Skip to content

The Peter McCormack Show Health Science

The Man Who Says Humans Can Live Forever | Aubrey de Grey

Aubrey de Grey opens by correcting the premise: death cannot be abolished, but aging can, because aging is accumulated molecular damage in a machine and machines can be maintained. He is the only senior figure in the field willing to give a number, and he gives it early, a 50/50 chance of longevity escape velocity within 12 to 15 years, noting that twenty years ago he said 25. The biology runs deep, from the indigestible waste that blinds you to the sugar bonds that stiffen your arteries, alongside a specific falsifiable target in mice: beat calorie restriction by two to three times starting at middle age. Peter McCormack spends as much time on the consequences, pushing hard on population, risk, inheritance, religion and whether death is what gives life meaning, which earns the single word bollocks.

Published Aug 6, 2026 1:17:16 video 75 min read Added Aug 7, 2026 Open on YouTube →

At a glance

Aubrey de Grey, president and chief science officer of the LEV Foundation, sits down with Peter McCormack in San Francisco and spends the first ninety seconds correcting the premise of the first question. Death cannot be abolished, he says, because there are a lot of ways to die. Aging can be, because aging is accumulated molecular damage in a machine, and machines can be maintained. From that single distinction the whole hour and seventeen minutes unspools: why the body ages at all, why evolution tuned different species to age at different rates, the seven categories of damage he sorted the field into twenty years ago, what the repair therapies will physically look like in a clinic, and why he thinks they will be free rather than a billionaire's toy.

De Grey is the only senior figure in the field who will put a number on it, and he gives it early and unprompted: a 50/50 chance of reaching longevity escape velocity within 12 to 15 years. Twenty years ago he was saying 25 years, so the target has slipped, but not much. His colleagues refuse to give any timeframe at all, which he calls a complete footshooting exercise and a prisoner's dilemma: each expert protects the next grant application, and the collective silence tells the public and the treasury that none of this is real.

McCormack does not nod along. He spends as much of the interview on the social consequences as on the biology, and he keeps pushing on the parts that unsettle him: what a 300 year old person actually is, whether a 300 year old should be dating his daughter, whether risk aversion swallows a society that can no longer be repaired by time, whether death is the thing that gives life meaning. De Grey answers each one, sometimes with a rebuttal that lands, once with the single word "bollocks."

What follows rebuilds the conversation in order, with every argument, number, mechanism, joke, and detour kept in place, attributed to whoever said it.

decades late life

GERONTOLOGY slow the damage down by messing with metabolism (calorie restriction, mimetics) DAMAGE REPAIR remove or reverse the damage periodically, whatever caused it (de Grey's programme) GERIATRICS treat the disease once it has already arrived (almost all of medicine today) PREVENTATIVE MAINTENANCE "we age because cars age"; the car lasts as long as you keep servicing it

the gaps exist because evolution never needed to close them: in the wild you get eaten first

Figure 1. The framing de Grey builds the entire interview on. Metabolism is the machine running, and it inflicts damage on itself as a matter of physics rather than biology. That damage does nothing visible for decades, then produces pathology. Traditional gerontology tries to slow the left arrow by tinkering with metabolism, which he argues hits a hard ceiling. Traditional geriatrics fights the right arrow after the disease has arrived, which he argues is a losing battle. His programme sits in the middle: leave metabolism alone, let the damage happen, and periodically take it out, the way you service a car.

"Is it inevitable that we will cure death?" No, and stop asking it that way

McCormack opens with the obvious question and gets it thrown back at him within a sentence. Is it inevitable that we will cure death? No. And then the correction, delivered as one of the most frustrating things about his interactions with the media: they constantly conflate the word aging and the word death.

The distinction is not pedantry, and de Grey spends real time on why. Death quite clearly cannot be completely eliminated by technology, because there are quite a lot of ways to die. Aging is something that absolutely can be eliminated by medicine in the future. When McCormack says that is kind of what he meant, de Grey cuts in: "that's what you meant, but don't say that."

Why it matters, in his account, is that there is a terror in society of thinking rationally about aging. Aging will be overcome by medicine in the fullness of time, but nobody knows how much time, and people find that unbearable to sit with. So they keep their emotional distance from the question. McCormack finishes the thought for him: people want to know whether it gets solved in their lifetime, nobody can tell them, and that is not good enough. So, de Grey says, people make up fantasies that let them put it out of their minds and get on with their miserably short lives.

He sorts the fantasies into two families. The desirability fantasy pretends aging is some kind of blessing in disguise. The immutability fantasy holds that aging is somehow not like disease at all, in some profound way that puts it permanently off limits to medicine. The media, he says, like to pander to what sells, so they perpetuate both, and they do it in subliminal ways. Using the word death when you mean aging is one of the worst of them. That is why he picked McCormack up on it at once.

Reset. Start with aging. Is aging something we will inevitably have control over? Yes. At the moment we have basically almost no control over it, and can slow it only very slightly. But we will without doubt come to the point where medicine is good enough that we can keep people biologically young, however long ago they were born.

That phrase, however long ago they were born, recurs throughout the interview and is the load bearing one. The goal is not a longer countdown. It is decoupling how you function from the date on your birth certificate.

The 12 to 15 year prediction, and why he is the only one giving numbers

"But we don't know when," McCormack says. That is right. Not maybe in your lifetime. And here de Grey volunteers the thing his colleagues will not.

One thing that results from the irrationality of society about aging is that most experts in the field feel unable to give even probabilistic timeframe predictions for how soon this happens. He calls that a huge problem, and diagnoses the cause bluntly: they feel they are in danger of being accused of saying irresponsible things, and it will hurt their next grant application. McCormack adds the commercial version, that investors want to know too, though de Grey notes a company can sidestep the problem by working on something that is a bit of aging and calling it a disease.

The consequence is a doom loop. If no expert will give any kind of timeframe, even a probabilistic one, then the rest of the world concludes that however optimistic people sound and however much noise they make about this or that breakthrough, it does not really mean anything and we are never going to get there. So it does not feel worth funding with taxpayers' money. He calls the silence a complete footshooting exercise, and a prisoner's dilemma, because each individual expert is rationally protecting their own funding while the collective outcome starves the field.

He has never had that problem. He has always given timeframes, and he restates the number:

a 50/50 chance of getting to a point that is for practical purposes having defeated aging, something that I call longevity escape velocity, within the next 12 to 15 years.

He then does something interviewers rarely get: he audits his own track record out loud. Twenty years ago he was saying 25 years. So the estimate has slipped a bit, but not much. Not a retraction, not a dodge, just the number moving less than the calendar did.

McCormack asks whether he would rather talk about the science or the social impact. De Grey talks about both all the time, so ask what you like. McCormack says he thinks a lot about the social impact, but they should probably do the science first, and then makes the admission that shapes the next twenty minutes: he does not understand the science of aging at all. He does not know why we age, why his beard has gone gray, why things are starting to not work anymore. Where is a good place to start?

The body is a machine, and machines wear out because of physics

De Grey's entry point is a fact everyone already knows and routinely glosses over: the body is a machine. A really, really, really complicated machine, one whose blueprints we do not have, but a machine, which means its function is determined by its structure, by what it is made of.

We have a pretty thorough understanding of what the body is made of. Cells, and stuff between cells. We know what cells are made of, DNA and proteins and so on. What we understand poorly is how it all goes together.

Then the pivot that does the work. Any machine with moving parts does itself damage as a consequence of its normal operation. This is not a fact of biology, it is a fact of physics. So the aging of a living organism is pretty much no different from the aging of a car or an airplane. Once you accept that, "why do we age" stops being a question. We age because cars age. It is entropy.

So why is that not obvious to people? Because a large part of the complexity of the body consists of a very sophisticated arsenal of automatic, built in damage repair machinery. Self maintenance machinery, if you like. That machinery is not 100 percent comprehensive, because if it were we would not age at all, and that would require perpetual motion. So there are gaps in what we have built into us.

And because the machinery covers most of it, we see no real consequences of the gaps until late in life. Which produces the central confusion: it looks like the body was repairing itself perfectly for a long time and then stopped. De Grey calls that nonsense, with a line that reframes the whole thing:

who blows the whistle to say, okay, now start aging please? That doesn't happen.

We have been aging throughout our lives, starting before we were born. There is simply no macroscopic consequence of it until middle age or later.

Why evolution built the gaps in

McCormack asks the deeper question. What is the biological reason for aging? Do we know it evolutionarily?

De Grey says yes, and that the answer has been settled for a long time. Since the body is a machine, and not aging at all would be perpetual motion, all organisms are going to age. All multicellular organisms with a fixed body size, anyway, which lets him set aside plants, which age in a very different way. They all age, but they age at different rates, because they have different degrees of comprehensiveness in that self maintenance machinery.

So the real question is not why evolution created aging. The real question is why evolution tried harder with some species than with others. And that we understand very well indeed.

Evolution happens in the wild, where there are many causes of death that have nothing to do with how long ago you were born: predation, starvation, hypothermia. There is no point in carrying machinery that makes you age as slowly as a human if you are low down the food chain and every individual in your species will be gone by the age of five anyway. So natural selection gravitates to a setting where there is some aging inside the time window that a typical member of the species actually reaches despite predation and starvation, but not much. Some minority of a species ages before it gets eaten. Never a majority, never zero.

"Huh. Never knew that. That's wild," McCormack says. De Grey notes this was all worked out in the 1950s and 1960s, a long, long time ago, which is his quiet way of saying none of this is fringe. The names he does not bother to say out loud are Peter Medawar and George C. Williams, whose work on the evolution of aging is exactly the mid century consensus he is pointing at.

Can the body just pause aging? Calorie restriction as the natural experiment

McCormack pushes on it from an angle worth keeping: a woman can create new life inside her, brand new fresh cells that start the aging clock over. Is there any biological reason our system could not just pause aging?

De Grey says some species do have systems for manipulating the rate of aging, essentially for trying harder in some circumstances than in others, activating more anti aging machinery. But it comes at a cost, most obviously in rate of growth.

His example is calorie restriction, discovered maybe a century ago: feed mice or rats less than they would like and they live longer than if you feed them as much as they want. Why would that be? Again, evolution explains it cleanly. In a famine, having offspring is pointless, because they will starve before they are old enough to reproduce. There is no selective advantage. Better to hunker down and outlive the famine so you can breed when there is food again. Conversely, if you are a mouse or a rat with plenty of food around, the best move is to eat as much as you can, grow as fast as you can, and reproduce as fast as you can before something eats you.

So it is a resource allocation trade off. Depending on environmental circumstances, there are different priorities for how much of your energy goes to self maintenance versus growth and reproduction. That trade off, tuned by mutation and selection over generations, is what sets a species' characteristic pace of aging. Genes in a mouse simply do not have the same level of sophistication in that part of the genome as a human's do, which is why mice age faster.

Does the body age differently in different parts? Yes and no. There are plenty of differences between organs in how regularly they regenerate. But everything is talking to everything else. There is a great deal of cross talk between organs and tissues with regard to the rate of accumulation of molecular and cellular damage and the consequent loss of function. That cross talk matters later, when he explains why repairing only a subset of the damage still moves the needle in a mouse.

How a theorem proving programmer ended up here

"I've got to ask, how did you get into this?"

De Grey was originally a computer scientist. He started programming at 15, found he was good at it, and had already decided by that age that he wanted to change the world, to do work of major humanitarian benefit. His reasoning at 15 was that since he was a good programmer, he would work on artificial intelligence, because that was the way to solve one of the biggest problems for humanity: the problem of work. The fact that we spend so much of our time doing things we would not do unless we were paid for them.

That went fine. Undergraduate degree in computer science in the early 1980s, then roughly seven years of what he calls pretty effective, pretty successful research in AI. He is careful to note that this was AI as nothing recognizably like what it is today, a completely different way of doing things. He was writing programs that would prove mathematical theorems.

Then the accident. During that period he met and married a biologist a lot older than him, a senior professor at the University of California San Diego who was in England on sabbatical. Over the next couple of years, into the early 1990s, he learned a lot of biology by accident over the dinner table.

And gradually it dawned on him that they were never talking about aging.

He had always known, since his earliest childhood, that aging was by far the world's biggest problem, considerably bigger than the problem of work. It kills the most people. It causes the most suffering. He had completely assumed everyone else thought the same and that biologists were working on it. You did not hear much, but it was a hard problem, so there you go.

The conversations that broke that assumption are worth quoting nearly whole, because they explain the last thirty years of his career:

I would say, you don't seem to be interested in aging. And she would say, no. And I would say, why not? And she would say, well, it's just decay, isn't it? What fundamental truths about the universe are you going to understand by studying decay? And I would say, well, but yes, sure, but it's bad for you. And she would say, well, that's not my problem. She would actually say that. And I would say, well, it kind of is. And that would be as far as we would get.

He quickly found that other biologists he met through his wife held the same view. He describes it as a complete bombshell, and says it took him a couple of years to come to terms with it to the extent of realizing he had to switch fields. That happened around 1994.

The logistics of the switch are the unglamorous part he tells anyway. He happened to be in a convenient job situation: a bioinformatics post at the University of Cambridge that was very undemanding. He had taken that job specifically because it left him spare time to do his AI research, having run out of funding for it. So he simply repurposed the spare time toward aging, and earned enough to pay his own way to conferences, since nobody was inviting him back then. "I started having ideas that were well received, and here we are."

"Tell me what day is the day you want to die on"

McCormack notes it is probably a good time to raise money on this, since a lot of rich people out there do not want to die. Then he tells the story that shapes the middle third of the interview.

On the drive up he called his wife. She asked who he was interviewing. He explained it was about longevity, about defeating aging, potentially in their lifetime. She said: yeah, but I don't want to live forever. And he said: well, tell me what day is the day you want to die on.

She was like, yeah, I don't have an answer for that.

That one exchange spawns the questions he stacks up for de Grey. What does it mean to be a 300 year old person? Are you much wiser, or are you traumatized? What does an accelerating growth curve of people do to a society? Do we get to the point where there are too many? And, the one he leads with, if you could effectively live forever, how does that change your risk profile in life, because an accident becomes really catastrophic.

De Grey takes the last one first.

Risk, Colbert, and the grandmother who helps you across the street

In 2007 de Grey published a general audience book, Ending Aging. When you write a book you end up on television, and in particular Stephen Colbert had him on. Colbert asked him this exact question: if we are all going to live to a thousand, won't that kill people's ability to take risks? You are not going to want to cross the street, because you cannot cure being hit by a truck.

De Grey's answer, then and now:

It'll be fine. Your grandmother will be able to help you across the street, because she'll still be healthy.

He adds, with visible pride, that he thinks to this day he is the only scientist who has managed to get Colbert out of character and make him spontaneously burst out laughing on air.

McCormack presses the serious version. He is not going to stop getting on planes, but perhaps as a society death becomes even more traumatic. De Grey concedes the direction: yes, we will probably be more risk averse. But there are two ways to lower your risk of anything. One is not to do the risky thing. The other is to use technology to make the thing less risky. And that is exactly what we are already doing. We are building self driving cars that will greatly reduce the risk of the single largest category of accidental death.

McCormack had taken his first Waymo ride the day before. Weird experience, he says. Good driver, says de Grey. Very safe.

That exchange gives de Grey his general template for the objections: the rebuttals to concerns about a post aging world usually come down to the fact that there are obvious technological solutions arriving on the same timeline.

