Mustafa Suleyman (with Michael Bhaskar) (Bodley Head)
David Runciman (Profile Books)
Igor Tulchinsky and Christopher E. Mason (MIT Press)
Advertisement
Edward Geist (Oxford University Press)
A SORCERER’S apprentice decides to use magic to help clean his master’s castle. The broom he enchants works well, dousing the floors with pails full of water. When the work is finished, the apprentice tries to stop the broom. He can’t, so he tries to smash it. The broom splits and regrows, working twice as hard as before, four times as hard, eight times as hard… until the rooms are awash and the apprentice all but drowns.
This is the basis of a poem by Johann Wolfgang von Goethe in 1797. I wonder if it sprang to mind as Mustafa Suleyman wrote The Coming Wave? Or perhaps the shade of J. Robert Oppenheimer, father of the atomic bomb, darkened Suleyman’s view of artificial intelligence and of his own role in its rise as co-founder of AI leader DeepMind and now as CEO of Inflection AI.
Having launched AI systems, Suleyman and his peers rightly tremble. At one point, he compares AI to an evolutionary burst like the Cambrian explosion 500 million years ago, the biggest ever eruption of speciation.
The Coming Wave is mostly about the destabilising effects of new technologies, how they create a wildly asymmetric world where, say, a single quantum computer might render the world’s encryption infrastructure redundant, or one keystroke might repurpose an AI mapping drugs to hunt for toxins.
Extreme futures beckon: there is subjection under a surveillance state, or radical community self-governance in a world where, says Suleyman,”an array of assistants… when asked to create a school, a hospital, or an army, can make it happen in a realistic timeframe”.
The predatory city states in this latter, neo-Renaissance future can seem attractive. Suleyman isn’t so sure. “Renaissance would be great,” he writes; “unceasing war with tomorrow’s military technology, not so much.”
A third possibility is infocalypse, where the information ecosystem that grounds knowledge, trust and social cohesion collapses (we will come back to this). But we need to focus on today, says Suleyman. He complains about meetings where he tried to raise questions about synthetic media and misinformation, or privacy, only to be asked about the singularity (a hypothetical time when advanced AI makes our civilisation obsolete) or machine consciousness.
Historian David Runciman makes an analogous point in The Handover: How we gave control of our lives to corporations, states and AIs, an impressive account of the limited liability company and the modern nation state. The rise of these “artificial agents” at the end of the 18th century was, Runciman argues, “the first Singularity”, when we tied individual fates to two distinct but compatible autonomous systems.
These “have a lot more in common with robots than we might think”, he writes. Our political systems are artificial and autonomous: fail to grasp this and we won’t understand what happens or what to do as they acquire new intelligences.
Long-lived, sustainable states with a healthy balance between political power and civil society won’t keel over under helpful AI, Runciman predicts. But as they embrace it, they will grow more automated and disconnected from human affairs. How will we ever escape this machine utopia?
Human freedom may still be a force to reckon with, say Igor Tulchinsky and Christopher E. Mason in The Age of Prediction: Algorithms, AI, and the shifting shadows of risk. They explore why the more predictable world ushered in by AI may not be safer. Humans evolved to take risks, and weird incentives emerge when predictability seems to increase and risk seems to decline.
Tulchinsky, who analyses data flows in financial markets, and Mason, a geneticist who maps dynamics across human and microbial genomes, make odd bedfellows. Mason welcomes any advance that makes medicine more reliable; Tulchinsky fears perfect prediction would render humans as docile and demoralised as cattle. The authors’ spirited dialogue illuminates their detailed survey of what predictive tech can do, from warfare to politics.
Let’s say they are right and individual free will survives governance by all-seeing machines. It doesn’t follow that human societies would survive their attentions. This is the unexpected sting in the tail in Edward Geist’s Deterrence Under Uncertainty: Artificial intelligence and nuclear warfare, a gripping analysis of AI’s role in such conflict.
Geist, steeped in the history and craft of deception as a specialist in defence policy and security, thinks even the smartest agent can be made self-destructively stupid by subterfuge. Fakery is so cheap and effective that Geist envisions a future where AI-driven “fog-of-war machines” create a world that favours neither side, but backs “those who seek to confound”.
In Geist’s hands, Suleyman’s “infocalypse” is a far cleaner and cheaper weapon than any bomb. Imagine future wars using mind games. In such a shifting world, people could be persuaded their adversary doesn’t want to hurt them. Rather than living in fear, they would see the adversary’s values are, and always have been, better than theirs. Depending on your politics and sensitivity to disinformation, you may feel this future is already upon us.
And, says Geist, at his most Machiavellian, “would it not be much more preferable for one’s adversaries to decide one had been right all along, and welcome one’s triumph?”
Simon Ings is a writer based in London
