The Rise and Fall of Nectome [Podcast Transcript]

The Rise and Fall of Nectome was first published in April 27th 2018 as a podcast. It can be obtained here.

Hi all, and welcome to Physical Attraction. You heard the intro music; that means this show is going to be about technology or the future! This is something of a bonus episode while we gear up to finish the series that has taken up most of the show’s running time so far, as you know I’m liable to throw out there from time to time. As such, don’t expect anything too miraculous — but I can’t stop thinking about this story and the various ideas and issues that it raises; in some ways it seems like a perfectly surreal indictment of a lot of futurist thinking that may be misguided, and in some other ways, maybe this is something that we might actually have to think about fairly soon. Not just yet, though.

Here’s the short version of the story. Earlier this year, a startup promised to kill you for $10,000 — then people raised concerns about the business model and the validity of the technology — and now the University, MIT, that had been associated with this startup has moved to distance itself from them. These are the facts, as far as I can make out. So here’s the long version.

Y Combinator, for those that don’t know, is a fairly prominent “seed accelerator”. To me that still sounds like a device that flings apple pips at people at high velocity — which is probably one of the devices that’s actually marketed at Y Combinator — but no, instead it’s a prestigious venture capital organisation that funds new startups. And it’s prestigious for a reason — the Silicon Valley elite like to hobnob there, and in the past it’s funded startups like Dropbox, Airbnb, Coinbase, Reddit, Twitch, etc… companies with a combined worth now of $80bn. It’s interesting to look at the list of companies they invested in, say, ten years ago, to see which ones have succeeded. I think we all have this concept that all you need is some brilliant idea, filling some niche that no-one could have previously anticipated — a concept like Uber, for example. But of course there are some people who seem to have perfectly good ideas — or at least, perfectly similar to other companies that go on to be successful — but it’s not enough for them to succeed in the longterm.

I think we all have an image of Silicon Valley startup companies. Maybe it’s based on glowing press releases: sleek offices, ideas bursting from the brightest minds of the soon-to-be billionaires. Maybe it’s just clips from The Social Network, where angry techno-yuppies yell at each other about stock prices and so on. But, increasingly of late, my vision is of a bunch of dudes throwing darts at a wheel-of-fortune dartboard with Black Mirror episode titles on it. They really are stealing ideas straight from dystopian sci-fi.

“Oh, so we’re going to be creating digital versions of people scraped from their social media profiles? Sure. Robot bees? We’ll leave it to WalMart.”

Both of those were Black Mirror ideas and startups; eterni.me promises to scrape your social media data to create a “digital ghost” of you for your loved ones to remember you by… colour me really really sceptical about that idea… while WalMart does have a patent for robot bees like the ones out of Black Mirror episode Hated In The Nation.

How else to explain Nectome, the new company that’s taken the social media world by storm? For the low, low, fee of $10,000, they promise to kill you. With a tagline like “100% Fatal”, how can you refuse? There’s a waiting list already. They promise a refund if you change your mind — although, of course, you’ll have to do so before they remove it.

The idea was first pitched at a Y Combinator Demo Day, where eager start-up founders try to get funded by — or at least, get attention from — various business moguls and investors. The business model jumps headlong into transhumanism. “What if we told you we could back up your mind?” booms the website. [What if I told you that I had a bridge in Brooklyn to sell you?]

Anyway, the idea is that your brain is preserved by a process they call vitrifixation. According to Nectome’s website, “the powerful chemical fixative glutaraldehyde [will] rapidly solidify synapses and prevent decay… transforming the brain from a soft, watery consistency to that of soft rubber.” This is where the 100% fatal part comes in; the co-founder Robert McIntyre, who falls somewhere on the spectrum between futurist visionary and James Bond villain, views this as a form of euthanasia. “The user experience will be identical to physician-assisted suicide,” McIntyre said. Sounds pleasant. “Product-market fit is people believing that it works.”

After they’ve turned your brain to rubber, Nectome promises to freeze it. The idea is then that at some point at a later date, technology will advance enough that you can be revived — in this case, most likely by “scanning” and “uploading” your brain.

