrative of AI in films like 2001: A Space Odyssey, Star Wars, Blade Runner, and more recently Her, Transcendence, and Ex Machina. We understand and sympathize with Jurvetson’s characterization of AI applications as magical. As economists, our job is to take seemingly magical ideas and make them simple, clear, and practical.

Cutting through the Hype

Economists view the world differently than most people. We see everything through a framework governed by forces such as supply and demand, production and consumption, prices and costs. Although economists often disagree with each other, we do so in the context of a common framework. We argue about assumptions and interpretations but not about fundamental concepts, like the roles of scarcity and competition in setting prices. This approach to viewing the world gives us a unique vantage point. On the negative side, our viewpoint is dry and doesn’t make us popular at dinner parties. On the positive side, it provides a useful clarity for informing business decisions.

Let’s start with the basics—prices. When the price of something falls, we use more of it. That’s simple economics and is happening right now with AI. AI is everywhere—packed into your phone’s apps, optimizing your electricity grids, and replacing your stock portfolio managers. Soon it may be driving you around or flying packages to your house.

If economists are good at one thing, it is cutting through hype. Where others see transformational new innovation, we see a simple fall in price. But it is more than that. To understand how AI will affect your organization, you need to know precisely what price has changed and how that price change will cascade throughout the broader economy. Only then can you build a plan of action. Economic history has taught us that the impact of major innovations is often felt in the most unexpected places.

Consider the story of the commercial internet in 1995. While most of us were watching Seinfeld, Microsoft released Windows 95, its first multitasking operating system. That same year, the US government removed the final restrictions to carrying commercial traffic on the internet, and Netscape—the browser’s inventor—celebrated the first major initial public offering (IPO) of the commercial internet. This marked an inflection point when the internet transitioned from a technological curiosity to a commercial tidal wave that would wash over the economy.

Netscape’s IPO valued the company at more than $3 billion, even though it had not generated any significant profit. Venture capital investors valued startups in the millions even if they were, and this was a new term, “pre-revenue.” Freshly minted MBA graduates turned down lucrative traditional jobs to prospect on the web. As the effects of the internet began to spread across industries and up and down the value chain, technology advocates stopped referring to the internet as a new technology and began referring to it as the “New Economy.” The term caught on. The internet transcended the technology and permeated human activity at a fundamental level. Politicians, corporate executives, investors, entrepreneurs, and major news organizations started using the term. Everyone began referring to the New Economy.

Everyone, that is, except economists. We did not see a new economy or a new economics. To economists, this looked like the regular old economy. To be sure, some important changes had occurred. Goods and services could be distributed digitally. Communication was easy. And you could find information with the click of a search button. But you could do all of these things before. What had changed was that you could now do them cheaply. The rise of the internet was a drop in the cost of distribution, communication, and search. Reframing a technological advance as a shift from expensive to cheap or from scarce to abundant is invaluable for thinking about how it will affect your business. For instance, if you recall the first time you used Google, you may remember being mesmerized by the seemingly magical ability to access information. From the economist perspective, Google made search cheap. When search became cheap, companies that made money selling search through other means (e.g., the Yellow Pages, travel agents, classifieds) found themselves in a competitive crisis. At the same time, companies that relied on people finding them (for example, self-publishing authors, sellers of obscure collectibles, homegrown moviemakers) prospered.

This change in the relative costs of certain activities radically influenced some companies’ business models and even transformed some industries. However, economic laws did not change. We could still understand everything in terms of supply and demand and could set strategy, inform policy, and anticipate the future using off-the-shelf economic principles.

Cheap Means Everywhere

When the price of something fundamental drops drastically, the whole world can change. Consider light. Chances are you are reading this book under some kind of artificial light. Moreover, you probably never thought about whether using artificial light for reading was worth it. Light is so cheap that you use it with abandon. But, as the economist William Nordhaus meticulously explored, in the early 1800s it would have cost you four hundred times what you are paying now for the same amount of light.⁴ At that price, you would notice the cost and would think twice before using artificial light to read this book. The subsequent drop in the price of light lit up the world. Not only did it turn night into day, but it allowed us...

Table of contents

• Cover
• Title Page
• Copyright
• Dedication
• Contents
• Acknowledgments
• 1. Introduction: Machine Intelligence
• 2. Cheap Changes Everything
• Part One: Prediction
• Part Two: Decision Making
• Part Three: Tools
• Part Four: Strategy
• Part Five: Society
• Notes
• Index
• About the Authors