A Multilevel Plan to Live Forever
Plan A – Fighting Aging. Plan B – Cryonics. Plan C – Digital Immortality. Plan D – Big World Immortality. — Antimortality
The idea of life extension is gaining traction. Several methods for reaching what I prefer to call potentially indefinite life extension — I use it as a synonym for “immortality,” which carries religious connotations for some — have been suggested: cryonics, mind uploading, digital immortality. But most proponents support just one of these methods, and are convinced that their chosen one is the best and most effective way to live forever.
I think that is a mistake. The future is very uncertain, and we don’t know what will work in life extension, or when. It is easy to be overoptimistic or over-pessimistic, and our ability to predict — let alone influence — the speed of technological progress is limited. There is personal uncertainty too: some people live naturally until 90, others die young, and we all start at different ages, which changes our chances of surviving until the bright future.
To counter this, I want to lay out a different approach: a multilevel “death-prevention” plan. We can think about death as an enemy that attacks the fortress of our survival from many different sides. So we need to defend all the sides at once.
The idea of “multilevel defense” is borrowed from engineering safety — the way we keep nuclear plants safe. There, a failure on one level of defense means the next level starts to act: normal regulating rods, then emergency shutdown rods, then the containment building, then the exclusion zone. My survival strategy works the same way. The plans are presented in their order of implementation. If I have to start Plan B, it means Plan A has failed. But — and this is the crucial part — all the plans must be prepared simultaneously, in advance. You cannot scramble for cryonics on your deathbed if you never signed up.
Each plan also has two faces: a personal one — the things I do to raise my own odds of survival — and a social one — the collective work of building these technologies and making them available, since none of this happens if nobody works on it. And remarkably, all four plans converge on the same destination at the end: indefinite survival as an uploaded mind inside an ecosystem created by a benevolent superintelligent AI.
Plan A: Surviving Until AI
The main idea of Plan A is to surf the wave of technological progress long enough to reach what Aubrey de Grey called longevity escape velocity (LEV) — the moment in history when personal life expectancy grows by more than one year for each calendar year that passes. Two exponential processes are racing each other here: the growth of my personal probability of death from aging, and the exponential growth of medical and computer technologies that could extend my life.
Aging is the main killer — it causes, directly and indirectly, up to 95 percent of deaths in developed countries. But stopping aging is not enough. Even if you froze a person at the age of minimal risk — around ten years old, with no aging, good parental care, and little risky behavior — accidents alone would still cap life expectancy at a few thousand years. To make the risk of death actually approach zero, we need to combine two things: stop aging, and create mind uploading technologies. Once a mind lives on a digital carrier, it is easy to copy and back up, and it can exist as long as some future analogue of the Internet exists — which could be billions of years.
The Immortality Warrior
On the personal side, a person first has to make a decision to actively fight for life extension. Some will survive a long time through pure luck and genetics, or through high-quality care. But as our understanding of aging grows, the power of personal choices grows with it.
I would suggest the term immortality warrior for a person whose primary goal is reaching indefinite life extension, balancing egoistic and altruistic behavior to get there. He does not try to survive by lowering anyone else’s life expectancy — but he also does not completely ignore his own needs. The logic is straight out of decision theory: if everybody “defects” and pursues only selfish survival, the global project collapses for lack of cooperation. But if everybody neglects their own health while nobly helping others, collective survival fails too.
There are many psychological obstacles to becoming a perfect immortality warrior — anyone who has tried to quit smoking or exercise every day knows this. A good warrior is rational, takes care of his mental health, and works on his own debiasing. And he should care above all about preventing damage to his own brain, because a brain that ages loses its ability to adapt and to rationally seek out the new technologies that will save him.
Low-Hanging Fruit, Then Aging
The warrior should start with the low-hanging fruit: getting rid of dangerous addictions — smoking, drinking, overeating — and investing in prophylaxis through regular check-ups, controlling chronic conditions like type-2 diabetes and hypertension, and vaccination. He can even calculate how many millimorts he prevents by installing a carbon-monoxide detector, not riding a motorbike, wearing reflective clothes at night, and fastening his seat belt.
Once those are handled, survival depends mostly on aging and genetics. As of writing, the available tools to slow aging include a fasting-mimicking diet, sport (slow running and dancing seem best), hot bathing, and a handful of generally safe geroprotectors that should be evaluated individually, like vitamin D and metformin. None of this is dramatic: all of these low-tech interventions together may buy maybe ten years. But even a small gain matters enormously, because it means hundreds of millions of people survive until the next technology arrives — and ultimately until immortality itself. We are not talking about merely healthy aging. We are talking about a lottery ticket where the prize is millions of years of future life.
Putting Yourself in the Best Place
The best chance of survival is to be an early adopter of new life-extension technologies the moment they appear and the risks look low. Several things help: knowledge of the field; actually creating the technology yourself, as a scientist or investor; participating in clinical trials, which may give a ten-year head start at the cost of some risk; living in a country with high life expectancy and advanced science; funding; and being part of a community of like-minded people for access, feedback, and motivation. One caution on biohacking: responsible self-experimentation should avoid self-replicating entities like gene-delivery viruses, which carry a real risk of causing a global catastrophe, and concentrate instead on data collection and non-replicating experiments.
