The Age of Prediction: Algorithms, AI, and the Shifting Shadows of Risk     
Igor Tulchinsky and Christopher E. Mason
The MIT Press (August 2023)

A brilliant analysis of two abstractions that are increasingly shaping our lives

According to Igor Tulchinsky and Christopher E. Mason, “To better understand this new era [in which almost everything can be predicted and risk and uncertainty appear to be diminishing in almost all areas of life], this book is focused on two abstractions that are increasingly shaping our lives: prediction and risk. Prediction speaks of what is to come; risk, less visibly, calculates the probability that a model is wrong and, like a grim accountant, totals up the costs of that error. Where prediction serves as a light projecting into the likely future, risk is the shadow of what cannot be seen or predicted. Their relationship is often paradoxical, particularly as prediction changes age-old apprehensions about risk and reshapes the very type of society and world we live in.”

These are among the subjects and issues on which Tulchinsky and Mason focus:

o The potential benefits of new tools for prediction
o Their possible/probable impact on the status quo
o The possible/probable consequences of that impact
o The unpredictability of human nature

o As humans become more predictable, the extent to which they could lose their humanity
o The especially serious ramifications of steadily enhanced predictive capabilities
o Responding effectively to economic disruptions caused by AI
o To what extent will humans become machines? To what extent will machines become humans?

In or near the central business district in most major cities, there is a farmer’s market at which some merchants (at least prior to COVID-19) offer slices of fruit as samples of their wares. In that same spirit, I now offer a few excerpts that suggest the thrust and flavor of Tulchinsky and Mason’s analysis of major issues:

o “The fact that prediction and risk are inextricably linked explains why the optimism that often characterizes the Age of  Prediction is often shadowed by anxiety fear, and active opposition to what accurately predicting the future will entail. At its most fundamental level, the enhanced ability to predict  create change, which can be destabilizing and anxiety provoking.”  (Page 15)

o “All models have imperfections, and these imperfections have to be balanced and normalized as well. As the statistician George Box famously said, ‘All models are wrong, but some are useful…However, all these limitations are just part of the modeling. Through arbitrage, which can test, model, and ruthlessly compare — separating true from false, and fantasy from realistic prognostication — it is possible to account for such risks and errors. Indeed, this leads us to the question of risk: the trade-offs, paradoxes, and unanticipated feedback loops that may emerge if prediction actually works.” (38)

o “In the twentieth century, physics revealed a quantum world that resisted determinism, consisting of particles defined by a probabilistic framework (for example, a wave function). Specifically, the stronger a prediction on the exact location of a particle, the weaker the measure will be for the momentum. Yet these wave equations can be analyzed statistically and measured with even greater precision than Newtonian mechanics ever could provide. As [Ian] Hacking points out [in The Emergence of Probability, 1975] the resurgence of indeterminism –of chance — has produced a powerful paradox that could be the banner of modern science: ‘The more the indeterminism, the more the control.'” (44-45)

o “In his article ‘On the Genealogy of Moral Hazard,’ the University of Pennsylvania law professor Tom Baker notes that insurers in the Victorian age felt the need to convince the public that insurance was above reproach, ‘not a form of gambling; a handmaiden to crime, or above all, a presumptuous interference with Divine Providence itself,’ but rather a scientific product designed for those who deserved it. One way to do so was to embrace the virtues of the day by selling only to the ‘right people’ and thus avoiding the perils of moral hazard.” (93)

o “So much of the burgeoning digital architecture — the IoT, social media, mobile telephony — has become both a tool for policing and security and a vector for criminals. The algorithm itself has become a tool for good and evil., just like the other powerful technologies. from fire to atomic energy and gene editing. These technologies represent a potential evisceration of traditional views of and expectations for privacy. Balanced against this negative cost is the law enforcement benefit.” (113)

To their great credit, Igor Tulchinsky and Christopher Mason are so clear in their thinking and in how they express their thoughts that a non-scientist such as I can absorb and digest the material, thereby gaining a much better understanding of how — and why — predictive capabilities can be viewed as “good news” and/or “bad news,” depending upon one’s assumptions, biases, unknown unknowns, etc. Their book is a brilliant achievement. Bravo!

In his business classic, Leading Change (1995), James O’Toole suggests that the greatest resistance to change is the result of what he so aptly characterizes as “the ideology of comfort and the tyranny of custom.” I cannot recall a prior era that better exemplifies that than does the Age of Prediction. Today’s change agents must often have the feeling that they are trying to manage a bubble bath in a tub filled with tomcats.