Cryopreservation, of course, has a long history — from a persistent urban legend about Walt Disney, who was never actually frozen, to the newer startup Alcor which has ‘preserved’ 156 people, with ten times as many queuing up to be frozen when they die. Hopefully Alcor manage to keep the freezers switched on and don’t go out of business.

You laugh, but this happened — while I was researching this topic, I stumbled upon an article about the history of cryopreservation via Alcor’s own website. The “Cryonics Society of New York” froze around half a dozen people in the 1960s and 1970s. A few of their “patients” thawed out when their surviving relatives decided they no longer wanted to make payments. Eventually, the early cryonics companies went bust and could no longer afford to maintain the storage facilities; so their “patients” thawed out entirely. At the time, the press were interested: “The stench near the crypt is disarming,” wrote one reporter, “strips away all defenses, spins the stomach into a thousand dizzying somersaults.” Pretty macabre — there was even talk of suing the guy who ran the show, but the whole question is very legally dicey.

Anyway, financial missteps aside, the hope is that one day future generations will revive those cryogenically preserved. One key concern has been preventing the formation of ice crystals in the brain; simply freezing it without vitrification beforehand will destroy cells and render any hopes of one day ‘retrieving’ the person impossible, even with technology that we can forsee being developed in the future.

Unlike previous cryonics attempts, Nectome inject the vitrifying substance (ethene glycol) while you’re still alive. The aim here is to preserve what the startup calls your connectome, and they claim that a similar process had good results with rabbits. This process won the “Brain Preservation Prize” and claims to have preserved each of the synapses in the rabbit’s brain, which were then imaged by an electron microscope. The road forward, presumably, is then to simulate the brain in totality. Once we have an unimaginably complex computer programme that can map out each of the synapses of the human brain — and manage to survive the ensuing singularity — you just need a means of scanning the vitrified brain, mapping each of its synaptic connections. Then you can upload that data into your brain emulator, and boom — you’re immortal. Sort of. Maybe.

Of course, like many transhumanist pitches, it runs headlong into that strange question that combines philosophy and medical science. What is consciousness? On one level, it seems apparent that if you perfectly replicate the structure of your brain and body — way down to the atomic level, including all of the synapses — the result is, well, you. Where else could the consciousness possibly be located?

But does the same hold true for just your brain? Or just the neural and synaptic connections? How detailed would a simulation, or physical reconstruction of the brain have to be to reproduce your consciousness? We struggle to define these questions, just as we struggle to develop a test or even a coherent definition for what consciousness is — and what it is that we’d seek to preserve. Anders Sandberg of the Future of Humanity Institute in Oxford has been one of the most vocal people about the philosophical problems that the concept of brain uploading entail. Can consciousness ever be embodied in an artificial intelligence? And, if it is possible to create a consciousness this way, is there a “continuity of consciousness?” That is to say, when they upload your brain into an emulator or artificial reconstruction and switch it on, is that the same thing as you — or just a copy of you, with all of your memories and experiences? An “imposter?”

Some hope to resolve this “identity problem” through a process of gradual transformation. We would become like the proverbial grandfather’s axe — piece by piece, we are replaced by immortal or synthetic alternatives, but the continuity remains. Replace the handle, replace the prefrontal cortex, merge the soft grey matter with digital enhancements, but it’s still my grandfather’s axe. Of course, no one knows if this would actually work — but it seems less possible with a setup like Nectome’s, where the idea is to scan a preserved brain.

In many ways, though, all of these questions are pretty premature — if they’ll ever be relevant. Many neuroscientists believe that “restoring a consciousness” from a preserved brain in this way will be decades away at best, if not completely impossible. Shelly Fan, my colleague at Singularity Hub, wrote an excellent article about the science behind the company. She notes that while the brain preservation technology used by Nectome is cutting-edge, and has won awards in the past, there are still significant technological challenges to overcome:

Part of the argument is technological. Despite billions of dollars and several large-scale brain-mapping programs, no one has (yet) been able to image an entire mammalian brain at the synapse level — not even that of a mouse.

We’re talking really big scale here: the human brain is jam-packed with millions of neurons, each connecting to thousands of others forming trillions of synapses. Today, even top-of-the-line imaging efforts are struggling with one cubic millimeter of a mouse brain.

