In Prediction Machines: The Simple Economics of Artificial Intelligence, Ajay Agrawal, Joshua Gans, and Avi Goldfarb “emphasize trade-offs. More data means less privacy. More speed means less accuracy. More autonomy means less control…The best strategy for your company or career or country will depend on how you weigh each side of every trade-off.”

Of course, AI can help to ensure that the measurements are accurate when obtained even as the nature and extent of each trade-off are certain to change. That is especially true now, when the global marketplace is more volatile, more uncertain, more complex, and more ambiguous than at any prior time that I can recall.

Why do I think that this book is a “must read” for business leaders? Because the co-authors share and examine several dozen insights, any one of which is worth far far more than the cost of the book. Here is a representative selection of three, each discussed in depth within the co-authors’ narrative:

o  “The impact of small improvements in prediction accuracy can be deceptive. For example, an improvement from 85 percent to 90 percent accuracy seems more than twice as large as from 98 percent to 99.9 percent accuracy fall by one-third, whereas the latter (an increase of 5 percentage points compared to 2). However, the former improvement means mistakes fall by a factor of twenty. In some settings, mistakes falling by a factor of twenty is transformational.” (Page 30)

o “Prediction machines scale. The unit cost per prediction falls as the frequency increases. Human prediction does not scale the same way. However, humans have cognitive models of how the world works and thus can make predictions based on small amounts of data. Thus, we anticipate a rise in human prediction by exception whereby machines generate most predictions because they are predicated as routine, regular data, but when rare events occur the machine recognizes that it is not able to produce a prediction with confidence, and so calls for human assistance. The human provides prediction by exception.” (69)

o “AI will increase incentives to own data. Still, contracting out for data may be necessary when the predictions that the data provides are not strategically essential to your organization. In such cases, it may be best to purchase predictions directly rather than purchase data and then generate your own predictions.” (178)

If you are eager to understand “the simple economics of artificial Intelligence,” look no further.

* * *

Prediction Machines was published by Harvard Business Review Press (April 2018).

To learn about the co-authors, please click on the name of each: Ajay Agrawal, Joshua Gans, and Avi Goldfarb.