Population: eight billion people and more than eight billion acres

The population objection is the one everybody reaches for. If hardly anyone is dying, and people carry on being born, population growth gets bad, and we already have too many people.

De Grey's first move is to ask in what sense we currently have too many people. His answer: we do not. There are only 8 billion of us, and if you count the acres on the planet, even excluding all the places where it is not terribly pleasant to live, there are more than 8 billion. Everyone could have their own acre right now.

So the actual problem is not people, it is pollution. We are releasing too much carbon, making plastics and putting them in the ocean, and so on. And technology is coming fast for those. He points at the XPRIZE Carbon Removal competition funded by Elon Musk, which targets removing carbon from the atmosphere rather than merely slowing how much we add. He points at engineered bacteria that eat plastics which cannot otherwise be eaten, and at cheap desalination. These, he says, are all easy technological problems compared with the stuff he works on. By the time bringing aging under control produces any demographic consequences at all, they will be completely done.

And the distant version? Yes, in theory, in the far future we will have too many people unless we go to other planets. But run the numbers, even on pessimistic fertility assumptions, and it is many hundreds of years before there is any difficulty finding space. His verdict on planning around that:

It's really not intelligent to make decisions today on the basis of one's assumption about how the world's going to be many hundreds of years in the future. We have no idea whether kids are going to go out of fashion in 100 years.

Choosing the age you look, and the 300 year old dating your daughter

Will we be able to choose the age we want to look? "Oh, sure," de Grey says. "In the same way that you can choose the age you want your car to look. It depends on how thoroughly and how frequently you do the preventative maintenance."

McCormack picks 35. He quite liked how he looked at about 35. Totally, says de Grey.

And then McCormack finds the uncomfortable corner. You could be a 300 year old man who wants to look 25, out there dating women who are actually 25. How would they know you are 300? Why would they care? Because, McCormack says, he has a 16 year old daughter, and when she is 20 he would not want her dating a 40 year old, so he certainly would not want her dating a 300 year old.

De Grey stops him. "Wait a minute." His first wife, the biologist, was 18 years older than him. She was 45 when he was 26. His current wife is 18 years younger than him. He has, as McCormack puts it, gone both ends. And he has had friends more extreme than that: his current wife's first husband was 40 years older than her, married when she was 25 and he was 65, and he was also a prominent gerontologist.

"Your dating in your world of dating is strongly aligned to your line of work," McCormack observes. "Kind of. Some people might say that."

McCormack still does not want a 300 year old dating his 20 year old daughter, and de Grey gives the answer that reframes the whole anxiety:

We won't have any 300 year old people for another 200 years, whatever happens. People will only get older at one year per year. The rate at which society will need to adapt to this kind of novelty is really slow, a lot slower than the rate at which humanity constantly adapts to new technology.

That is the underrated structural point of the interview. Even total success in the next 15 years produces no 300 year olds for two centuries, and every intermediate step arrives at the pace of an ordinary human year. There is no shock, only a very long adjustment.

McCormack objects that it would still be a vastly different world. De Grey narrows it hard:

I'll tell you how it'll be vastly different. There won't be any sick people. That's the only big difference. Other differences will exist, but they will be the result of other technologies, like AI.

We will have injured people, not sick people, McCormack offers. Even injured people will not stay injured for long, says de Grey, any more than they do now. When a young person gets injured, a footballer ruptures an ACL, they might be out for six or eight months, and that is about it. Youth already is the repair capacity.

Fertility, pensions, and the social care bill that stops existing

Will this extend fertility? Yes, totally. And then the twist that runs against intuition: de Grey expects it to lower fertility rates, because you can wait longer. McCormack plays it out: I'll wait until I'm 85, get my career together first.

De Grey grounds it in the demographic transition, which has happened in every single country in the world that has reached a certain level of female emancipation, education and prosperity, with the single exception of Israel. Fertility rates plummet. And at the same time as choosing fewer children, women have them later. They only have them a little bit later today because there is currently a deadline. Remove the deadline and, for exactly the same reasons they have them a little later now, they will have them a lot later.

Are there negative social consequences he considers? "No." It is all positive. McCormack asks what the positives are that they have not covered.

Prosperity, mostly. At the moment we spend the vast majority of the medical budget of the western world on the health problems of late life, which will not happen. The medicine to maintain people in youth costs money, of course, but a small fraction of what we spend today. Prevention is always better than cure, in medicine as in everything else. That is one of the big ones for him.

McCormack brings it home to the UK, where there is a live national argument about the provision of social care against an aging population and a falling birth rate. How are we going to pay for social care? De Grey's answer is that social care for age stops being a thing. And on pensions:

People say, how will we pay the pensions? They kind of forget that we've got this thing called AI coming along which is going to basically eliminate almost all jobs anyway. So the idea that we would have to make people work longer because they can makes no sense whatsoever.

McCormack steps back for a beat. Is it not interesting that we get to live in a time where all the stuff we saw in the films, all the stuff we were promised about the future, seems to be arriving at once? Abundant cheap energy, AI, longevity, all in one lifetime.

De Grey deflates it politely. That is one way of looking at it, and it is easy to see it that way from an outsider's perspective. When you are inside any particular field, you see the stuff that is too technical to explain to the wider world, so you watch an exponential rate of progress that started long before the dramatic breakthroughs that get the experts on television. "Yeah, but we're still getting to live it," McCormack says. "We're still going to see it ourselves."

"That's a theory."

The seven categories: what a heretic actually said

McCormack asks the question he has been circling: in AI they talk about the breakthroughs required, the recursive one they are chasing right now. What are the breakthroughs required in longevity?

Because the body is an insanely complicated machine, de Grey says, it inflicts on itself a large number of different types of damage. And here he identifies the moment he became known as a heretic in the field, roughly twenty years ago. "I'm not nearly so much a heretic now, because I basically won the arguments."

The heresy was this: there are all these types of damage, but they can be classified into a rather manageable number of categories. He identified just seven. Seven categories of molecular and cellular change that accumulate and eventually contribute to functional decline, mental and physical. So the breakthroughs needed are ways to repair each of those types of damage.

That is the entire strategic claim. Not seven diseases, seven categories of damage, each with its own repair approach, applied together.

Category of damage	What goes wrong	Repair strategy	In this interview
Cell loss and tissue atrophy	Cells die and, in some tissues, are not automatically replaced by division of the cells already there	Put in cells that know how to divide and become the missing type: stem cell therapy	Explained in detail
Intracellular junk	Waste molecules made too slowly for evolution to have bothered evolving a way to break them down, so they pile up over decades	Import enzymes from other species that can digest them	Explained in detail, the eye case
Extracellular junk	Waste that accumulates outside cells, poisoning the tissue around it	Extract it or break it down; the artery case he describes fits here	Explained via atherosclerosis
Extracellular crosslinks	Sugars in circulation form chemical bonds between long lived matrix proteins that were never meant to be joined, stiffening the tissue	Break the specific bonds, which are chemically unlike any bond that is supposed to be there	Explained via wrinkles and stiff arteries
Death resistant cells	Cells that should have died stay around and misbehave	Kill them selectively	Not covered on air
Mitochondrial mutations	Mutations in the small separate genome inside mitochondria	Put backup copies of those genes into the safety of the cell nucleus	Not covered on air
Nuclear mutations and epimutations	Damage to the main genome, whose chief late life consequence is cancer	Attack cancer at the level all cancers share rather than one tumour type at a time	Not covered on air

Figure 2. The seven category classification de Grey refers to but does not enumerate in this interview. The categories and their paired repair strategies are the framework he set out in Ending Aging and in the SENS programme; the right hand column marks which ones he actually walked McCormack through on air. Four of the seven get a full mechanistic treatment in this conversation, which is why the biology below runs so much deeper on junk and crosslinks than on mitochondria or cancer.

Stem cells: the easy one to describe

De Grey starts with the category everyone has already heard of. What is stem cell therapy? You put cells into the body of a particular type so that they can divide and transform into replacements for cells that are needed. Why would cells be needed? Because cells sometimes die, and in some tissues they are not automatically replaced by division of the cells already there. So you put in cells that know how to do the dividing, restore the number of that kind of cell, and you are back where you started. That is one type of rejuvenation, one type of damage repair.

Intracellular garbage, and why you go blind

The harder categories start with waste. Cells run a lot of different chemical reactions all the time, with functions like replicating DNA, and many of those reactions also produce waste products. Those waste products accumulate unless they are destroyed or excreted. Everything produced at a respectable rate does get destroyed or excreted, because otherwise the cell would fill up and die long before we could reproduce.

But some waste products are generated very, very slowly. They can accumulate for decades and decades before the cell really notices. And for that reason, evolution has not taken the trouble to invent machinery to excrete or destroy them.

It's just waste products. Same as not taking out the garbage in your kitchen for a month.

McCormack asks whether this is dead cells. No, molecules. And then de Grey gives the most detailed piece of biology in the interview, on vision.

Vision works through vitamin A in the retina at the back of your eye, which absorbs a photon and changes its molecular shape. There is a cycle, the visual cycle, in which the vitamin A molecule goes through four different shape changes and returns to where it started. One of those changes is where the light is absorbed, and another is where the signal from that absorption is pumped into the optic nerve so the brain knows a photon arrived. That is sight at the molecular level.

Very occasionally, one of those four reactions goes a bit wrong, a bit squiffy, and creates a different product, something that is not one of the four molecules supposed to be made. That new molecule is indigestible. It cannot take part in the visual cycle anymore, and the photoreceptor cells, the rods and cones, do not know how to do anything about it. So it accumulates. The molecule he is describing without naming is A2E.

The eye is interesting here, because it has a sophisticated workaround. The photoreceptor cell gets rejuvenated by having the garbage transferred into another cell behind it, a retinal pigment epithelial cell. The photoreceptor no longer has the garbage and can carry on working.

Unfortunately that is the end of the story. The RPE cells now have the garbage, and they too are unable to break it down. So they eventually end up chock full of this stuff, and eventually they die, and the result is macular degeneration, the number one cause of blindness in the elderly. "We'd like that not to happen, really."

McCormack, deadpan: "Is this why my eyesight's going?" There are various other reasons eyesight goes downhill, but this is a very important one.

The fix de Grey thought of nearly 30 years ago is the template for the whole intracellular junk category: identify enzymes in other species that are able to break this stuff down, and inject the genes for those enzymes into the eye so that the RPE cells can break it down after all. After many, many years of work, this has been shown to succeed. It has been taken up by a company, which he says is about to be bought by a big pharmaceutical firm. He does not name it on air; the programme traces back to the enzyme therapy work funded out of his foundation at Ichor Therapeutics.

The concept, he insists, is very straightforward. Garbage accumulates just the way it does in your kitchen. You need a way to extract it or break it down.

Atherosclerosis: the number one killer, same idea, different molecule

The second case is atherosclerosis, the number one killer in the western world. It is caused by the accumulation of a different molecule, an oxidation derivative of cholesterol if you really want to know, which is 7 ketocholesterol. It accumulates inside a type of white blood cell in the artery wall and poisons it.

Here the strategy is to extract rather than break down, and they have found ways to do it. That one is in clinical trials in Australia right now, and he says he is quite proud of it. Both of these, he notes, are spinouts from his foundation, from the nonprofit he has led over the years. The artery programme is the cyclodextrin work now carried by Cyclarity Therapeutics.

What the treatment actually looks like: mostly injections, eventually 300 in one syringe

McCormack jumps forward. In the future, are we going to have a series of injections and pills we take? How will this work?

Mostly injections. Some of it may come as pills, but most of what is needed is gene therapies and cell therapies, and you cannot eat those. They get broken down in the stomach.

In the early stages there may be things that have to be done with surgery, such as replacing organs with lab grown organs. De Grey does not know whether that will be necessary, and says that even if it is it will not be necessary for long, because it is very invasive and people do not like it, so there will be a lot of pressure to refine the treatments until they can be done by injection after all.

So, a lot of injections? Or one master injection? Kind of the latter, and his analogy is the MMR vaccine:

MMR is not a vaccine, it's three vaccines in one. One for measles, one for mumps and one for rubella. It's just that you get them all in the same inch. So I can totally envision a time in which we get 300 things all in one syringe.

McCormack makes it personal. Neither of them is the youngest person in the world. Say you solve this and he chooses to be 35. How does it reverse his aging?

That is what damage repair is, de Grey says. That is what rejuvenation is. And he takes a small detour to reclaim the word, which he says was pretty much owned by the cosmetics industry. Twenty odd years ago he was asked to edit a journal, and he named it Rejuvenation Research, precisely because rejuvenation means reversal, reduction, of biological age.

Would it be like choosing a percentage of rejuvenation? "Well, if you like. Yeah, that's a fine way to put it."

McCormack works out his own order: he does not mind having a gray beard, but he does not want gray hair. So that would be a certain injection he would have. "Sure."

Wrinkles, stiff arteries, and the bonds nobody could break

Then McCormack asks what is actually causing the wrinkles, and gets the fourth category.

One of the main reasons we get wrinkles, and the skin becomes less elastic, comes down to a type of chemical reaction distinct from the waste production described earlier. Rather than creating garbage that must be destroyed or excreted, this reaction damages molecules that are simply long lived.

The body is made not only of cells but of a lattice of proteins secreted by some cells. That lattice is the extracellular matrix, and it is laid down in a very regular way that gives tissue its elasticity. Elasticity matters for wrinkles, but it also matters in life threatening contexts. He separates two words people conflate: atherosclerosis is the plaque process he already described, while arteriosclerosis is a completely different process, the stiffening of the arteries, and it is the major reason we get high blood pressure in the elderly.

The chemical reactions causing both the wrinkles and the stiff arteries are the same, just in different places. They happen between these long lived proteins and sugar molecules in the circulation, and they create new chemical bonds between proteins that would not normally be joined together. That stiffens the material.

Inside the cell this basically does not matter. Some intracellular proteins have a half life of an hour, some a half life of a decade, but none have a half life comparable to a human lifetime. So when a protein is destroyed and replaced by resynthesis, the extra bonds go in the bin with it.

The extracellular matrix is not like that. It lives for a long, long time, and in particular the proteins responsible for elasticity, the main one being elastin, are basically not turned over at all. They live for the lifetime of the organism. So they accumulate this damage, and that is why they get stiffer.

YOUNG MATRIX fibres slide past one another elastin, laid down once, never replaced

AGED MATRIX fibres bonded together, no give sugar in the blood welds fibre to fibre

decades IN SKIN loss of elasticity wrinkles IN ARTERIES arteriosclerosis, stiff vessels high blood pressure in the elderly

the repair target: break the bond, which is chemically unlike anything meant to be there

Figure 3. The crosslinking mechanism de Grey walks through, rebuilt. Long lived matrix proteins are laid down early and effectively never replaced, so any chemical bond that forms between them is permanent. Sugars in the circulation form exactly such bonds. The cosmetic result is wrinkles; the lethal result is arteries that no longer flex. Because these bonds have a chemical structure unlike any bond that is supposed to be present, a molecule that breaks them selectively is at least theoretically possible without side effects, which is the opening he has been aiming at for two decades.

There are two ways out. One is to find a way to cause turnover, resynthesis of the stuff that is not normally turned over. That, he says, turns out to be really hard, but it might be possible. The other is simply to break the accumulated bonds. And that is not obviously impossible, because those bonds have a different chemical structure from any of the bonds that are supposed to be around, so one could theoretically find ways to break them that would not have side effects.