So this is a huge barrier at the moment, but in principle, it’s surmountable. We have seen exponential developments in lots of previous technologies; it used to take millions of dollars and teams of researchers decades to sequence the human genome, and now it can be done swiftly for a few hundred dollars by various startups. It’s not impossible to imagine that, with a few decades of technological progress, we’ll be able use new techniques to image the brain. But creating a static image of the brain is not the only barrier to overcome to understand, preserve, and even revive consciousness. If this technology turns out to one day be feasible — and if it turns also out to be impossible to resolve this identity problem for a setup like Nectome’s — it adds to an underlying sense of exasperation with people who take these things seriously. Neuroscientists are amongst those exasperated; this is from MIT Technology Review:

“Fundamentally, the company is based on a proposition that is just false. It is something that just can’t happen,” says Sten Linnarsson of the Karolinska Institute in Sweden.

He adds that by collaborating with Nectome, MIT had lent credibility to the startup and increased the chance that “some people actually kill themselves to donate their brains.”

“It is so unethical — I can’t describe how unethical it is,” says Linnarsson. “That is just not something we do in medical research.”

Not all Silicon Valley types should be tarred with the same brush, but would you trust your afterlife to the neon village that came up with Juicero, the widely-mocked $120m startup that sold a $400 machine to squeeze packets of juice that could be done just as easily by hand? When we still can’t master and perfect voice over internet, are we ready for immortality? And what about ethics, is it just an English county? After all, isn’t it vanity to want to live forever? The field of cryonics in general has come under harsh criticism from prominent neuroscientists like Michael Hendricks. He writes in MIT Technology Review that the idea of preserving a connectome doesn’t work for one of the most well-studied of organisms — C. Elegans, a roundworm whose 302 neurons have been thoroughly mapped for decades. If it can’t be done for the most simple brain we know of, how can we possibly hope to expand that to humans?

“Yet even with the full connectome in hand, a static model of this network of connections lacks most of the information necessary to simulate the mind of the worm… It is this purposeful conflation of what is theoretically conceivable with what is ever practically possible that exploits people’s vulnerability,” writes Hendricks.

In other words, as Shelly Fan puts it, it’s really not clear that Nectome’s technology is going to be anywhere near sufficient to do what they suggest is possible. The Nectome website includes a TedTalk entitled “I am my connectome” — but it’s far from clear that, even if you could extract the entire human connectome from a person’s brain, you’d have any of the thoughts, memories or conscious processes that you’d seek to preserve.

“In the end, it’s unclear what needs to be preserved to retrieve “you” from the massive tangle of neural connections in your head. Are synapse structures enough? Do we need to capture memory-related proteins too? What about non-neuron cells called glia, which are involved in memory? Or is it more feasible to model synaptic strengths of a living brain inside a computer, essentially achieving “mind uploading” before death?

Nectome has a rebuttal: by 2020 they hope to extract a “high-level bit of memory” from a preserved mouse brain. So far, however, they’ve made no mention of attempting that feat in a preserved
human brain.”

There’s also the elaborate question of whether one single static frame of the human brain can really capture what you’d want to — can really capture consciousness. Here, the philosophical problem of what consciousness is, isn’t just philosophical: it’s biological. To say you can preserve it, you need to be sure that you have captured everything relating to it. When you look at a still photograph from your youth, you bring a lot of context to it — information about the day it was taken, the people concerned, how you were feeling at the time — which is more than what’s contained in a single snapshot. Similarly, even if you could instantaneously reconstruct your brain down to the most accurate detail, there’s no guarantee that a single thought or memory would be transferred. Shelly explains:

“The living brain, after all, is constantly in motion. Neuroscientists often capture a fleeting neural process by sticking electrodes into the brain to pick up electrical signals. Or they use glow-in-the-dark protein sensors to monitor neural activity.

In other words, like any other biological processes, a thought is dynamic. It’s impossible to reconstruct an entire human using his DNA letters alone — it’s the expression of his DNA, based on complex interactions between himself and his environment — that make him, him.

