Prediction Machines: The Simple Economics of Artificial Intelligence

by Ajay Agrawal

Each startup in our lab is predicated on exploiting the benefits of better prediction.

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.

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...

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.

Although economists often disagree with each other, we do so in the context of a common framework.

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.

Where others see transformational new innovation, we see a simple fall in price.

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.

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

Technological change makes things cheap that were once expensive. The cost of light fell so much that it changed our behavior from thinking about whether we should use it to not thinking for even a second before flipping on a light switch. Such significant price drops create opportunities to

do

things we’ve never done; it can make the impossible possible. So, economists are unsurprisingly obsessed with the implications of massive price ...

Tim Bresnahan, a Stanford economist and one of our mentors, pointed out that computers do arithmetic and nothing more. The advent and commercialization of computers made arithmetic cheap.5 When arithmetic became cheap, not only did we use more of it for traditional applications of arithmetic, but we also used the newly cheap arithmetic for applications that were not traditionally associated with arithmetic, like music.

That is precisely what happened. When, a century and a half later, the cost of arithmetic fell low enough, there were thousands of applications for arithmetic that most had never dreamed of. Arithmetic was such an important input into so many things that, when it became cheap, just as light had before, it changed the world. Reducing something to pure cost terms has a way of cutting through hype, although it does not help make the latest and greatest technology seem exciting. You’d never have seen Steve Jobs announce “a new adding machine,” even though that is all he ever did. By reducing the cost of something important, Jobs’s new adding machines were transformative.

Lovelace had the same thought, but quickly dismissed it. At least insofar as a computer was concerned, she wrote, it “had no pretensions to originate anything. It can do whatever we know how to order it to perform. It can follow analysis; but it has no power of anticipating any analytical relations or truths.”7 Despite all the hype and the baggage that comes with the notion of AI, what Alan Turing later called “Lady Lovelace’s Objection” still stands. Computers still cannot think, so thought isn’t about to become cheap. However, what will be cheap is something so prevalent that, like arithmetic, you are probably not even aware of how ubiquitous it is and how much a drop in its price could affect our lives and economy.

We emphasize that each of these methods is about prediction: using information you have to generate information you don’t have. We focus on helping you identify situations in which prediction will be valuable, and then on how to benefit as much as possible from that prediction.

When prediction is cheap, there will be more prediction and more complements to prediction. These two simple economic forces drive the new opportunities that prediction machines create. At low levels, a prediction machine can relieve humans of predictive tasks and so save on costs.

As the machine cranks up, prediction can change

and

improve decision-making quality. But at some point, a prediction machine may become so accurate and reliable that it changes how an organization does things. Some AIs will affect the economics of a business so dramatically that they will no longer be used to simply enhance productivi...

At some point, as it turns the knob, the AI’s prediction accuracy crosses a threshold, changing Amazon’s business model. The prediction becomes sufficiently accurate that it becomes more profitable for Amazon to ship you the goods that it predicts you will want rather than wait for you to order them.

readers may be surprised to learn that Amazon obtained a US patent for “anticipatory shipping” in 2013.10 Instead, the salient insight is that turning the prediction dial has a significant impact on strategy.

What does this mean for strategy? First, you must invest in gathering intelligence on how fast and how far the dial on the prediction machines will turn for your sector and applications. Second, you must invest in developing a thesis about the strategic options created from turning the dial.

We lay the foundation in part one and explain how machine learning makes prediction better. We move to why these new advances are different from the statistics you learned in school or that your analysts might already conduct. We then consider a key complement to prediction, data, especially the types of data required to make good predictions, and how to know whether you have it. Finally, we delve into when prediction machines perform better than humans

and when people and machines might work together for even better predictive accuracy.

We turn to strategy in part four. As we describe in our Amazon thought experiment, some AIs will have such a profound effect on the economics of a task that they will transform a business or industry. That’s when AI becomes the cornerstone of an organization’s strategy. AIs that have an impact on strategy shift the attention on AI from product managers and operations engineers to the C-suite. Sometimes, it’s hard to tell in advance when a tool will have such a powerful effect. For example, few people predicted, when they tried it for the first time, that the Google search tool would transform the media industry and become the basis of one of the

Popular discussion seems to focus on the risks AI poses to humanity, but people pay much less attention to the dangers AI poses to organizations. For instance, some prediction machines trained on human-generated data have already “learned” treacherous biases and stereotypes.

PREDICTION is the process of filling in missing information. Prediction takes information you have, often called “data,” and uses it to generate information you don’t have.