Then the news. Another spinout from his foundation has very recently published what he calls a huge breakthrough in this area. They have not solved the problem completely, but they have taken a step that everybody thought was impossible. "I'm pretty happy about that." The crosslink he is referring to is glucosepane, and the spinout is Revel Pharmaceuticals.

Is there a peak age?

McCormack asks whether there is a peak age, a point where the body is in its best shape.

Sure, though sometimes you have to take a little care in order to measure it. Athletes obviously start going downhill after a while, though he notes the age at which people win world records in a lot of sports has gone up over the years, and there are many reasons for that which are not necessarily about aging at all. Improved training, for instance.

The general shape is that one reaches a peak, stays close to it for a while, and then goes downhill at an accelerating rate. But for people who are not exerting themselves like elite athletes, he does not think it matters very much whether one is at peak or 20 percent below. And one does not get to be 20 percent below until middle age.

"Would it be expensive?" Two answers, and the second one is free

"So basically, no." But he wants to answer it in two different ways.

The first question is whether it would be expensive to deliver. Would the actual process of manufacturing the stem cells or the gene therapies be expensive? Moderately expensive, yes, as any new technology is, and it would come down as all technologies do. But, as he said earlier, vastly less expensive than the alternative of letting people get sick and then trying to keep them alive in a poor state of health, which is what we do today.

The second answer is that it will be free.

Even in this crazy country, the USA, which doesn't like taxes and thinks it's okay to provide healthcare through insurance, it will be free, simply because it will pay for itself at the level of national prosperity so many times over so quickly. It will just be economically suicidal not to make sure that everybody who is old enough to need these treatments can get them irrespective of their ability to pay.

McCormack imagines that to begin with it will be a luxury. De Grey says not at all, and the reasoning is worth following, because it is the least obvious argument in the interview.

Achieving this is a big project, he says. Like the Apollo program, but much bigger, and it will be going on for a decade or so once humanity finally sorts itself out and starts to care about aging. What he works on right now is the aging of mice. And he believes that once we make sufficiently dramatic breakthroughs in making mice live healthier for longer, and therefore live longer overall, that will wake the world up.

Not because of the mice. Because of who notices.

It'll wake up people like you, influencers, people who people listen to. People don't listen to scientists and people don't care about mice, but they do for whatever reason care what Joe Rogan says, or Oprah Winfrey, or Kim Kardashian.

At that point the world's leading influencers come out and say the experts have said we are in a different universe now, we have a fighting chance of bringing aging under medical control fairly soon, so let us make it a little bit sooner and save 110,000 lives a day, which is how many people die of aging worldwide.

110,000 a day. That is the number the entire interview keeps returning to. He predicts a reaction that will dwarf the reaction to covid, and offers his favourite slogan of the moment: "I want to make aging the new covid."

The mechanism that gets from a mouse study to free treatment therefore runs through public attention rather than through price. By the time the treatments actually arrive, governments will have sorted this out, understood the economic benefits, and frontloaded the investment in infrastructure and training of personnel, so it will be possible to give absolutely everybody these things pretty much as soon as they become available.

So no, to answer your original question, it will not be a luxury even for a year.

"Not everyone will want it, though"

This is the objection McCormack keeps coming back to, and it is where the interview gets its best exchange.

"Well, wait, wait, wait, wait, wait, wait," de Grey says. Think about it. Who wants to be sick?

"Oh, people don't want to be sick. But I just think there won't be people who want to live forever. Some people, I've had enough."

Right. But if you don't want to be sick, there is this side effect that you tend to wake up tomorrow.

That is the whole rhetorical move in one line. He is not selling immortality. He is selling the absence of sickness, and pointing out that longevity is the arithmetic consequence. McCormack gets there himself: "so you won't have a choice, because the choice will be sick, not sick."

De Grey allows the obvious exit and then closes it. Of course you could always kill yourself, but we have an opinion about that in society. We have organizations like the Samaritans, and we are fairly pleased that they exist. He has never rung a suicide hotline, but he is fairly sure the first question they ask is not your date of birth so they can put the phone down if you are over 80. So if people do decide they no longer want to live, even though they are mentally and physically healthy, we will continue to try to change their minds however long ago they were born.

McCormack wonders whether our attitude shifts. We consider life precious partly because death is so final. If death becomes less regular, and we have this option, some people might just say, I have done 600 years, I am done.

De Grey deals with boredom first. Of course it is possible. But what do people do today when they get bored and are in peak condition? They do not say, I am bored, I am going to go and kill myself. They go and find something new to do. It seems perfectly reasonable to him that this continues.

Then he flags his own limits: this is about the distant future, and it is crazy to have opinions about it, crazy even to think about it, because we have no idea what context the world will be in. If people end up wanting to commit suicide, maybe we should just let them, and maybe society will take a different view. "But I doubt it personally."

Religion, omnipotence, and the Tibetan monk

McCormack raises the religious objection through the Tibetan monk, whose role is to spend their whole life prepared for death.

De Grey brings it straight back to the correction he opened with, the difference between aging and death:

Last I checked, in every religion God is supposed to be omnipotent, which means he's perfectly capable of striking you down with a thunderbolt, however healthy you are. And also, in every religion, in every holy scripture, it is very clear that you're not supposed to make the choice yourself. You're supposed to let God decide when he wants you upstairs, and not to hasten the whole thing any more than you're supposed to kill other people.

"Fair. Fair," says McCormack. Curing aging does not take the decision away from an omnipotent God; it removes one particular mechanism of decay. The scriptural prohibition, as he reads it, runs against hastening death, not against maintaining health.

McCormack turns the question around. Do you want to live forever?

De Grey does not. He wants to be here tomorrow. He thinks further ahead than one day, but certainly not further ahead than 10 years, and says it is pointless to try, because things are going to be so different 10 years from now. It is a quietly consistent position: the man arguing for indefinite healthy lifespan plans on a decade at most.

What nobody asks him, and the funding paradox

McCormack asks the good closing question early: given the media has been a bit crappy on this, what are the areas people do not ask about that he wishes they did?

"Where do I send the check?"

So how expensive is the work? This, he says, is the craziest thing. Medical research in general, not just for aging, runs a long process. It starts in a petri dish, goes through laboratory animals, and eventually reaches clinical trials, which themselves have stages, first for safety and then for efficacy. Each step is considerably more expensive than the last. But the curious thing is that the later, more expensive steps are easier to fund, purely because of de risking. Every step you complete raises the probability that you reach the finish line.

Which means the work he does, the very early stage in mice, is the hardest to fund even though it is not very expensive. Big experiments with a couple of thousand mice cost far less than even an early stage clinical trial. He puts the number at 5 or 10 million dollars.

McCormack, who has spent the interview noting that there are a lot of rich people out there who want to live forever, says exactly that again. And de Grey gives the line that explains three decades of fundraising:

Every billionaire has a different reason for not writing me a check.

Some have written checks over the years, and that is how he has been able to do as much as he has. But not very many. Some do not because their spouses will not let them, because they think it is evil. Some say they are too old and it will not be in time for them, which he says really blows his mind, because last time he checked it was quite fashionable to care about your kids. Some say they are too young and it will come along in time for them anyway. In both cases, he points out, you are only caring about yourself.

He does not think about himself at all. What gets him out of bed is the number:

110,000 a day. I'm doing as much as more or less anybody in the world to hasten the defeat of aging. So every day that I bring forward the defeat of aging, which I probably do about once a month, is 110,000 lives. That's 30 World Trade Centers. It's quite easy to get out of bed for that.

McCormack notes that his son, who produces the show and will be editing this, is going to hear all of this and think: does that mean I do not get an inheritance? "Yeah, going to be around forever, dude." (Sorry, Con. He is only joking.)

The mouse experiment: four interventions and the additivity result

So how are the mice getting on? What are the breakthroughs?

De Grey warns that this answer is slightly complicated, and the complication is the strategic core of his programme.

Because aging is such a complicated phenomenon, with so many different types of damage accumulating, the way we bring it under control is by addressing a lot of things at the same time. Giving a lot of different damage repair treatments to the same individual, together. That is true in mice as it is in humans, though perhaps a little less true in mice, because mice do not live very long in the first place. Two and a half years, basically. In a short lived animal you can get away with repairing only a subset of the damage, because there is enough cross talk between the various damage accumulation systems that you still get an effect.

So they ran a study, starting a bit over three years ago and ending a bit over one year ago, involving four different interventions. What they wanted to demonstrate was additivity: that you get more benefit in terms of lifespan by doing all these things together than by doing any one of them individually.

That is what happened. They used only interventions that other people had already reported to have some life extension benefit individually, put them together, and got an additive effect.

But the magnitude was not enough to be earthshattering for the expert community, which is the audience that matters, because the goal is to convince the experts that this is the way to go.

Why was it not earthshattering? Because there is a very simple, century old way to make mice live quite a bit longer: feed them less than they would like. Feed them 30 to 40 percent less and they live 30 to 40 percent longer. Any new approach is measured against that benchmark.

Why starving yourself will not work, and what the animal data says

Does calorie restriction translate to humans? "Unfortunately, it does not." And once again the explanation is evolutionary.

Long famines do not happen as often as short famines. So the selective pressure to optimize your metabolism in response to short famines exists, but the pressure to optimize for long famines does not. The practical consequence is that the amount of life extension you can get by starving an animal to just the right degree in just the right part of its lifespan ends up being about the same absolute number of years, roughly one, rather than a proportion of the natural lifespan.

Which is devastating for a species that lives 80 years and wonderful for one that lives three weeks. He runs the ladder:

worms at least 5x lifespan

mice 30 to 50%, about 1 year

dogs about 10%

monkeys a couple of percent, if you are lucky

humans hardly anything

short famines were common enough to select for; long ones were not

Figure 4. The numbers de Grey gives on air, drawn to scale. Worms living about three weeks can have their lives multiplied by at least a factor of five with this kind of trickery. Mice get 30 to 50 percent if you are lucky. Dogs get about 10 percent. Very expensive, very long monkey experiments returned a couple of percent. The pattern is not proportional, it is roughly a fixed number of extra years, which is why the same trick that transforms a worm does almost nothing for a human. This is the ceiling on the whole "slow the damage" approach and the reason he works on repair instead.

He also disposes of the obvious objection that people like eating. That has more or less been solved, he says, because drugs have been found, and they used one of them in their own study, that trick the body into thinking it is in a famine when it is not. The cells behave as they would in an actual famine. These are called calorie restriction mimetics. But you cannot get more benefit from the drugs than from the restriction itself. So it does not help.

The breakthrough that changes everything: beat calorie restriction by a factor of two

What is the most important breakthrough he needs to make? The one he thinks about the most?

I want to do a lot better than calorie restriction in mice. If we can find a cocktail of interventions of damage repair, rejuvenation interventions, that we can put together in mice that will give maybe twice as much as calorie restriction does, then that's going to completely change the world.

Every expert would agree we are in a different universe. They would say this really shows that animals with legs and fur, not fruit flies, can actually be made to live a lot longer healthily with things we already have at our disposal today.

And then he gives the target in round numbers, which is the most concrete falsifiable claim in the whole interview. Mice typically live two and a half years. What matters is being able to treat people who are already in middle age, so the study starts mice at one and a half years of age. Start calorie restriction at that age and you get about four months of life extension. What he wants is eight months, or even twelve.

And if we can, then yeah, I can retire, basically. My job will be done.

That is the whole thesis compressed into a mouse: not a longer countdown, but a middle aged animal given a cocktail of repair therapies that beats the century old benchmark by two or three times, starting late. Because this is what leads toward escape velocity.

therapy 1 therapy 2 therapy 3 therapy 4 therapy 5

each round buys enough time to reach the next, better round

biological age falling while calendar age rises

Figure 5. The concept behind the phrase de Grey coined. The upper line is the default: biological age climbs with calendar time. The lower line is a body under periodic maintenance. Each round of repair removes some accumulated damage, and crucially each round is better than the one before it, because the science keeps moving during the years the previous round bought. Escape velocity is the point where the improvement per round outpaces the damage accrued between rounds, so remaining life expectancy grows faster than time passes. That is what his 50/50 within 12 to 15 years refers to. It is not a cure administered once; it is maintenance that keeps winning.

"Death gives life meaning." "Bollocks."

McCormack wonders aloud what this does to the meaning of life, because in some ways death gives life a lot of meaning.

"Bollocks."

"Yeah, no, it does. It does."

"Absolute bollocks. Go on then. Persuade me. Give me an argument for why death gives meaning to life."

McCormack takes the challenge seriously and speaks personally. He says there are two parts of death he thinks about, and asks whether de Grey is religious at all. No, but keep going. So McCormack explains: he is partly religious, born and raised a Catholic, has come back and forth with religion, but definitely feels a spirituality that is just part of him. He thinks a lot about the morality that comes with religion, and the potential judgment that comes with it. And if there is a chance he does not die, he does not have to think about that.

De Grey interrupts on the technicality that is also the point: there is no chance you will not die.

McCormack revises: there is a chance he could effectively live forever, and he thinks about that differently.

Effectively is a weasel word. What do you actually mean? If you die at age 100 or a thousand or a million, you still die, and that's up to God.

McCormack concedes and reaches for the far horizon, the heat death of the universe. De Grey allows we might be able to get away before then, but yes, heat death, things like that.

Then McCormack makes his real argument, which is not about religion at all. There are things you put off in life. When you are young you think you will live forever, and then you get old, and life changes, and you start asking: have I lived a good life? I have got a hundred years, hopefully, maybe 70 of which I have used, maybe 25 good years left. What am I going to do in that period? How am I going to be around my kids?

De Grey says he thinks about what he has done in life too, but it does not slow him down. He is not tempted to rest on his laurels.

"It's unwinding as I try and talk about it," McCormack admits.

"I should hope so."

And then the rebuttal, which is the strongest counterargument he makes all interview because it points at people who already live this way:

What do kids do when they don't see death? They think they're going to be living functionally forever. What do they do? They still have priorities. They still make decisions about what to do and what not to do, just the same as older people do. And I don't think there's anything somehow inferior about the decision making that youngsters make relative to the decision making that people in middle age make.

The existence proof for a life without a felt deadline is childhood, and childhood is not notably short of meaning.

Bryan Johnson, and the bet that AI arrives first

McCormack is still stuck on the timeline. "I just, 10 to 12 years, man." De Grey corrects him: 12 to 15.

There are other people working on this. McCormack mentions Brian Armstrong, who built Coinbase and is now working on longevity through NewLimit. Then he asks about Bryan Johnson.

De Grey knows him quite well. "He's an acquired taste, of course. He has chosen to be extremely outspoken in a rather unusual way about his desire not to die." And then the joke that is not entirely a joke: "I think he would do better by writing me a large check. And I've told him so numerous times."

But he states Johnson's actual position fairly. Johnson thinks AI is going to be everybody's salvation fairly soon, so really the only big issue is not to die of anything stupid before then. That is the logic of the whole Blueprint project: hold the line until the cavalry arrives.

De Grey's response is not a refutation, it is a risk assessment:

Of course he may be right, but he may be wrong. And I'd prefer not to take the risk.

Where AI helps, and where it has nothing to work with

Is AI helping de Grey's work? It helps the field quite a lot. He and his foundation do not use AI to speak of, but plenty of the people they work with do.

Many of the ways AI is being leveraged in medical research sit far outside longevity, covering everything. The most conspicuous example is AlphaFold, which tells you what shape a protein will take if you give it the amino acid sequence. People in aging use AlphaFold all the time. He points back at the crosslink breakthrough he mentioned earlier: that was made with heavy use of AlphaFold and would have been very much harder otherwise.