Similarly, it’s entirely possible that you can’t reconstruct a history of neural activity from a fixed brain. Hell, it’s likely that you can’t even isolate a single memory trace from a mapped connectome.”

And as far as the ethics of something like this is concerned for Nectome precisely, Hendricks was even more critical:

“Burdening future generations with our brain banks is just comically arrogant. Aren’t we leaving them with enough problems?” Hendricks was quoted as saying. “I hope future people are appalled that in the 21st century, the richest and most comfortable people in history spent their money and resources trying to live forever on the backs of their descendants. I mean, it’s a joke, right? They are cartoon bad guys.”

And, recently, it has turned out that MIT — whose graduate founded the project — have quietly agreed with the whole “cartoon bad guys” assessment. They’ve said that “after reviewing “the scientific premises underlying the company’s commercial plans, as well as certain public statements that the company has made,” MIT will “terminate the subcontract between MIT and Nectome in accordance with the terms of their agreement.”” The Nectome website, once so triumphalist, now carries a massive disclaimer at the top: “response to recent press”, which points out that they don’t yet know whether their technique could be used on humans, or if it could preserve all of the levels of detail necessary to actually “scan and reproduce” the human brain in any meaningful way. It also comments on the controversy:

“Feedback from neuroscientists and thoughtful discussion from medical ethicists must be incorporated. We believe that clinical human brain preservation has immense potential to benefit humanity, but only if it is developed in the light, with input from medical and neuroscience experts. We believe that rushing to apply vitrification today would be extremely irresponsible and hurt eventual adoption of a validated protocol.”

A far cry from the startup pitch which obviously leaves out the caveats. People will be glad that the $10,000 is refundable.

Nectome, and probably Alcor as well, almost certainly can’t offer the service they claim to provide. There are perfectly valid reasons to be interested in medically preserving bodies and brains to exquisite detail, as well as imaging brains and attempting to reconstruct them. After all, one of the reasons brain preservation is considered is to help understand neurological diseases like dementia; and lots of AI architecture draws at least conceptual information from the brain structure. But as far as bringing people back to life goes: I doubt anyone involved with either of these projects is ever going to be revived. Sure- if you’re a selfish billionaire, leaving aside a few million for a 1% chance of someday being brought back to life by a 22nd century Victor Frankenstein probably seems like a pretty decent wager, but it seems to be out of reach at the moment — maybe forever.

Will it be this way forever? If technology continues to march forwards, maybe not. And then we’ll have to address the questions that the startup raises. Unanswered — and perhaps unanswerable — ethical and philosophical questions, as many of the ideas beloved of the transhumanist movement raise. The idea of spending money so that future generations will create a precise replica of you to live in whatever world there is then seems a hopelessly sad and vain one. After all, the you that is reading these words is, in all likelihood, long dead at this point. Perhaps our philosophical perspectives and attitudes will shift as technology advances, and the lines between human and post-human begin to blur. Perhaps we romanticise mortality because we have to live with it, and because we’re afraid of death — in awe of death, which is elevated as a great unquestionable inevitability, even as we seek to delay it for as long as possible with medical science. It’s not inconceivable that, someday, the idea of being limited to a lifespan of the length we have today will seem to be the real tragedy. But — naïve, romanticised notion that it may be — it’s hard to shake the feeling that everything should, when it’s time, end.

Thank you for listening to this episode of Physical Attraction. I’d like to thank Shelly Fan for her excellent article on Nectome that gave a much deeper dive into the neuroscience than lots of other places; you can read her work over at Singularity Hub, where I also have a weekly article dealing with similar themes and topics. That’s our show for this week. You can follow us on Twitter @physicspod, we have a Facebook page if you’re still on that website, and you can always contact us with your comments, questions, and concerns via the contact form over at www.physicspodcast.com; I read everything and respond to everything that makes sense, and if you want to tell me what you’d like to hear from the show, I’d love to hear it. You can donate to the show via links on the website to help us cover hosting costs and in case you think it’s worth a few dollars, pounds, yen or euros for hours of free, independent content. If you don’t want to go through the hassle of doing that, though, please do tell as many people as you can about the show. Upcoming episodes include the apocalypses that didn’t happen, the ways we might hope to avoid them in the future, and where to go next.