Sparks Brain Preservation

At seventy-seven, Ray Kurzweil takes eighty supplements a day, has an artificial pancreas to manage his diabetes, and appears to be in great shape. As he waits for technology to tide humanity onward, he is bewildered that some people have other plans. He used to have lunch with the economist Daniel Kahneman, who won a Nobel for his work on how irrational factors cloud our decisions. Over lunch in late 2023, when Kahneman was eighty-nine but in decent health, they debated the wisdom of extending your life. “We’re at a point where problems can be reversed,” Kurzweil said. “Your kidneys are failing? Well, so what? It could be solved next week.” Kurzweil told me that Kahneman wasn’t persuaded. “He said, ‘Look at history! Billions of people have lived, lost most of their capabilities, and then died! Nobody’s escaped that!’ ” Three months later, Kahneman wrote to his family and friends that he had decided to die by assisted suicide: “I have believed since I was a teen-ager that the miseries and indignities of the last years of life are superfluous, and I am acting on that belief.” Kurzweil sighed, recalling the loss. “People really don’t want to be nonfunctional. But it’s death that’s the tragedy. Death is the loss of information, beauty, love—everything we know!”

For many biohackers, the ultimate goal is simply the preservation of their own consciousness. One constraint on retaining information forever is that our neurons mostly don’t replicate themselves; they just shrink, deteriorate, and die. But Kurzweil has a fix for that, too. Decades back, he predicted that we’d soon be able to scan and copy our connectome—the intricate web of a hundred billion neurons that constitutes the human brain.

At A360, a nanotechnologist named Michael Andregg declared that Kurzweil’s prophesy was all but realized. Andregg co-founded a startup called Eon Systems. His hypothesis is that, because consciousness arises from bioelectric signals, you’ll be able to turn your digital brain on in the cloud and experience a thoroughly satisfying life up there (or out there, or wherever), because A.I. will predict your neuron’s next signal: brain autocomplete. He showed an emulated fruit fly whose connectome had been copied to a Kinko’s level of fidelity. It skittered around onscreen, in a somewhat staccato fashion, and it knew enough to groom itself and to avoid bitter tastes (though how there could be bitter tastes in the digital realm was never explained).

Andregg said that the human connectome was only a million times more complex than a fruit fly’s, so all we have to do is model our brains’ structure and activity and, voilà, emulation. We could be scanned and wake up in a digital body by 2030. “This is a whole new body, a whole new brain—this is transcendence!”

No one can fully emulate a C. elegans worm yet, despite fourteen years of trying; the best program can’t even make the worm move backward. Yet Andregg told me that the difficulties he foresaw were chiefly psychological: “The first person to do the upload, it will be destructive. We pause you, lock down all your proteins, and slice your brain with a big deli slicer—but much finer, at the hundred-micron level.” Nonetheless, he said, “We have at least ten volunteers who want to be the first to do it. Well, the second.”

When I asked Diamandis about the plausibility of whole-brain emulation, he pointed out that Andregg was featured in the “Moonshots” part of his program. He was ambivalent about eventually uploading his own connectome: “Destroying myself to upload feels like suicide. And if I’m somehow still here after the upload, destroying myself because my peer in the cloud says, ‘I’m good, you can kill yourself’—I’m not sure I’m ready for that.” Uncharacteristically, he acknowledged having qualms about Kurzweil’s nanobots, too, particularly if they cross the blood-brain barrier. “I’d probably do nanotechnology,” he said. “But it begins to enter a reëngineering of the brain, and of our personas. You have to ask, ‘At what point do we stop being human?’”