So why does his own foundation not use it? Because different areas of research need AI and others do not.

You don't need AI to tell you whether a mouse is alive or dead.

And this leads to the sharpest thing he says about the limits of AI in his field. In aging, the difficulty is that we have so little data of the right kind. We have plenty of biological data on genomics and proteomics and so on. What we do not have is data on what works and what does not work. That has to come from the boring wet lab experiments, giving mice this or that and seeing what happens. Otherwise the AI has nowhere to start from.

Which turns into a direct answer to a specific class of donor:

People who say, oh, there's no point in writing you a check because AI is going to do it all for us, I think that's misguided.

Fifteen years of AI, from a paper the field is ashamed of to a step function

Given that he worked in AI in the 1980s, McCormack wants to know what he makes of the last three years.

De Grey says he has been following the growth of AI as we know it today from pretty much the time it began, about 15 years ago. And he starts the story earlier, with something the field is rather ashamed of now: one academic paper by a couple of very prominent researchers, showing a mathematical result that a certain type of system was incapable of a certain type of function. The paper he is describing without naming is Perceptrons, by Marvin Minsky and Seymour Papert.

That paper was over interpreted. It was assumed that because this type of system could not do this function, other types of system, even somewhat more sophisticated ones of broadly the same kind, would also not be able to do interesting things. So everyone gave up on the approach and went in the direction de Grey himself was working in, which was often called good old fashioned AI: all very formal, writing programs that would prove mathematical theorems.

But the abandoned approach, neural networks, was not abandoned by absolutely everybody. A few professors around the world doggedly kept going. One of them, Geoffrey Hinton, who got the Nobel Prize a year or two ago, decided to run what de Grey calls a hail mary experiment with very large amounts of data, essentially by hiring a postdoctoral fellow who knew how to program GPUs. And it worked, far better than anyone expected. (De Grey says Stanford; Hinton's lab was at the University of Toronto, and the result was AlexNet.)

People started trying variations on that theme, and very rapidly you got the fundamentals of what you see today. Someone else in Hinton's lab invented the generative adversarial network, which he describes as a mainstay of how AI works now.

Since that fundamental breakthrough, he says, there has been more or less no surprise. And he reaches for aviation:

It's a bit like what happened with aviation starting in 1905 or whatever. People had been saying for a long time that they thought we ought to be able to fly. Eventually they got there, but once they got the fundamental breakthrough, the refinements that happened thereafter happened at a very rapid and very steady pace. We had perfectly respectable aircraft flying in World War I.

It has been the same with AI over the past 15 years. It has gone pretty much the way he would have expected technologies to go when no additional fundamental breakthroughs are needed and you just carry on refining.

Has anything been unexpected? Yes, and his answer is precise about what counts as a surprise: the main surprises are deviations from the curve of incremental refinement, when there is more of a step function.

His example is AlphaFold again, and it is a good one. The protein shape problem had a competition run every two years for decades, the CASP assessment, where people submitted their programs. There was basically no progress at all for decades. Then AlphaFold came along in its first iteration and won the competition, but not in an earthshattering way, just marginally better than what had been done before. Two years later the next version came back and wiped the floor with everybody. "That was very sudden, and I guess the sudden things are the surprising ones."

AGI, and the people he trusts to worry about it

What about AGI? Will it be achieved? McCormack admits he does not even know exactly what it means.

De Grey notes the terminology is being refined, with people now distinguishing AGI, artificial general intelligence, from ASI, artificial super intelligence, essentially by how generally the machine is better at things than humans. In practice, he thinks, there may not be much distinction between the definitions in terms of timing, because everything is moving so fast. But he is willing to state the destination:

I certainly think that for all practical purposes we will get to a point where everything that we would ever have thought that we wanted a machine to do, a robot or anything, they will be able to do.

And do you think it will kill us all?

"Hope not."

McCormack finds the irony immediately: you do all this work, you get us living forever, and then the robots come and kill us.

De Grey's position on AI risk is essentially delegation, and he is explicit that this is what it is. The debate about whether AI gets out of control and kills us all has been around a long time, and he has been friends with many of the people who led it. He names Jaan Tallinn, Nick Bostrom and Anders Sandberg, and says he does not know Nate Soares, whom McCormack has interviewed on this same show. His conclusion:

It's an important question, but at the end of the day the critical thing is that the expert community is well aware of the possibility and is doing better than I could do to stay on top of that question and make it not happen.

There are a lot of very knowledgeable and very dedicated people working to stop it. He does not know whether they will succeed, but he has no better way to decide what he should do other than trust them and get on with what he is good at. Which, as McCormack points out, means raising money.

Why funding is so hard: Thiel, then four years, then six

"Is it that hard?"

"Oh, it's insane."

They have a lot of small donors and are very grateful to them, but the fact is this remains a minority pursuit, which means the occasional big donor makes most of the difference. Twenty years ago Peter Thiel became his first big donor, hooked after seeing his TED talk and meeting him at TED.

Then the shape of two decades of fundraising, in one sentence: it was another four years before the next big donor came along, and another six years before the one after that. Very stochastic, very hit and miss.

So he keeps trying. And he explains, at the end of an hour and seventeen minutes on a podcast, exactly why he is on the podcast:

The reason I do so many podcasts and so many talks at conferences is precisely to get the word out, to educate people to understand that we are within striking distance of doing this, and therefore the sooner we do it the better, and that's actually quite a good way to spend one's money.

Is there anything McCormack did not ask that he wishes he had? "I don't think so. You've been fairly wide ranging."

Where should people go? levf.org, the LEV Foundation website, which has a nice friendly donate button and all the information about what they do, what they plan to do, and why.

McCormack closes by hoping he lives to a thousand so they can do this again. De Grey: "That's the idea." If they make it, McCormack will celebrate his thousandth birthday doing this interview. Then the sign off to everyone listening, and to beautiful San Francisco.

1970sStarts programming at 15, having already decided he wants to do work of major humanitarian benefit. Picks AI, because he thinks it solves the problem of work.
early 1980sUndergraduate degree in computer science, then roughly seven years of AI research writing programs that prove mathematical theorems, the era he calls good old fashioned AI.
early 1990sMarries a senior biology professor from UC San Diego and learns biology over the dinner table. Discovers that no biologist he meets is interested in aging. "It's just decay, isn't it?"
1994Switches fields. Takes a deliberately undemanding bioinformatics job at Cambridge and repurposes the spare time toward aging, paying his own way to conferences because nobody is inviting him.
~1997Proposes importing enzymes from other species to break down the indigestible waste that kills retinal cells, the idea that eventually becomes a working therapy roughly 30 years later.
~2005Peter Thiel becomes his first big donor after seeing the TED talk. The next big donor arrives four years later, the one after that six years later.
2007Publishes Ending Aging, goes on television, and gets Stephen Colbert to break character laughing at the grandmother who helps you across the street.
~2022Launches the mouse study with four combined interventions, designed to test whether damage repair treatments are additive.
~2025Study ends. Additivity confirmed, magnitude not yet earthshattering. Separately, a foundation spinout publishes a step toward breaking crosslinks that everybody thought was impossible.
nextThe target: start mice at 1.5 years and beat calorie restriction's four months by getting 8 to 12 months. "Then yeah, I can retire."
12 to 15 yrs50/50 odds of longevity escape velocity. Twenty years ago the same claim was 25 years out.
+200 yrsThe earliest any 300 year old can exist, whatever happens, because people only get older at one year per year.

Figure 6. The arc de Grey lays out across the interview, from a teenage programmer who picked the wrong biggest problem to a mouse study with a specific numeric target. Everything after "next" is his forecast rather than a result.

Key takeaways

Aging and death are different problems. Death cannot be eliminated by technology, because there are too many ways to die. Aging, in de Grey's account, can be, because it is accumulated damage in a machine. Conflating the two is the error he says the media makes constantly, and it is the reason the public cannot think clearly about the subject.
Aging is physics, not biology. Any machine with moving parts damages itself through normal operation. The body's built in self maintenance covers most of it but not all, and the uncovered gaps produce nothing visible until middle age. Nobody blows a whistle to start the process.
Evolution set the pace, not the existence. Every multicellular species with a fixed body size ages. Species age at different rates because natural selection only invests in maintenance out to the age a typical individual actually reaches in the wild. Worked out in the 1950s and 1960s.
Seven categories, not a thousand diseases. The heresy that made his name is that all the damage sorts into a manageable number of categories, each with its own repair strategy. Four of the seven get a full walkthrough here: cell loss, intracellular junk, extracellular junk, and crosslinks.
Two concrete programmes are already in the clinic or close to it. Enzymes from other species injected to digest the indigestible waste that causes macular degeneration, and a molecule that extracts oxidized cholesterol from artery walls, in clinical trials in Australia. Both are spinouts from his nonprofit.
The treatment is mostly injections, delivered together. Gene therapies and cell therapies cannot be swallowed. The model is the MMR shot scaled up: eventually 300 things in one syringe, given periodically.
Calorie restriction is a dead end for humans and the benchmark to beat in mice. It buys roughly a fixed number of extra years regardless of natural lifespan, which is a fivefold gain for a worm and almost nothing for us. Mimetic drugs do not beat the restriction itself.
The falsifiable target is specific. Start mice at 1.5 years of age, where calorie restriction gives about four months. Get 8 to 12 months with a cocktail of repair therapies and, he says, the expert community concedes the argument and his job is done.
50/50 within 12 to 15 years for longevity escape velocity. He is the only senior figure who will give a number, and he notes that twenty years ago the same claim was 25 years out.
He expects it to be free, and the argument is economic rather than charitable. Late life healthcare is the majority of the western medical budget. Maintenance costs a fraction of that, so refusing to provide it universally would be economically suicidal.
The bottleneck is funding, and it is worst exactly where the work is cheapest. Early stage mouse work costs 5 or 10 million dollars and is the hardest money to raise, because every later stage is more de risked. First big donor twenty years ago, next one four years after that, the one after that six years later.
AI helps the field but cannot substitute for it. AlphaFold made the crosslink breakthrough tractable. But aging lacks data on what actually works, and only wet lab experiments generate it. "You don't need AI to tell you whether a mouse is alive or dead."

Where it stands

The interview is a rebuild of de Grey's case in his own frame, which is the right way to hear it. It is also worth knowing which parts of that case the wider field shares and which parts are his.

Broadly mainstream. The claim that aging is progressive accumulation of molecular and cellular damage is not controversial. Neither is the evolutionary explanation for why species age at different rates, which is the standard Medawar and Williams account from the mid twentieth century and is taught as such. The specific biology he walks through checks out: A2E accumulation in retinal pigment epithelium really is central to macular degeneration, 7 ketocholesterol really is implicated in atherosclerotic plaque, and sugar driven crosslinking of long lived matrix proteins really is why tissues stiffen. His point that calorie restriction scales poorly to long lived species is well supported, including by the primate studies he alludes to. His complaint about the funding valley for early stage work is a structural fact of biomedical research, not a grievance unique to him.

His own framework, contested in its organization. The seven category classification is de Grey's, from SENS. The framework most biogerontologists organize around instead is the hallmarks of aging, which lists twelve and is built around causal mechanisms rather than repair targets. The two overlap heavily in content and differ in intent: hallmarks describe, SENS prescribes. The historical flashpoint was the 2005 MIT Technology Review challenge, in which a panel of judges concluded that SENS was too speculative to merit serious debate but that no critic had actually demonstrated it was wrong. The prize went unclaimed. That verdict, that it is unrefuted rather than validated, is still roughly where the argument sits.

His alone, and unproven. The 12 to 15 year figure at 50/50 is far more aggressive than any survey of the field would produce, and the most useful evidence about it is the one he supplies himself: two decades ago he said 25 years. No comprehensive damage repair cocktail has extended maximum lifespan in a mammal by the margins he is targeting. His own study confirmed additivity but at a magnitude he concedes was not earthshattering. The benchmark venue for claims like these is the NIA Interventions Testing Program, which replicates lifespan results across three independent sites, and where rapamycin remains the strongest reproducible single intervention. Until a repair cocktail clears that kind of bar, the mouse target he names is a plan, not a result.

Outside his field entirely. The claims about population, pollution, pensions, universal free provision and the political response to a mouse result are assertions rather than modelled forecasts, and he says as much about the far future ones. "It will be free" is a prediction about political economy from a biologist. The argument behind it, that maintenance is cheaper than late life sickcare, is sound as far as it goes; whether health systems act on sound arithmetic is a separate question with a discouraging track record.

None of that touches the part of the interview that is hardest to answer, which is the question McCormack's wife could not answer on the phone. Tell me what day is the day you want to die on.

Chapters

0:00:00 Trailer
0:00:48 Can We Cure Ageing?
0:04:24 The 12–15 Year Prediction
0:06:17 The Body Is a Machine
0:15:21 How Aubrey Entered Longevity
0:18:45 Would You Want to Live Forever?
0:20:15 Does Longevity Change Risk?
0:21:53 Will Population Explode?
0:24:22 Can You Choose Your Age?
0:30:19 The Seven Types of Ageing Damage
0:36:54 What Rejuvenation Treatment Looks Like
0:45:35 Will It Be Expensive?
0:48:39 Who Would Refuse It?
0:53:00 Why Longevity Research Lacks Funding
0:56:02 The Mouse Experiment
1:00:18 The Breakthrough That Changes Everything
1:01:35 Does Death Give Life Meaning?
1:05:22 Bryan Johnson and AI
1:12:32 AlphaFold, AGI and AI Risk
1:15:13 Why Funding Is So Hard

Notable quotes

"One of the most frustrating things about my interactions with the media is that they constantly conflate the word aging and the word death." Aubrey de Grey, 1:05

"People make up these fantasies that tell them that they can put it out of their minds and get on with their miserably short lives." Aubrey de Grey, 3:10

"I always say we have a 50/50 chance of getting to a point that is for practical purposes having defeated aging, something that I call longevity escape velocity, within the next 12 to 15 years." Aubrey de Grey, 5:15

"We age because cars age. It's like entropy." Aubrey de Grey, 7:20

"Who blows the whistle to say, okay, now start aging please? That doesn't happen." Aubrey de Grey, 9:10

"It's just decay, isn't it? What fundamental truths about the universe are you going to understand by studying decay?" Aubrey de Grey, quoting his first wife, 16:45

"Tell me what day is the day you want to die on." Peter McCormack, recounting the phone call with his wife, 18:55

"It'll be fine. Your grandmother will be able to help you across the street, because she'll still be healthy." Aubrey de Grey, on Colbert's risk question, 20:45

"We won't have any 300 year old people for another 200 years, whatever happens. People will only get older at one year per year." Aubrey de Grey, 28:35

"I'll tell you how it'll be vastly different. There won't be any sick people. That's the only big difference." Aubrey de Grey, 29:25

"It's just waste products. Same as not taking out the garbage in your kitchen for a month." Aubrey de Grey, 32:35

"I can totally envision a time in which we get 300 things all in one syringe." Aubrey de Grey, 38:15

"It will just be economically suicidal not to make sure that everybody who is old enough to need these treatments can get them irrespective of their ability to pay." Aubrey de Grey, 46:45

"I want to make aging the new covid." Aubrey de Grey, 48:25

"But if you don't want to be sick, there is this side effect that you tend to wake up tomorrow." Aubrey de Grey, 49:45

