This is our paradox: We know that uncertainty is the substrate of the universe, is the crucible of creativity, is the path to truth, and yet we spend our lives searching for certainty. The word auspicious derives from the Latin auspex — one who examines the flight of birds for omens and divinations of the future. Today, our birds are made of bits — wind-up larks of our own making we call algorithms — but they serve the same function: to let us believe that we can divine the future in order to respond to it; to sedate our anxieties of living by selling us the illusion of control over the unknowable.

All power structures are prisons we choose to live inside and all prediction technologies are power structures. Prophets have always been the true governors of empires, first counsels to kings, final ventriloquists of the future.

We can never break free from the uncertainties of living, for we ourselves are bright violations of probability, but we can break free from the prison of prediction in order to be more alive: The great mercy of life is that we can change, we can reroute, we can surprise ourselves — that our future is not the computational sum of our past and our destiny not a deterministic equation but an ongoing process constantly creating itself, the way a memory reconstitutes itself with each retelling. Every algorithm that tries to tell us what we will like or do next based on what we have liked or done previously is stealing our self-respect — the respect for the fundamental capacity for self-surprise that makes life worth living — and selling it back to us at the price of the product.

How to defend this vital aspect of our humanity, of life itself, from the merchants of prediction is what Carissa Véliz explores in Prophecy: Prediction, Power, and the Fight for the Future, from Ancient Oracles to AI (public library) — a book that, in an epoch when it is difficult to say anything original about AI, offers a revelatory look at the prison we have chosen to live inside and a key out.

She writes:

The computer scientists of today play the same role in power and prophecy as the oracles of the ancient world, the astrologers of the Middle Ages, and the social scientists of the nineteenth century, as we will see… Prophets are merchants of prediction… Artificial intelligence is the new Oracle of Delphi. The likes of Elon Musk, Sam Altman, and Eric Schmidt sharing drinks with presidents and prime ministers is reminiscent of Nostradamus advising Catherine de’ Medici, and Rasputin counseling Nicholas II.

[…]

Predictions are often power moves disguised as quests for knowledge. Even when predictions are sincere, they can create currents of political influence that change the world. They become power moves.

She offers a fascinating history of how such technologies of power became possible, became desirable, became the norm — a history that chronicles the shift from the spiritual to the statistical, from faith to blind trust:

With God gradually moving into a background position, the grip of fate loosened. Along with worldviews that could make sense of and contemplate the notion that chance might play an important role in the world, the invention of printing with movable type in the middle of the fifteenth century facilitated the development of abstract numerical thinking, and with it, scientific inquiry and bureaucratic organization.

[…]

Trust in numbers was partly born out of distrust in people.

It was not only the Church’s disapproval that led to a decline in traditional forms of divination; common people were tired of depending on experts to catch a glimpse of the future. Reading animal entrails or the flight of birds is not obvious, and astrology is technical. The opaqueness of these methods facilitated fabrication, errors, and corruption. Divinatory lots such as dice are ways of potentially communicating with the gods in an open way, visible and readable to all. They also invite asking questions about randomness and probability, which might have contributed to probabilistic thinking. But dice were unfit for purpose in the Enlightenment and beyond. Games of chance were associated with gambling and vice, and leaving decisions to pure chance was counter to rationality. Statistics came to the rescue.

After getting a kick start from astronomers, statistics developed mostly thanks to bureaucrats who were responding to challenges in public life. It was a practical endeavor, not a scientific one. Merchants, bankers, farmers, militaries, and tax officials needed to count in standardized ways. Extensive trade needs prices and measures, and therefore extensive quantification. As bureaucracies grew, standardization and public justification were necessary. There was also the question of how to hide power struggles to make them look as rational as possible. Numbers, standards, and statistics make politics and power invisible. That’s convenient to anyone wanting to justify what they do as free from values.

Behind our trust in numbers is a closely related philosophical idea: that values cannot be derived from facts.

Which, of course, they cannot. None of the things we most value — not mountains, not music, not love — can be reduced to a set a facts. Numbers and figures can contour for us what the world is — truly or falsely — but they can never shade in how to live in it, how to share it, how to reshape it into the kind of world we want to live in. Statistical probability will never tell us anything about the reach of possibility. Probability theory pioneer Marc Kac himself recognized that all creative acts — those triumphs of our humanity — spring from “a world with its own passions, elations and despairs, and it is here that… one may even hear the voice of God.”

