I WRITE this while Pope Francis is still alive, and I hope that that’s still true when you read it. But every paper and every broadcaster is checking over their obituaries, and has them now ready to run if his death is announced. Some people find this ghoulish: I greatly upset a Swedish thriller writer I knew by asking him for an appreciation of Henning Mankell after he had announced that he had terminal cancer; but I think of it as one of the few opportunities to report on a story without time pressure.
Of course, what everyone will really want to know is who the next Pope will be, and here, I think, The Economist has already beaten everyone with a story from last December which looks at the chances of a non-white Pope. It will, no doubt, be ruthlessly plundered by the daily press. If you see references to the last non-European Pope (Gregory III, a Syrian who died in 741), or to the outsider Korean candidate, Cardinal Lazarus You Heung-sik, you’ll know where they’ve been cribbed from. Me, I’m hoping for Cardinal Anders Arborelius, who’d be the first Swedish Pope, and I have already downloaded his performance on the Swedish equivalent of Desert Island Discs in preparation for my career as an authority.
COMPARED with these speculations, the things that have actually happened in the world are almost too painful to contemplate, since there is nothing we can do about them. Instead, I have been thinking about original sin, as it shadows the optimism of the progressive world. Professor John Naughton, who writes a tech column for The Observer, and a kind of public diary on Substack, drew my attention to a longish essay on the potential benefits of artificial intelligence (AI) by Dario Amodei (darioamodei.com), who really knows his stuff, having worked on AI at Google, then at Open AI, which he left to found Anthropic, an AI firm whose chatbot, Claude, I use myself.
By the standards of Silicon Valley, Amodei is a realist. He is concerned about some of the potential dangers of AI, and he does not expect the imminent transformation of the world or the departure of humanity’s elite to Mars. But he does imagine that, as early as next year, we might see an AI model with powers that include “taking actions on the internet, taking or giving directions to humans, ordering materials, directing experiments, watching videos, making videos, and so on. It does all of these tasks with, again, a skill exceeding that of the most capable humans in the world.”
This much reads like familiar science fiction. The important twist that he adds is that only the first one of these AI models will be prohibitively expensive to build and train. Once that training is accomplished, the model is just software, which can be infinitely duplicated at a trivial cost. He reckons that the hardware needed to run millions of these will be available in a couple of years, and then these millions can collaborate with each other, much as humans would.
This collaboration will solve, he believes, almost all human problems, largely by inventing new scientific instruments and techniques — and that is where I want to start screaming. His assumption that these immense powers will be wielded by benevolent people like him is nowhere seriously examined. He knows that there are bad people in the world, but is confident that they can be overcome. He hopes for a world “in which democracies lead on the world stage and have the economic and military strength to avoid being undermined, conquered, or sabotaged by autocracies, and may be able to parlay their AI superiority into a durable advantage. . .
“In this environment democratic governments can use their superior AI to win the information war: they can counter influence and propaganda operations by autocracies and may even be able to create a globally free information environment by providing channels of information and AI services in a way that autocracies lack the technical ability to block or monitor. It probably isn’t necessary to deliver propaganda, only to counter malicious attacks and unblock the free flow of information.”
If very smart, well-informed people can be quite so stupendously wrong about the world after Trump, it’s very difficult to be optimistic — but then he was writing from California, where the horizon is so promising that you need never look down and notice that the streets are paved with dying junkies.