"Every billionaire has a different reason for not writing me a check." Aubrey de Grey, 54:20

"Every day that I bring forward the defeat of aging, which I probably do about once a month, is 110,000 lives. That's 30 World Trade Centers. It's quite easy to get out of bed for that." Aubrey de Grey, 55:15

"If we can find a cocktail of interventions of damage repair that will give maybe twice as much as calorie restriction does, then that's going to completely change the world." Aubrey de Grey, 1:00:25

"Bollocks. Absolute bollocks. Go on then. Persuade me. Give me an argument for why death gives meaning to life." Aubrey de Grey, 1:01:45

"Effectively is a weasel word. What do you actually mean?" Aubrey de Grey, 1:03:15

"What do kids do when they don't see death? They still have priorities. They still make decisions about what to do and what not to do, just the same as older people do." Aubrey de Grey, 1:04:30

"I think he would do better by writing me a large check. And I've told him so numerous times." Aubrey de Grey, on Bryan Johnson, 1:06:05

"You don't need AI to tell you whether a mouse is alive or dead." Aubrey de Grey, 1:09:00

"I have no better way to decide what I should do other than just sit back and trust them and get on with what I'm good at." Aubrey de Grey, on AI risk, 1:14:40

Resources mentioned

The guest and his organization

Aubrey de Grey, biomedical gerontologist, president and chief science officer of the LEV Foundation
LEV Foundation, the nonprofit he directs people to at the end of the interview, with the donate page he mentions by name
Robust Mouse Rejuvenation, the combined intervention mouse study discussed at length
Aubrey de Grey on X
SENS, the seven category damage repair framework he built his career on
Ending Aging (2007), the general audience book that put him on television
Rejuvenation Research, the journal he founded and named to reclaim the word from the cosmetics industry
His TED talk, the one that brought in Peter Thiel

The host and this show

Peter McCormack and The Peter McCormack Show
Nate Soares on this same show, the AI risk conversation de Grey is asked about here

The biology

Longevity escape velocity, the term he coined
Evolution of aging, the mid century theory he refers to, from Peter Medawar and George C. Williams
Calorie restriction and calorie restriction mimetics, the century old benchmark and the drugs that fake it
Visual phototransduction, the four step vitamin A cycle he describes
A2E, the indigestible byproduct of that cycle
Retinal pigment epithelium, the cells that take the garbage and then die of it
Macular degeneration, the number one cause of blindness in the elderly
Atherosclerosis and 7 ketocholesterol, the number one killer in the west and the molecule behind it
Arteriosclerosis, the separate stiffening process he is careful to distinguish
Extracellular matrix and elastin, the lattice that is never replaced
Glucosepane, the crosslink at the centre of the recent breakthrough
Stem cell therapy, the damage repair category everyone already knows
MMR vaccine, his model for combined delivery
Hallmarks of aging and the NIA Interventions Testing Program, the mainstream framework and the mainstream replication venue

Companies and spinouts

Ichor Therapeutics, the enzyme therapy programme for retinal waste
Cyclarity Therapeutics, the artery programme now in trials in Australia
Revel Pharmaceuticals, the crosslink breaking spinout behind the recent publication

People named

Stephen Colbert, who broke character
Peter Thiel, his first big donor
Bryan Johnson, who is betting on AI arriving first
Brian Armstrong of Coinbase, now funding longevity through NewLimit
Elon Musk, who funded the XPRIZE Carbon Removal competition
Joe Rogan, Oprah Winfrey and Kim Kardashian, the people the world actually listens to
Geoffrey Hinton, the hail mary experiment, and AlexNet
Jaan Tallinn, Nick Bostrom and Anders Sandberg, the AI risk thinkers he counts as friends
Nate Soares, whom he does not know

Technology referenced

AlphaFold and the CASP competition it wiped the floor with
Perceptrons, the over interpreted paper that stalled neural networks for a generation
Generative adversarial networks, invented in Hinton's lab
Waymo, McCormack's first ride, the day before
Samaritans, the organization he uses to make his point about who we try to keep alive