Véliz counters the cultural myth we have been sold:

The world is not inherently statistical; we made it so by quantifying it, the better to control it.

Statistics invented society. Since the statistical regularities in phenomena like crime and suicide could not be traced to individual actions, these patterns became attributed to “society” itself. Society became a living entity, above and beyond the individuals constituting it. From 1830, these numerical regularities were interpreted as the most compelling evidence of society’s existence. Statistics invented normal people too.

[…]

Probabilistic thinking changed our views of reality (our metaphysics), how we think about what we know (our epistemology), what we count as a good argument and an explanation (our logics), and how we think of the good life (our ethics). It might have even changed the way we think about how we think.

It didn’t just change us, it killed something essential of our nature — the shimmering irregularity that makes each of us who we are, a singular constellation of atoms and story unmatched in all the history and future of the universe. She writes:

The murderers were philosophers, gamblers, mathematicians, sociologists, insurers, and, above all, bureaucrats. The murder weapons were quantification, data collection, and statistical tools like the normal curve… The motive was power, understood broadly: from astronomers wanting to predict the skies for the purposes of navigation and timekeeping, to insurers wanting to make a profit, to bureaucrats vying for authority and wanting to have more control over the population. We killed individuals with names, personalities, and idiosyncrasies. We killed valuing people for who they are, without attaching categories, numbers, or, worse, rankings to them. We killed our confidence in free will, a topic to which we will come back in chapter eight. We killed trust in people and replaced it with trust in numbers. We gained order, but it came at a price. What we lost is rarely subject to quantification. But that it cannot be quantified doesn’t mean that it didn’t exist or matter.

In a passage that calls to mind Ursula K. Le Guin’s eternal insistence that “words are events, they do things, change things,” she adds:

We do things with words, other than describe the world. Predictions are not mere assertions; they are acts in the world, and that is a further reason why they are not facts. Predictions are speech acts: assertions that do something.

[…]

When someone makes a prediction, depending on the context and their methods, as well as their will to power (what they want), they might be doing very different things: expressing their degree of confidence about something, expressing the chances of something happening in the world, expressing what they want the world to look like in the future, trying to manipulate their audience into buying what they sell, or a combination… What they are not doing, however, is uttering a true statement about the world.

[…]

Predictive societies tend toward oppression through surveillance and self-fulfilling prophecies because predictions are speech acts that are closer to commands than to truth claims.

Given our cognitive apparatus is wired for our expectations to shape our reality, the predictions we believe end up shaping our lives:

A prediction about people has reflexive qualities, acting like a magnet that bends reality toward itself. When the force of the magnet is strong enough, the prediction becomes the cause of its becoming true.

[…]

Predictions are commands disguised as descriptions… Predictions change the reality they purport to predict, and sometimes their accuracy is evidence not of knowledge or understanding but of despotism.

Our greatest frontier of resistance is to remember that we can be the makers of our own destiny with our choices, but our choices are driven by our beliefs, which are the easiest thing to manipulate and imprison in a power structure, and therefore the first thing that must break free. Véliz — who delves deeper into these questions in this excellent Where Shall We Meet interview — insists that the key is in our hands if we keep in mind — our gloriously fractal, wildly associative, defiantly illogical mind — a few truths about being human that the prophets of prediction try to obscure:

• What we deem inevitable often never comes to pass, while what we thought impossible awaits around the corner.
• Spaces of indeterminacy are where creativity, humor, and innovation flourish.
• Allowing defiance is the right way to treat human beings, who are ends in themselves, with their own idiosyncratic views and preferences. Every life is worth living, every life a work of art to be created much more than discovered.
• We shape our ideas and our ideas shape us

Reflecting on the tapestry of paradoxes threading through our yearning for control and our vulnerability to exploitation, she asks:

Unpredictability, arising from partial knowledge, social complexity and reflexivity, scientific and technological breakthroughs, flukes, and blindness, is never going away. Given this unsettling fact of life, how should we live?

The best answer I know is this Marie Howe poem.