Full transcript

======================================== I always say we have a 50/50 chance of getting to a point that is for practical purposes having defeated aging. At the moment, what I say is the next 12 to 15 years from now, we will without doubt come to the point where medicine is good enough that we can keep people biologically young, however long ago they were born, save 110,000 lives a day, which is how many people die of aging worldwide. >> Not everyone will want it, though. >> Well, wait, wait, wait, wait, wait, wait. Who wants to be sick? Oh, people don't want to be sick. But I I just think there won't be people who want to live forever. >> But if you don't want to be sick, there is this side effect that you tend to wake up tomorrow. >> Well, tell me what day is the day you want to die on. >> I don't think about it. I genuinely don't. >> We're good. Right. Aubrey, how are you, man? Good to see you. >> I'm well, thank you. Thanks for having me. >> Um, is it inevitable that we will cure death? >> No. Um, one of the most frustrating things about my inter interactions with the media is that they constantly conflate the word aging and the word death. >> Okay? >> It's painful. And the reason it's painful is because death is quite clearly something that cannot be completely eliminated by technology. Yes. Because you know there's quite a lot of ways to die. Um whereas aging is something that absolutely can be eliminated by medicine in the future. >> I mean that's kind of what I meant in terms of >> that's what you meant but don't say that. >> Okay. Um but yeah I mean [laughter] uh I mean the reason it's important is [clears throat] because there is this terror in society of thinking rationally about aging because even though it is something that will be o overcome by medicine in the fullness of time we don't know how much time and people find that very uncomfortable to think about. So they like to try to keep their kind of emotional distance from the question. >> Well they want to know if it was solved in their lifetime. That's right. And they can't tell them whether or not it will. And that's not good enough. So, people make up these fantasies that tell them uh that they can put it out of their minds and get on with their miserably short lives. And um and some of those fantasies are about desiraability. You know, trying to pretend that aging is some kind of blessing in disguise. But also there's the idea that aging is somehow not like diseases at all uh in some profound way such that it is some in some you know permanent manner um off limits to medicine right um and uh you know the media like to pander to you know what sells right so in other words they like to kind of perpetuate these ridiculous fantasies whether it's the desiraability one or the immutability one And um they do it in subliminal ways and using death when they mean aging is one of the worst. So that's why I picked you up on it at once. >> Right. Okay. So let's start with aging. >> Mhm. >> Is aging ultimately something we will inevitably have control over. >> Yes. >> Yeah. Okay. >> We will at the moment at the moment we've got basically almost no control over rating. We can just f very slightly slow down. Um but yes, we will without doubt come to the point where medicine is good enough that we can keep people biologically young however long ago they were born. >> Wow. But but we don't know when. >> That's right. >> Not not maybe in my lifetime. Well, so actually let me um say a little more about that because one thing that results from this terrible irrationality of society about aging is that most experts in the field feel unable to give even probabilistic time frame predictions for how long how soon this is going to happen. And that's a huge problem. Uh essentially it's um it it resolves from basically they feel they're in danger of being accused of saying irresponsible things and it will hurt their you know next grant application or whatever. >> But it would be it's not useful if you're looking for investment as well. Investment investors would want to know. >> Well that's also true. Yes. But of course you can I if you're running a company then you can address something that's not really aging or it's a bit of aging but you can call it a disease or something like that you know. So that doesn't that's not really the same problem. Um but yeah, so the point is hardly any of my colleagues are willing to give any kind of time frame predictions and this is enormously problematic because if people are not willing to give any kind of time frame predictions, even probabilistic ones, then the rest of the world just says, well, however optimistic people sound and however, you know, much noise they make about this or that breakthrough that's been made, um it doesn't really mean anything. We're never going to get there. And so they don't really feel that it's worth funding with taxpayers money. >> So um this is basically a complete footshooting exercise not giving these time frame predictions. But yet it's a kind of prisoners dilemma thing. You know the individual experts feel that they're going to hurt their own funding if they talk about time frames. So I've never had that problem. I've always given time frames and my time frames are not necessarily correct. Um, and they have slipped a little bit over the years, but I always say we have a 50/50 chance of getting to a point that is for practical purposes having defeated aging, something that I call longevity, escape, velocity. Um, within at the moment what I say is the next 12 to 15 years from now. Um, 20 years ago I was saying 25 years. So [snorts] um, it's a bit but not much. >> 12 to 15 years. Okay. Do you prefer to talk about the science of this or the social impact of this? >> Oh, I talk about both of those things all the time. You're welcome to ask me what you like. >> Well, I think a lot about the social impact, but I think we should probably do the science first. >> I don't know anything about this. >> All right. >> As far as I know, aging was something that the beauty industry cared about most of [clears throat] all. >> Um, >> I don't even understand the science of aging. I don't know why we age. I don't know why my hair, my beard's gone gray. I don't know why I'm starting to feel old and things don't work anymore. I don't know any of it. So, where's a good place to start? >> Um, the way I tend to start is I I start by emphasizing that we can start out with one fact that we all know, but that is often kind of glossed over, which is that the body is a machine. It's a really really really complicated machine and of course it's one that we don't have the blueprints of or anything like that but the fact is it's a machine which means that its function is determined by its structure by what it's made of right so we have a pretty thorough understanding these days of what the body is made of you know cells and stuff between cells and we know what cells are made of you know DNA and proteins and stuff um but the they all go together in this complicated way that we have a very poor understanding of um and the body functions um Now any machine you know with moving parts anyway does itself damage as a consequence of its normal operation. This is not a fact of biology. It's a fact of physics. So the aging of a living organism is actually pretty much no different from the aging of a car or an airplane. Once you understand that it becomes a nonquest you know why do we age? We age because cars age. You know it's like entropy. >> Okay. Now the reason that this is not obvious to people is because a large part of the complexity of the body consists of um very sophisticated arsenal of automatic built-in damage repair machinery, self-maintenance machinery if you like. Um and this self-maintenance machinery is not 100% comprehensive. If it were, then we wouldn't age at all. But that would require basically perpetual motion. Um so uh there are some gaps in what we have built into us. But uh for the most part this this is what happens and that means that we don't see any real consequences of the gaps of the things that the body can't repair on its own um until late in life. The um the confusion then comes. It looks like somehow um the body was repairing itself perfectly for a long time and then it stopped doing so. But that's nonsense because you know who blows the whistle to say okay now now start aging please. That doesn't happen. So the reason why is because we have been aging throughout our lives even starting before we're born. It's [clears throat] just that there's no macroscopic if you like um consequences of that until middle age or later. What what what is the biological reason for aging? Do we know this? Do we know evolutionary why? >> Well, sure. [gasps] Um, so coming back to the the fact that the body's a machine and that you know it would be a case of perpetual motion which is against the law of physics not to age at all. Then we can put it this way. We can say that all organisms are going to age all multisellular organisms with a fixed body size anyway. So you we can leave out plants which age in a very different way and things like that but um uh they all age but they age at different rates because they have different degrees of comprehensiveness of that self-maintenance machinery I was talking about. >> Um so the um real question is not why has evolution created aging. The re the real question is why has evolution tried harder with some species than with other species not to age to get close to non-agging? In other words, to slow down aging. And we understand that from an evolutionary perspective very well indeed. It's simply because the uh because evolution happens in the wild where there are lots of causes of death that have nothing to do with how long ago you were born. things like, you know, predation and starvation and hypothermia and so on. And there's no point in having machinery to make you age really, really slowly so that you'll live as long as a human being if you're rather low down the food chain and all and every individual in the species is going to be gone by the age of five because of predation or whatever. And for that that reason, mutations well evolution, sele natural selection um gravitates [clears throat] to a point where there is some aging in the time in the t time window, the average age that your um typical member of a species gets to despite predation and starvation and so on. Um but not much. So that basically some minority but not zero. And not a majority of any species will age before they get eaten. >> Huh. Never knew that. That's wild. >> And this was this was all worked out in the 50s and 60s a long long time ago. >> Okay. So So it's a there is an evolutionary process with aging, but also um a you know a woman can birth a child. It can create new life inside of her and create new fresh cells that you know that start the aging process. Is there any reason why the body cannot just pause aging? Is there any biological reason that our system could not just pause aging? >> Well, I mean, some species do have various systems for manipulating the rate of aging, essentially for trying harder in certain circumstances than in other circumstances to slow down aging. Essentially, activating more anti-aging machinery. But that comes at a cost of for example rate of growth. So um a fine example of this is the phenomenon of life extension by calorie restriction. It was discovered maybe a century ago that if you feed mice or rats less than they would like, then they live longer than if you feed them as much as they would like. Why would that be? Well, again, we can explain that from an evolutionary perspective pretty easily. We can say, well, if you're in a famine, then it's pretty pointless having offspring because those offspring are going to die of starvation before they get old enough to have their own offspring. So, there's no evolutionary selective advantage to doing so. It's better to hunker down [clears throat] and do your best to out outlive the famine so that you can have your offspring when there's food again. Uh, conversely, however, if you're low down the food chain like a mouse or a rat and um there's plenty of food around, then the best thing to do is to grow as quickly as possible to eat as much as you can and grow as quickly as possible so that you can have offspring as quickly as possible before you get eaten, right? Um so that's the kind of trade-off. Uh depending on the circumstances, depending on the environmental um you know vagaries um there will be different priorities for for how much of one's energy resources and so on to allocate to this self-maintenance process versus allocating it to growth and reproduction. So, so the the kind of average age of any kind of species of creature, animal or us, that is an evolutionary mutation. >> Well, I mean, mutations happen all the time from one generation to the next. And that's how that's why people that's why individuals within a species have differences and it's also why species themselves have differences. Yes. >> But it's the mut it's a mut it's still part of the mutating process, the age. >> Uh oh. Yeah. I mean basically we've got our genomes which encode a whole bunch of genes a lot of which are involved in making sure that we age as slowly as we do. So um genes in a mouse for example there are not there's there's not there's not the same level of sophistication of that part of the genome of those those parts of the genome as there is in humans and that's why they age more quickly. And does the a you know the way our body ages is it is it different in different parts of the body is it different systems? Well, I mean, yes and no. There are plenty of differences between different organs in terms of, you know, how regularly they regenerate, things like that. But at the same time, everything's talking to everything else, you know. So, there is a great deal of cross talk between different organs, different tissues with regard to rate of accumulation of molecular and cellular damage and consequent loss of function. This show is brought to you by my lead sponsor, Iron, the AI cloud for the next big thing. Iron builds and operates next generation data centers and delivers cuttingedge GPU infrastructure, all powered by renewable energy. Now, if you need access to scalable GPU clusters or are simply curious about who is powering the future of AI, check out iron.com to learn more, which is irre.com. I' I've got to ask, how did you get into this? Um, so I was originally a computer scientist. When I was 15, I started programming, found I was good at it. Thought I'm going to work. I had already decided by that age that I wanted to change the world. That I wanted to do stuff that was, you know, of of major humanitarian benefit. [snorts] And so I thought, well, I'm a good programmer. So what I'll do, I'll work on artificial intelligence because that's the way to solve one of the biggest problems in for humanity, the problem of work. the fact that, you know, we have to spend so much of our time doing stuff that we would not do unless we were being paid for it. >> Um, and that went okay. I did my undergraduate degree in the early 80s in computer science and I did for I think seven years um pretty effective, pretty successful research in AI. Uh that was when AI was nothing recognizably like what it is today. It's completely different way of doing things. I was writing programs that would prove mathematical theorems, things like that. Um um but then during that period I met and married a biologist who was a lot older than me. She was a senior professor at University of California San Diego and she was in England on sobatical. >> Okay. >> And so over the next couple of years, we're talking the early '9s now. I, you know, learned a lot of biology by accident over the dinner table. Um, and gradually it began to dawn on me that we were never talking about aging. I had always known since my earliest childhood that aging was certainly by far the world's biggest problem, considerably bigger than the problem of work. >> It kills the most people. >> Well, it Yeah. And it causes the most suffering and so on. So, um, uh, so basically, you know, I had completely assumed that everybody else thought the same and that biologists would be working on this. And of course, you didn't hear anything much, but you know, it was a hard problem. So, there you go. And that it never dawned on me until my first wife um and I were talking about this. And I said, I was we would have conversations that went like, you don't seem to be interested in aging. And she would say, no. And I would say, why not? And she would say, well, it's just decay, isn't it? You know, what fundamental truths about the universe are you going to understand by studying decay? And I would say, well, but yes, sure, but but it's bad for you. And she would say, well, that's not my problem. she would actually say that and I would say well it kind of is and that would be as far as we would get and you know quite quickly I began to find that other biologists that I was meeting through my wife had the same view and it was like a complete bombshell to me um took me a couple of years to come to terms with it to the extent of realizing that I had to switch fields but that's eventually what happened around 94 um [snorts] so yeah so I I happened at that point to be in a rather convenient um uh job situation. I had a a job at the University of Cambridge doing a bio working on a bioinformatics project and the job was very undemanding. So um you know I had plenty of spare time. I'd taken the job specifically because of that so that I could essentially do my AI research in my spare time because I'd run out of other funding for that. Um so really I was just repurposing my spare time and I was earning enough to be able to pay my way to go to conferences and so on. So, because of course I wasn't being invited back then. Um, and um, yeah, so that's that's where it started. I started having ideas that were wellreceived and here we are. >> It's interesting because uh, I mean, well, it's a good time probably to raise money on this because a lot of rich people out there probably don't want to die, who want to extend their lives as much as possible. But I was I was chatting to my um, my wife on the the way up. I gave her a call and she said, "Oh, who are you interviewing today?" And I explained I was interviewing you, Aubrey. And she said, "You know what's about?" And I said, "Well, it's basically about longevity, you know, you know, defeating aging. Um, you know, potentially in our lifetime we we might defeat aging." And she said, >> "Yeah, but you didn't want to I don't want to live forever." And I said, "Well, tell me what day is the day you want to die on." And she was like, >> "Yeah, I don't I don't have an answer for that." >> Quite an interesting because I I was thinking about the so many questions came through my head from that one thing. Mhm. >> What What does it mean to be a 300-year-old person? >> Mhm. [clears throat] >> Are you are you w Are you much wiser? Are you are you traumatized? >> What does it mean to have uh essentially a accelerating growth curve of people? >> Do we get to the point where there's too many? >> There's so many parts of the social side that I'm fascinated by. Another one I was thinking about is that if you could uh essentially essentially kind of live forever, >> how does that change your risk profile in life >> because an accident becomes really catastrophic? >> All right. So, let me deal with a few of those. >> Yeah. Where should we start? Let's do that. Let's do that last one because um I wrote a book in 2007, a general audience book called Ending Aging. And um when you write a book, you often get, you know, on TV and in particular Steven Cobear had me on his show. >> Awesome. >> Um and he asked me this exact question. He said, "Right, I mean, if you if we all going to live to like a thousand, you know, I mean, won't that kill people's ability to take risks? I mean, you're not going to want to cross the street because you can't cure being hit by a truck," he said. Um and I said, "It'll be fine. your grandmother will be able to help you across the street because she'll still be healthy. [laughter] And um I I I feel quite proud that I I still I think to this day I am the only scientist who has managed to get Colar out of character and get him to actually spontaneously burst out laughing on on air. [laughter] Um so yeah, that's that. Um then the population thing. Well, just on that one though, just to just to Yes. I I I mean, look, I'm not going to not get on planes and I'm not going to not do things, but perhaps as a society, death will become even more traumatic because it's >> right. Exactly. I mean, but the thing is I mean, yes, I think we will be more risk averse, but there are two ways to um lower one's risk of this or that. One is not to do the risky thing and the other is to use technology to make the thing less risky. Yes. Right. And of course, that's exactly what we're doing. We're building, you know, self-driving cars that will, you know, very greatly reduce the risk of the major cause of accident, the thing that kills more people than any other type of accident added together. >> I had my first trip in a Whimo yesterday. >> Yeah. >> Hadn't done it before. I thought it was a really weird experience, but it is >> good driver. >> Very safe. >> Yeah. Well, yeah. Uh um so uh so I mean Yeah. So that's that's a big thing and generally the answers the rebuttals of these various concerns that people have about a post-agging world do often come down to the fact that there are obvious technological solutions. So for example let's look at population. People say oh dear if hardly anyone's dying cuz not no one's dying of aging then um you know when people are going to carry on being born then population growth is going to be bad and we've already got too many people. But if we ask, you know, in what sense do we currently have too many people? The answer is uh we've got we haven't got too many people. We've got only 8 billion people. And if you count the number of acres on the planet, right? Even excluding all the places where it's not terribly pleasant to live, it's more than 8 billion. So, you know, everyone could have their own acre at the moment, right? So, the actual problem is simply pollution. It's the fact that, you know, we're burning too, we're releasing too much carbon and we are, you know, making plastics and and putting them in the ocean and, you know, and so on. But technology is coming fast to solve these things. You know, with even an X- prize for carbon capture that was funded by Elon Musk. Um, you know, removing carbon from the atmosphere, not just slowing down how long how much we put into it. Um, and similarly, you know, I mean, people are getting bacteria that that eat plastics that can't be eaten otherwise. And, you know, cheap desalination, things like that. These are all easy technological things compared to the stuff that I'm working on. So by the by the time we get any kind of demographic consequences for um you know of bringing aging under control all of these things are going to be completely done. And of course one can say well in the distant future we will have too many people unless we like go to other planets or whatever. Um, and in theory that's true, but if you do the numbers, even in pessimistic, you know, assumptions about fertility rates and so on, it's going to be many hundreds of years before we get to a point where there's any kind of difficulty finding enough space for people. And, you know, I think it's really not intelligent to make decisions today on the basis of one's assumption about how the world's going to be many hundreds of years in the future. It just makes no sense. We have no idea of whether kids are going to go out of fashion in 100 years, things like that. >> Will we be able to choose the age we want to look? >> Oh, sure. I mean, in the same way that you can choose the age you want your car to look. You can, you know, it depends on how thoroughly and how frequently you do the preventative maintenance. >> I I quite liked how I looked to about 35. >> I could choose 35. Go back to 35. >> Totally. And say that. That has its own weird consequences for it as well because you could be a 300y old man. You want to look like a 25-y old >> and you're out there trying to date girls who are 25 who are actually 25 and there's >> they How would they know you're 300? Let's >> Why would they care? well, I mean, I've got a I guess I've got a 16-year-old daughter. When she's 20, I wouldn't want her dating a 40-year-old. So, I certainly wouldn't want her dating a 300-y old. >> Wait a minute. Wait a minute. Um, so my first wife that we talked about a moment ago was 18 years older than me. I was 26. She was 45 when >> I think it works that way. Okay. >> My current wife is 18 years younger than >> you you've you've you've gone both both ends. >> That's right. Yeah. >> Yeah. And I've had friends who are um more extreme than that. So in fact, my current wife's first husband was 40 years older than her. They got married when she was 25 and he was 65. He was also a prominent gerontologist actually. >> Your dating in your world of dating is strongly aligned to your line of work. Kind of. [snorts] >> Some people might say that. I suppose. >> Yeah. But but I I don't know. I still don't think I want a 300y old dating my 20-y old daughter. >> Well, look at it this way. >> We won't have any 300y old people for another 200 years. Whatever happens. >> Yeah, we're No, no, no. Seriously think no seriously think about this >> because people will only get older at one year per year society will adapt at uh the rate at which society will need to adapt to this kind of you know novelty is really slow really a lot slower than the rate at which humanity constantly adapts to new technology >> it'll be a vastly different world though I mean you could be having so >> I'll tell I'll tell you how it'll be vastly different there won't be any sick people That's the only big difference. I mean, other differences will exist, but they will be the result of other technologies like AI. >> We'll have injured people, not sick people. >> Even injured people won't stay injured for very long anymore than they do now. You know, when when when someone young gets injured, you know, they might, you know, footballer ruptured his ACL, maybe out for six or eight months, but that's about it. >> So, will this this be able to, I guess, extend fertility? >> Oh, yeah. Yeah, >> totally. >> And so >> which which paradoxically will probably lower fertility rates >> because you can wait longer. >> I'll have I'll wait till I'm 85, get my career together first. >> That's right. If there is any we have of course seen this in every single country in the world with the single exception of Israel that has reached some certain level of female emancipation and education and prosperity. You have this thing called the demographic transition, this plummeting of fertility rates. But at the same time as women choosing to have fewer children, they also have them later. And they only have them a little bit later because currently there's a deadline. But when there isn't, you'd imagine that for the same reasons that they have for having them a little bit later today, they will have them a lot later. >> Yeah, that's really Yeah. What What are the So what are the social consequences that you think about? Are there negative social consequences that you consider? >> No, >> it's all positive. >> Right. So what are the other positive ones that I've not mentioned here? >> Oh well I mean lots more prosperity. I mean at the moment we spend a vast the vast majority of the medical budget of the western world on the health problems of late life which won't happen. [snorts] Of course the medicine to maintain people in youth will cost money but it will cost a small fraction of what we're spending today. You know prevention is always better than cure and medicine is in everything else. So that's really for to me one of the big ones. >> Well, actually we have a big conversation in the UK at the moment about social care, the provision of social care. We have an agent population. >> Um we have a lower in birth rate and uh how are we going to pay for social care? But I guess >> social care for age won't be a thing. >> Well, that's right. And um you know people say how will we pay the pensions you know if people are return they kind of forget that we've got this thing called AI coming along which is going to basically eliminate almost all jobs anyway. And so the idea of having, you know, the idea that we would have to make people work longer because they can makes no sense whatsoever. >> Isn't it interesting that we get to live in a time where every one of these technologies seems to be reach like all the stuff we saw in the films, all the stuff we were promised about the future all seems to be coming at the same time. Like abundant cheap energy, AI, longevity, it's all happening in our lifetime. >> Well, I mean, that's one way of looking at it. I mean I think it's easy to look at it that way from an outsers's perspective. When you're in any particular field, you see the stuff that is too technical to explain to the wider world. So you can see this this exponential um you know rate of progress in the technology that started a long time before uh you got the dramatic breakthroughs that you know get the experts on television as well. >> Yeah. But we're still getting to live it. We're still going to see it ourselves. >> That's a theory. >> It's a It's a wild time to to to go through. Okay. So, talk to me then about what is the science? What you know, I know when I've interviewed the guys in the world of AI, they talk about like the breakthroughs that are required or the breakthroughs they've had to have reached and like at the moment they're looking for like the recursive breakthrough with AI, you know, different conversations. What are the breakthroughs in longevity that are required? So because the body is this insanely complicated machine, it accumulates uh inflicts upon itself a large number of different types of damage. The first big um time when I started to become known as a bit of a heretic in the field. >> Okay, you're a heretic. >> Oh yeah, come on. Well, I'm not nearly so much a heretic now because I basically won the arguments, but um but I certainly was 20 years ago. And the main thing that um that I said back then was well there's all these types of damage but they can actually be classified into a rather manageable number of categories. I identified just seven categories of damage molecular and cellular changes that happen that eventually contribute to um functional decline mental and physical and uh so we've got to solve all of those things. So the breakthroughs that we need are in developing ways to repair these various types of damage. So some of the categories are quite simple to describe. For example, everyone's heard of stem cells and stem cell therapy. Okay, what is stem cell therapy? What you're doing is you're putting cells into the body that are of a particular type so that they can divide and transform into replacements for cells that are needed. Now why would cells be needed? Answer is because cells sometimes die and sometime and in some tissues they are not automatically replaced by the division of the cells that are already in the body. Right? So we put cells in that know how to do that dividing thing and restore the number of that kind of cell and we're back where we started. That's one type of rejuvenation of damage repair. Um there are other ones which are much more complicated to describe but some of them are pretty simple in in basics. So waste products the cell does a lot of cells do a lot of diff different chemical reactions all the time and those chemical reactions have of course functions like you know replicating DNA and things like that but they also in many cases produce waste products. Now those waste products accumulate unless they are either destroyed or excreted, right? And all waste products that are created at a respectable rate um are either ex either destroyed or excreted because they you know if they weren't that they would just accumulate in the cell and the cell would die pretty quickly and we wouldn't live long enough to reproduce. But it turns out that there are some waste products that are only generated very very slowly. so that they can accumulate in the cell and it takes decades and decades before they get to a level of abundance that the cell really notices. And for that reason, evolution has not taken the trouble to invent machinery to excrete or to destroy those things. But it's just waste products. Same as not taking out the garbage in your kitchen for a month, right? >> How does it exist? Is it just dead cells or >> Oh, no. This is these are molecules I'm talking about. So molecules that are just for so for example let's talk about vision okay the way the vision works is that there is a well vitamin A basically vitamin A in the back of your eye in the retina that is able to absorb a photon right absorb light and when it does so it changes its molecular shape in a certain way so there's this thing called the visual cycle where the vitamin A molecule goes through actually four different shape changes and goes back to where it started. And one of those shape changes is where the light is absorbed and otherwise where the um signal from the absorption is pumped into the optic nerve so that the brain knows that a photon has been absorbed. All right, that's site at the molecular level without any details. Now, it just so happens that very occasionally one of those four reactions that makes this cycle um goes a bit wrong, goes a bit squiffy, and you create a different product, something that's not one of the four molecules that are supposed to be made. And that new molecule is indigestible. It it it can't take part in the visual cycle anymore, but it also the cell that this all happens in, the photo receptor cells, the rods and cones, they don't know how to do anything about this stuff. So it accumulates. Now actually the eye is a little interesting in this way. It's got a rather sophisticated structure. What happens is that the cell is kind of rejuvenated by the garbage being transferred into another cell behind the photo receptor cell called a retinal pigmented epithelial cell. And when it's in there um you know the the photo receptor cell no longer has the garbage. So it can carry on working and that's what happens. Unfortunately that's the end of the story. the the RP cells, the atomized epithelial cells um now have the garbage and they too are unable to break it down. So they eventually end up being chock full of this stuff and eventually they die and um the result is macular degeneration which is of course the number one cause of blindness in the elderly. Uh we'd like that not to happen really. >> Is this why I'm my eyesight's going? >> Well, there are various other reasons why eyesight goes downhill but this is a very important one. Um so um yeah so this stuff accumulates so so we'd like it not to happen and the way that I thought of to do this and this was nearly 30 years ago now was to identify enzymes in other species that are able to break this thing down this stuff and to know inject basically the genes for those enzymes into the eye so that the these cells the RP cells would be able to break this stuff down after all. And after a lot of work, many many many years of work, uh this has been shown to succeed. Uh and so, you know, this has been received by a company now. It's about to be bought by a big farmer. You know, it's quite a nice thing. But the concept is very straightforward. Garbage accumulates just the way it does in your kitchen. And one needs a way to extract the garbage or to break it down. And so, um that's what we do. Another case is atherosclerosis. That's the number one killer in the western world. that's caused by the accumulation of another molecule, different molecule. It's a very it's an oxidation derivative of cholesterol if you really want to know. And um this accumulates inside a type of white blood cell in the artery wall and poisons it and so on. And again, we've been able to find ways to extract this stuff in this case to actually extract rather than break down. Um and that's in clinical trials in Australia right now. That's actually I'm quite proud of that one because it's a well actually both of these are spinouts. spin out companies from my foundation from the nonprofit that I've led over the years. >> So, so okay. So, I want to then jump forward. In the future, are we going to have a series of injections and pills we take? How how how will this work? >> Mostly injections. Most >> um now I mean there's going to be a lot of details, right? Um, some of it may be available um with pills, but mostly it's going to be injections because a lot of what's going to be needed is gene therapies and cell therapies and you can't you can't eat those, >> right? >> Get broken down in the stomach. So, um, so that's part of it. Now, in the early stages, there may be things that we have to do with surgery like, you know, replacing organs with lab grown organs. Um but I don't know whether that's going to be necessary and even if it is it won't be necessary for long because of course that's very invasive and people don't like it and so there will be a lot of pressure to you know refine the treatments so that they can be done by injections after all but as I said there's a lot of different things to fix so there's a lot of injections or will there be not necessarily >> one master injection >> well kind of yeah I mean the example I tend to use is MMR which is a well-known thing that people get when they're young uh MMR is not a vaccine it's three vaccines in one. One for measles, one for MS and one for reubella. And um it's just that you get them all in the center inch. So I can totally envision a time in which we get 300 things all in one syringe. >> And so how would it work in terms of I mean look, you and I are we aren't the youngest people in the world >> and you start to solve this and I you know choose I don't know what age you would choose you want to be. You might be happy at the current age but I'd want to say be 35. How will it reverse my reverse my aging? >> Well, that's what damage repair is. Um, you know, this is what this is what rejuvenation is. You mentioned right at the beginning you were talking about how, you know, most people think of in terms of aging at a cosmetic level. You know, um, the word rejuvenation was pretty much owned by the cosmetics industry and I kind of reclaimed it. Uh, 20 odd years ago, I was asked to edit a journal and I named the journal rejuvenation research. Uh because rejuvenation means reversal reduction of biological age. >> So would it be like a percentage rejuvenation we'd be choosing? >> Well, if you like. Yeah, that's that's a fine way to put it. Yeah. >> Yeah. And so I like I don't want um >> I don't mind having a gray beard, but I don't want gray hair. That would just be a certain injection I would have. >> Sure. >> Huh. And then help me understand a little bit more about then the the aging of the cells because you know there's a number of things that are happening as I get older. Like the grayness is off it. I'm getting a few wrinkles starting to happen around here. I mean what what is causing the wrinkles? Why why does that happen in aging? Well, one of the main reasons why we get wrinkles, why the skin becomes less elastic, if it comes down to is another type of chemical reaction that I haven't mentioned yet, which um rather than just creating something from garbage that we need to destroy or excrete, um in this case, it damages molecules that are just longived. So the body is made not only of cells but also of a kind of latis of proteins secreted by some cells. This latice is called the extracellular matrix and it is laid down in a very regular way um that gives the gives that tissue elasticity. Elasticity matters quite a lot. It certainly matters in terms of the skin for wrinkles but it also matter matters in life-threatening context. So I mentioned atherosclerosis already. The completely different process that happens in the arteries is called arterio sclerosis which is basically stiffening of the arteries and that's the major reason why we have high blood pressure in the elderly. So um turns out that the chemical reactions that cause these things are the same um just different places in the body. The chemical reactions are between these proteins and sugar molecules that are in the circulation and they cause new chemical bonds between proteins that would not normally be joined together. So that stiff stiffens the stuff the the material. Um now in um proteins inside the cell this basically doesn't matter because proteins inside the cell some of them have you know a half life of an hour and some of them have a half life of a decade but none of them have a half life that's you know comparable to a human lifetime. So you know when a protein gets destroyed and replaced by this by reynthesis um then you know the damage of this kind this kind of extra chemical bonds don't matter very much well they don't matter at all cuz the stuff just gets thrown away. Um but the extra matrix is not like that. It lives for a long long time and in particular the proteins that are responsible for the elasticity there's a the main one is a protein called elastin in fact um you know these proteins are basically not turned over at all they live for the lifetime of us and so they accumulate this damage and that's why they get stiffer. So what we'd like to do is either um find a way to cause reynthesis turnover of this stuff that's not normally turned over. It turns out that's really hard, but it might be possible. The other alternative is just to break these uh chemical bonds that have spontaneously accumulated. And that's not obviously impossible because it um turns out that those bonds have a different chemical structure than any of the bonds that are supposed to be around, right? So, one could theoretically find um you know, ways to break them that would not have side effects. Turns out that another spin out from my foundation has actually just very recently published a huge breakthrough in this area. They haven't solved the problem completely, but they've done they've gone they've taken a step towards that which was way which everybody thought was impossible. So, I'm pretty happy about that. >> Okay, so you guys know that I'm not just an equities guy. I'm a Bitcoin guy, too. And for people like me, just holding Bitcoin has been the winning strategy. But for others, they like a little diversification. Whatever the strategy, each cycle, I'm always like, I wish I knew how to trade this. I just never try because I'm not a trader and I don't know what the hell I'm doing. So, this is where Arch Public comes in. They can help you implement a custom strategy that works for you. Because if you've held Bitcoin or Salesforce stock over the last year, man, [music] you have taken every draw down on the chin. But you can make that volatility work for you. They built a platform that automates layer trading strategies across stocks, ETFs, Bitcoin and commodities, handling buying, selling, profit taking and accumulation, all directly in your own exchange account. So when they asked to sponsor the pod, I said I have to see this first and they introduced me to their market way product where you can customize buys and sells based on the triggers, risk profile, and time scales that work for you. And that is what I've done. Now listen, I'm not telling you to go and blindly ape in. But if you're looking to protect your downside and make volatility work for you, give them a call. Ask to see every trade and make them prove the math. So if you're trying to accumulate smarter, take profits, use volatility or tax lost harvests alongside your CPA, take a look at Arch Public by heading over to archpub.com/peter. That's archpub.com/peter. Is there a peak age we have? Obviously when we're born we're you know we're a baby but we're hopefully a healthy baby and and then we grow and mature and our body doesn't is it like a peak age where our body is in the best shape is it? Oh, sure. I mean, um, [snorts] you know, sometimes you have to take a little care in order to measure that. Uh, you know, obviously athletes, you know, start going downhill, uh, after a little while. Though, if you've noticed, uh, the pe performance of athletes in a lot of sports has, you know, the age at which people win world records and so on has gone up a bit over the years. Um and so that you know there are reasons why that there there are many reasons why that might be and they're not necessarily to do with aging. They may just be to do with the improved training and so on. But um one one re one can reach a peak and stay within close to that peak for a while and then eventually one goes downhill at an accelerating rate. So yeah, I mean I wouldn't say I wouldn't say it really for people who are not exerting themselves like athletes, acite athletes that it really matters very much whether one is at peak or 20% below. Um and one doesn't get to be 20% below until middle age. >> Uhhuh. Yeah. Yeah. It's so fascinating like the consequences of this. I would it be expensive? So basically no. Um but of course I should answer that in two different ways. >> Yeah. The first question is would it be expensive to deliver would the actual um you know process of manufacturing the stem cells or the gene therapies or whatever would that be expensive and it would be you know moderately expensive because um you know any technology is of course it would go down as all technologies do uh but um yeah moderately so but as I was mentioning earlier vastly less expensive than the alternative of letting people get sick and then trying to keep them alive in a poor state of health which is what we do today. >> And for that reason, the other answer I wanted to give to expense is it'll be free. >> Even in this crazy country, the USA, which doesn't like taxes and thinks it's okay to provide healthcare through insurance and so on, it will be free simply because it will pay for itself at the level of, you know, national prosperity so many times over so quickly. It will just be economically suicidal not to make sure that everybody who is old enough to need these treatments can get them irrespective of their ability to pay. >> Yeah. I think I I imagine though to begin with it will be a luxury. >> Well, no, not at all. Think about it. If we once we achieve this we I mean the achievement of it is it's a big big project. It's like you know the Apollo program but much more so than that much bigger than that. And it'll be going on for a decade or so once humanity finally sorts itself out and starts to care about aging. So my what I what I work on right now is the aging of mice. >> And I believe that once we make sufficiently dramatic breakthroughs in making mice live longer, live healthier for longer, and therefore as a side effect of health live total longer, um once we do that, that will wake the world up. It'll wake the world up basically because it'll wake up people like you, influencers, people who people listen to. You know, people don't listen to scientists and people don't care about mice, but they do for whatever reason care what Joe Rogan says or Oprah Winfrey or Kim Kardashian. Yeah. And uh so yeah, there will be a point in which the world's leading influencers will come out and say, "Right, well, I mean, the experts have said we're in a different universe now. We have made enough progress that we have a fighting chance of bringing aging in humans under medical control fairly soon. Let's make it a little bit sooner and save 110,000 lives a day, which is how many people die of aging worldwide. >> 110,000 a day. >> That's right. Um so um so I believe that there will be this massive I mean it'll dwarf the reaction that there was to co it'll I my my favorite slogan these days is I want to make aging the new co >> [laughter] >> um um so >> not everyone will want it though. >> Well wait wait wait wait wait wait. So um what this means is that by the time these treatments actually arrive governments are going to are going to have sorted all this out. you know, they're going to have understood what the economic benefits are and so on, and they're going to have been able to frontload the investment in, you know, um, infrastructure and training of personnel and so on, so that it will be possible to give absolutely everybody these things pretty much as soon as they become available. Um, so no, to answer your original question, it will not be a luxury even for a year. >> Um, so and when you talk about people not not necessarily wanting it, well, think about that. Um, you know, who wants to be sick? >> Oh, people don't want to be sick. But I I just think there won't be people who want to live forever. Some people I've had enough. >> Right. Right. Right. But if you don't want to be sick, there is this side effect that you tend to wake up tomorrow. Right. >> Yeah. No, I know. Yeah. Oh, so you won't have a choice because the choice will be sick, not sick. >> Well, of course, you could always kill yourself, but we have an opinion about that in society. We have these organizations like the Samaritans that um, you know, we're fairly pleased that they exist. Yeah. And you know, I'm fairly I've never rung a suicide hotline, but I'm fairly sure that if you do, then the first question they ask you is not your date of birth, so that they can put the phone down if you're over 80, you know. So, I I'm fairly sure that if people do actually decide they want to not live any longer, even though they are mentally and physically healthy, we will continue to try to change their minds however long ago they were born. >> Yeah. But I wonder if we will have a [laughter] a shifted attitude to it in that, you know, we consider life so precious because death is so final, right? We we consider it this most precious thing for somebody to just not want to live anymore. You we have to deal with the emotions and the trauma of of death in a different way. But but if death is less regular >> and we have this option to live together, some people might just go, I've done I've done 600 years, I'm done. I actually um I mean there may be people who are against this for religious reasons. >> I'll come to that in a moment. But first of all, people just getting bored, which is actually what you were saying a moment ago. Um >> so of course that's possible. But you know what do people do today when they get bored if they are in peak condition? Yeah, they're not, I'm bored. I'm going to go and kill myself, >> right? They go and find something new to do, right? It seems it seems perfectly reasonable to me that that will continue to happen. However, you know, as I say, this is about the distant future. Crazy to have opinions about it. Crazy to even think about it because we just have no idea what context the world is going to be in. If people end up wanting to um wanting to commit suicide, you know, maybe we should should just let them and maybe society will take a different view of that. But um I doubt it personally. >> Yeah. But it might it might have a religious consequence because >> Okay. So let's talk about that. >> The Tibetans the the the role of the Tibetan monk is to spend their whole life prepared for death. >> Mhm. Yeah. But but but hang on. You see the thing is uh this comes back to what I was lambasting you for at the beginning of chat. Uh the difference between aging and death. you know, um, last I checked, in every religion, um, God is supposed to be omnipotent, which means he's perfectly capable of striking you down with a thunderbolt, however healthy you are. And also, in every religion, in every holy scripture, it is very clear that you're not supposed to make the choice yourself. You're supposed to like, you know, let God decide when when he wants you upstairs. Um uh and not to not to hasten the whole thing any more than you're supposed to kill other people. >> Fair. Fair. I Yeah. Okay. Okay. I It was like the like the conversation I had on the way up. >> I was like, "Pick the day you want to die." >> But it is a It is It is a question that people do ask. It's like or they do say, "I don't think I want to live forever." >> Well, right. Exactly. And so, >> do you want to live forever? don't. I >> you just want to you just want to be here tomorrow. >> Yeah. Well, I mean I think I think further ahead than one day, but I certainly don't think further ahead than 10 years. Um it's like pointless because things are going to be so different 10 years from now. >> Okay. So, look, if the media has been a bit on this asking that crappy question I started with, what are the areas that people don't ask you about that you wish they did? >> Where do I send the check? [laughter] Yeah. So, how expensive is this work? >> Well, this is the craziest thing. So, in medical um research in general, not just for aging, there's obviously a long process starts in a petri dish, goes through, you know, uh laboratory animals, eventually clinical trials, right? And there are various stages of clinical trials. First of all, testing for safety and then testing for efficacy. Um and each step is considerably more expensive than the last. But the curious thing is that the later steps are easier to fund to to obtain money for and this is just because of d-risking. Yes, every step you know shows that you might get to the actual finish line you have a higher probability of doing so. Um so that means that the work I do which is the very early stage in the pet tradition in mice um is actually the hardest to find even though it's not very expensive. We do big experiments with a couple of thousand mice, but you know, the price is far less than even an early stage clinical trial. We're talking like, you know, 5 or 10 million. >> Wow. Okay. Okay. I mean, there's a lot, like I say, there's a lot of rich people out there who want to live forever. >> Well, this is right. And I talk to a lot of them all the time. Um, and every billionaire has a different reason for not writing me a check. Um uh you know I mean of course some of them have written me text over the years and that's how I've been able to do as much as I have. Uh but not very many and yeah I mean you know some of them it's because their spouses won't let them cuz they think they don't want to they think it's evil or whatever. Uh some of them it's because >> people think this is evil >> sometimes. Um yeah I mean well you were talking about religion a moment ago right? Um uh you know some people will say oh I'm too old you it won't be in time for me so I don't care. I mean that one really blows my mind because last time I checked I thought it was quite fashionable to care about your kids. You know I mean but this is what this happens. People say I'm too old or people say I'm too young this is going to come along for in time for me anyway. You know in both of these cases you're just caring about yourself. I don't think about myself at all when I when I what gets me out of bed in the morning is that number I gave you a moment ago. 110,000 a day. >> Yeah. You know, I I I um you know, I I'm doing as much as more or less anybody in the world to hasten the defeat of aging. And so every day that I bring forward the defeat of aging, which I probably do about once a month, is, you know, it's 110,000 lives. That's 30 World Trade Centers. You know, it's quite easy to get out of bed for that. >> I think my son, cuz my son's my um producer on the show, he'll be editing this. He's going to listen to this and go, "Is that does that mean I don't get an inheritance?" So, yeah, going to be around forever, dude. You're not You're not going to get that. >> He will uh Yeah, he Sorry, Con. I'm only joking. Um, okay. All right. So, how are the [clears throat] mice getting on? What What What are the breakthroughs you've had? The big ones. >> So, it's I'm going to have to give a slightly slightly complicated answer for this one. I'm ready. Um, so as I said, the fact that aging is such a complicated phenomenon with so many different types of damage accumulating means that the way in which we're going to bring it under control is by addressing a lot of different things at the same time. So basically giving a lot of different damage repair treatments to the same individual at the same time. >> And that's true in my same as it is in humans. Um, however, it's perhaps a little less true in mice. mice because they don't live very long in the first place. >> Somehow, you know, you can play you you can >> How long was a mice live for? >> Two and a half years basically. Um uh you can get away with repairing only a subset of the damage and there's enough like if you like cross talk between the various damage accumulation systems that um that you still get an effect. So we're putting to we did we did a study starting let me see now three years ago three in a bit um ending about one and a bit years ago which um involved four different interventions and what we wanted to show was this additivity. We wanted to show that you got more benefit in terms of lifespan if you did all these things together than if you did any one of them individually. And that is what happened. Um pretty much we um you know we only used interventions which had which other people had already reported to have some life extension benefit individually and we put them together and we got an additive effect. Um but the magnitude that we got was not enough to know to be earthshattering for the expert community which is where we're starting of course we want to convince the expert community that this is the way to go. Um why wasn't it earthshattering? Well, basically because there's a way that you can make mice live quite a bit longer than um than they normally do. And it's very simple. It was discovered about a century ago. All you need to do is feed them less than they would like. If you feed them 30 40% less than they would like, then they'll live 30 40% longer than otherwise. Um >> does that translate to humans? >> Well, this is the thing. Unfortunately, it does not. It, and again, this is easy to explain from an evolutionary perspective. Um uh remember I talked about famines earlier. >> Um essentially long famines don't happen as often as short famines. So the selective pressure evolutionary selection pressure that exists to optimize your um metabolism in response to short famines does not exist for long famines. So that means that basically um if you well to a good approximation anyway the amount of life extension that you can get by starving an animal to just the right degree in just the right part of their lifespan um ends up being about the same number of years like roughly one year um rather than um like proportion to the natural lifespan. So yeah, we get hardly anything from calorie restriction and this goes right through the animal kingdom. The it's kind of thing has been done on little worms that live about 3 weeks on average and you can multiply their lives by at least a factor of five doing this kind of trickery. Um but with mice as I say it's 30 40% maybe 50 if you're lucky. Um with dogs it's like 10%. you know miss a couple of very very expensive very long experiments were done on monkeys some time back and you got a couple of percent of your lucky you know so not not the answer there is also of course the issue that people like eating uh but that issue has been more or less solved because drugs have been found and we actually used one of them in our study that um essentially trick the body into thinking it's in a famine when it isn't. Um so the cell you know the the the the what the cells do is altered in the same way that it's altered in an actual famine. Uh but as I say yeah you can't get more benefit using these drugs which are called calorie restriction mimetics than what you can with calorie restriction itself. So it doesn't help. >> So what is the most important breakthrough that you you need to make? What is the one that you think about the most? Is there one that you think about? >> Well really this is what I want to do. I want to do a lot better than calorie restriction in mice. If we can find a cocktail of interventions of damage repair, rejuvenation interventions that we can put together in mice that will give maybe twice as much as uh calor restriction does, then that's going to completely change the world. Every expert is going to agree that we're in a different universe. they're going to say this really shows that animals with, you know, legs and fur and stuff, not like fruit flies, um can actually be um made to live a lot longer healthily um with things that we have at our disposal today. At the moment, what what that means in round numbers, I told you these mice typically live 2 and a half years. >> What we care about is to be able to treat people who are in our kind of age, right? So, we want to treat my kind of thing. So we start at one and a half years of age when they're already in middle age and if you start calorie restriction at that age you can get about four months life extension right so what we want to get is like 8 months or even 12 months and if we can then yeah I can retire basically my job will be done >> this because this leads us towards the escape velocity >> that's right >> interesting I wonder what it does to the the meaning of life like because in some ways death gives life a lot of meaning >> bollocks Yeah. No, it does. It does. >> Absolute bollock. >> I I think it I don't know. I think it does. I I think because I think >> Go on then. Persuade me. Give me an argument for why death gives meaning to life. >> So, I mean, I only speak personally. Um, but uh there's two parts of death that I think about. One is are you religious at all? You saw >> No, but keep going. >> Okay. So, I'm partly religious, right? I'm kind of I was uh born I was born as a Christian. Uh uh my raised as a Christian actually raised as a Catholic and and I've come back and forth with religion, but I definitely feel a spirituality. It's just it's just part of me. >> Um and I think a lot about the the morality that comes with religion and the potential judgment that comes with >> All right. Um, but if I'm not, if there's a chance I don't die, I don't have to think about >> Well, there is no chance you won't die. >> There's a chance, but there's a chance I could effectively live forever. I think about that differently. >> Well, effectively is a weasel word. What do you actually mean? I mean, if you die at age 100 or a thousand or a million and you still die and that's up to God. >> Well, yeah. And eventually the the sum will become a big >> Well, we might be able to get away before then, but yeah, heat, death of the universe, things like that. But I [clears throat] guess there are things that you you put off in life as well because you think, you know, when you're young, you think you're going to live forever and and then you get old and you like life changes >> and you know, have I lived a good life? You know, I've got a hundred years hopefully probably maybe 70 of which I've got now maybe I don't know 25 good years left. Like what am I going to do in that period? Like how am I going to be around my kids? and you know like have I lived a good life? >> So I mean I think about what I have done in life as well but it doesn't slow me down. It doesn't I'm not tempted to you know rest on my laurels. >> It's unwinding as I try and talk about it. >> I should hope so. The um I mean u I mean the point is yeah the um what do kids do when they um don't have don't see death? You know they think they think they're going to be living functionally forever. What do they do? They still have priorities. They still make decisions about what to do and what not to do just the same as older people do. And I don't think there's anything somehow inferior about the decision making that youngsters make relative to the decision making that people in the middle age make. >> All right, let's talk to you about my sponsor, Leen. Now, if you're borrowing against your Bitcoin, there's one thing that matters more than anything else. Is it actually safe? Well, Leen have just dropped their lowest rates ever. But more importantly, they haven't changed the thing that matters most. Your Bitcoin is never lent out. So, they're not chasing yield. There's no hidden risk and there's no rehypothecation. Just collateralized loans done properly. Now, they've done over $10 billion in loans. And they've done this through bull markets, through bare markets, and everything in between. And they've done it without ever losing any client assets. So, now you get lower rates, which means the bigger the loan, the lower the rate. and full transparency before you apply. There's no monthly payments or early repayment penalties. And they give you tools to stay in control from auto topups to LTV alerts. So, you're not choosing between a good rate and safety anymore because with Leen, you get both. Now, you can check out your rate using the calculator at leen.io/a. >> I think this whole thing's wild. I just 10 to 12 years, man. >> I said 12 to 15. Oh, 12 to 15. Yeah. Um, there's a lot of people working on this as well. I know Brian Brian Armstrong who uh built Coinbase, he's now working on this. And then we also have Mr. What do you make of Brian Armstrong? Uh, sorry. Um, Brian Johnson. >> I know Brian Johnson quite well. Um, >> he's an acquired taste of course, you know, he has chosen to be extremely outspoken in a rather unusual way about his uh desire not to die. Um, I think he would do better by writing me a last check. Uh, and I've told him so numerous times. Um, but you know, everyone's got their reasons to think the way they think. Um, Brian Johnson thinks that AI is going to be everybody's salvation fairly soon. Really, the only big issue is not to die of anything stupid anytime before then. Um, and of course he may be right, but he may be wrong. and I'd prefer not to take the risk. >> So, but is AI helping you with your work? >> It helps the field quite a lot. Yes. I myself in my foundation, we don't use AI to speak of, but plenty of the people we work with do. Um Yeah. So, so um many of the the ways in which AI is being leveraged to help um medical research are way outside of longevity. They cover everything. So the most conspicuous example of that is of course alpha fold which uh tells you what type of protein it's going to be if you give it the amino acid sequence and um you know people in in aging alpha fold all the time. This breakthrough I mentioned in uh restoring the stiffening of the extra matrix that was made with heavy use of alpha fold would been very much harder otherwise. Um >> so why aren't you using it in your foundation? Oh well, different areas of research need AI and different and other ones don't. You don't need AI to tell you whether a mouse is alive or dead, >> you know. Exactly. >> Well, so what's Brian Johnson's thesis then that AI? >> Well, well, I mean, basically he just thinks the singularity is coming, you know, and and you know, AI will solve everything and you just got to not die before then. >> Do you think it's coming? >> I have no idea. I mean, you know, the debate about whether AI will get out of control and whether it will kill us all before before we can stop it. You know, that debate has been around a long time. I've been friends with many of the people who have led that debate. >> Nate s >> um I don't know him, but Yan Talin, Nick Bostonramm, Anderson Sandberg, people like that. Um and uh of course it's you know it's it's an important question but at the end of the day the critical thing is that the expert community is well aware of the possibility and is you know doing better than I could do to stay on top of that that question and make it not happen. So then the question is what will AI do if it doesn't get out of control and kill us all? And it's pretty reasonable to suppose that AI will do a great deal in every walk of life to improve improve the human condition. Now um in aging the difficulty is that we've got so little data. We've got plenty of bi biological data on you know genomics and proteomics and so on but there are other types of data we don't have about like what works and what doesn't work and we just have to do the boring wet boring lab experiments you know just give mice this or that and see what happens or otherwise the AI doesn't have anywhere to start from >> so yeah people who say oh there's no point in writing or a check because um AI is going to do it all for us I think that I think that's misguided >> being someone who works worked in AI you said quite early on was it in in the 80s we >> yeah what do you make of the last three years >> well I've been following the growth of AI as we know it today from pretty much the time it began which is about 15 years ago >> okay >> uh what happened basically was that um >> well it it it all started with what honestly the field is rather ashamed of now which was one academic paper that came out um by a couple of very prominent researchers uh showing a mathematical result that showed that a certain type of system was incapable of doing a certain type of function. Um and basically that paper was over interpreted. It would assume that because this type of thing, this type of system could not do this function then other types of system even if they were a bit more sophisticated but they were broadly the same kind of system would also not be able to do interesting things. Um and so everyone basically completely gave up on this approach and went in the general direction that I was pursuing when I was working in AI which was called often called good old-fashioned AI >> old fashioned AI >> but basically what it meant was it was all very formal you know you were as I say you were writing programs that would prove mathematical theorem stuff like that. Um now the thing is that the um the the previous system which was basically called neural networks um was not 100% a banner by absolutely everybody. There were just a few professors around the world who you know doggedly kept going and um eventually it worked. One of them a guy named Jeffrey Hinton who of course got the Nobel Prize a year or two ago um he was working at Stanford. he decided to do a basically a hailmary experiment um with very large amounts of data essentially by hiring a post-doal fellow who knew how to program GPUs and um it worked and it worked like like far better than anyone had expected and people started trying variations on that theme and very rapidly you got the fundamentals of what you see today you know things like like someone else in his lab invented this thing called a generative adversarial network which is a one of the real you know um uh main stays of how AI works now but yeah so what's happened you know as time's gone on has been more or less no surprise since then since the fundamental breakthrough bit like what happened with aviation starting in 1905 or whatever right I mean you know people had been saying for a long time that they thought we had to fly probably eventually didn't think it was going to take 400 years from from when he was around. Um but eventually they got there but once they got the fundamental breakthrough then the refinements of that breakthrough that happened thereafter were you know happened at a very rapid and very steady pace you know so we had you know perfectly respectable aircraft flying in World War I um and it was it's been the same with AI over the past 15 years you know it's gone pretty much what what I would have expected the way that technologies tend to do when no additional fundamental breakthroughs are needed and you just basically need to carry on refining. >> Is there anything that's been unexpected >> for you? Have you been surprised or wowed? >> Well, I guess the main things that are surprising are when that curve of ref incremental refinements when there's a deviation from the curve when there's more of a step function. So, um, and that's more in the eye of the beholder. So let's go back to alpha fold again. So alpha fold this the the problem it was trying to solve um the shape of a protein. There had been a competition for that particular thing uh for decades run every two years. People should would submit their program say okay you know what what um uh what is uh the shape of these proteins and there was basically no progress at all for decades. Then Alpha Fall came along in its first iteration and it won the competition. But it didn't win it like in a completely earthshattering way. It just like was marginally better than what had been done before. And then two years later the version of Alpha Fog came back and it wiped the floor with everybody. Um so that was very sudden and I guess the sudden things are the surprising ones. What >> what about Agi? Do you think that will be achieved? What even mean? I still even don't actually know exactly what it means. >> Well, that's right. I mean we've now got a situation where the terminology is being refined a bit. So now people talk about ASI and AGI. AGI artificial general intelligence, ASI artificial super intelligence. Um you know basically defining distinguished in terms of how generally things are better than humans better [snorts] at things than humans. So in practice there may be not all that much distinction between these various definitions because in terms of timing because everything's moving so fast. Um but we will just have to see you know I mean I I I certainly think that for all practical purposes we will get to a point where everything that we would ever have thought [snorts] that we wanted a machine to do a robot or anything they will be able to do. >> And do you think it will kill us all? >> Hope not. You know, I mean, um, you know, all >> you do all this work, you get us living forever and then the robots come and kill us. >> Yeah. I mean, all one can say is, you know, there's a lot of very knowledgeable and very, you know, dedicated people who are working to stop that from happening. And um, I don't know whether those people are going to succeed, but I have, but I I have no better way to decide what I should do other than just sit back and trust them and get on with what I'm good at. >> Raise some money. and raise some money to do so. >> Is it that hard? >> Oh, it's insane. I mean, you know, >> we have a lot of small donors and we're very grateful to them, but the fact is this is still a minority pursuit and um therefore, you know, the the occasional big donor that comes along makes most of the difference. So, 20 years ago, Peter Thiel, I guess you've heard of him, he became my first big der. he was hooked on this. Saw my TED talk and well met me at TED actually and and took it from there and um and uh then it was another four years before the next big der came along and it was another six years before the one after that. So it's very very stocastic and hit and miss. Um but I'm constantly trying you know I'm always out there. The reason I do so many podcasts and so many talks, so many you know talks at conferences and so on is precisely to get the word out. Um you know to educate people to understand that we are within striking distance of doing this and therefore the sooner we do it the better and that's actually quite a good way to spend one's money. >> Well I I do I hope you do it man. Um is there anything I've not asked you to wish I had? >> I don't think so. You've been fairly wide ranging. >> Yeah. Look loved it man. Well look I hope you do it. Um, and uh, we sit down and do this again in another 300 years. >> That's That's the idea. [laughter] Nice idea. >> Good luck with it. Where do you want people to go? >> Um, I want people to go to levf.org, which is our website. Um, so you'll put it in the show notes. >> Yes, we'll put it in the show notes. >> Yeah, there is a nice friendly donate button. Um, but there's also, of course, all the information about what we do, what we plan to do, why we do it. Um, you know, all that. >> Well, I I again, I hope I hope I live to a thousand and we're back here and we do it again. Let's do that. If if we make it to a thousand, I'll celebrate my thousandth birthday with you doing this. >> All right. Cheers. Thank you to everyone for listening and thank you to everyone in San Francisco. Beautiful San Francisco. Yeah. Thank you, man. Good luck. Go crush it.