The Technological Ladder
Behind the personal strategy sits a long technological ladder for the whole civilization, presented in logical sequence much like the space program went from satellite to animal to human to the Moon. The early rungs are simpler, cheaper, and safer but give smaller gains; later rungs are more powerful but harder. Briefly, the steps run from a weak slowdown of aging (geroprotectors, diet, removing risk factors), to a strong slowdown based on real understanding of its mechanisms (gene delivery, stem-cell and microbiome replacement, clearing senescent cells), to negligible senescence (a “vaccine against aging,” SENS-style repair), to genuine rejuvenation (Yamanaka factors, signals like Klotho), to a new body (a head on life support, a cyborg, a cloned biological body, or a nanotechnological one), to cyborgization of the brain, and finally to mind uploading itself. The social side, meanwhile, is mostly about money and law. I estimate that roughly $100 billion a year could be usefully spent on honest longevity research — which sounds enormous until you learn the world already spends around $300 billion a year on unproven anti-aging snake oil. The other obstacles are an FDA that still does not recognize aging as a disease, and a general human — often religious — unwillingness to live longer at all.
Plan B: Cryonics
The honest truth is that the chance of surviving until 2100 is not high, even without any global catastrophe, and it depends heavily on your current age. Most people alive today will probably not make it. But biological death is not information-theoretical death. We can preserve the human brain in the hope that future AI and nanotechnology will be able to bring the person back.
Cryonics — cooling the brain to very low temperature while preventing ice formation, in a process called vitrification — is currently the only commercially available option. And it is surprisingly unpopular. Only around 250 people are cryopreserved, across three facilities, two in the US and one in Russia. To put that unpopularity in perspective: more bodies are eaten by birds each year under the Zoroastrian funeral tradition in Mumbai. Because so few sign up, prices stay high, the legal status remains murky almost everywhere outside Arizona, and research stays thin.
The personal strategy is refreshingly straightforward. Sign up for cryonics — you can fund it through life insurance, so it is far cheaper than people assume. Make a will with clear instructions. Wear a cryotoken so that strangers know what to do if you die unexpectedly. Have a trusted relative ready to organize the process. And as you get older, live near a cryofacility and contact it immediately in any life-threatening situation. On the social side, the single biggest need is for legal stability: ideally, suspended bodies would have the same legal protection as a person in a coma, so that destroying them would count as murder.
Plan C: Digital Immortality
Cryonics can fail too. Maybe I won’t be preserved in time, maybe my brain will thaw, maybe crucial information will be lost. Given all the uncertainty, I put the chance of cryonics working at around 10 percent. So there needs to be a further layer of defense.
By digital immortality I mean something specific: collecting data about a person now, in the hope that a future superintelligent AI will be able to reconstruct a working model of that person from it. This runs straight into the notoriously hard “personal identity problem” — we don’t know how much fidelity such a model needs, or whether the reconstruction would really be me even if it were conscious. Since we cannot solve that now, I take a deliberately conservative approach: save as much information as possible, and hope that something in it turns out to be enough.
Not all data is equal, though. The information worth saving should be unique (the fact that I have two legs says nothing; my name is unique), valuable (connected to my actual preferences — my credit-card PIN is important but is not part of my personality), and predictive (it should actually shape my future behavior). There are two ways to gather it: passively, by recording everything you do, and actively — spending a focused month or two writing your life story, archiving your social networks, photographing your documents and home, running psychological tests, recording EEG, and capturing yourself in different modes of behavior, from public talks to fun with friends.
Digital immortality is Plan C rather than Plan B because being reconstructed from a real cryopreserved brain is better than being rebuilt from a mere model. But the two help each other — information lost in cryopreservation might be restored from the digital record. And the endpoint is identical to the other plans: a human mind uploaded into a computer overseen by an advanced AI.
Plan D: Big World Immortality
Plan D is a collection of ideas in which immortality appears with no conscious action on our part — but which all have very low probability. For religious people, this is actually Plan A: they are already certain that an afterlife exists. For me it cannot be, because the evidence is thin and the concept has real theoretical problems. Still, there are a few more rational versions worth keeping in reserve, none of which require a soul or a God.
The best known is “quantum immortality,” built on the many-worlds interpretation of quantum mechanics: at the moment of death there is always a branch in which I did not die. It is deeply controversial — we don’t know if the universe is truly infinite, whether distant copies count as me, or what to do about the shrinking “share of existence.” Worse, raw quantum immortality could mean surviving as an infinitely old, barely conscious, suffering being. But it pairs surprisingly well with cryonics, which tilts the most likely surviving branch toward a good outcome rather than a horrible one.
Other Plan D ideas include resurrection by a future superintelligent AI — perhaps reconstructing the dead from traces of the past, or via quantum randomness in a vast simulation; resurrection by an alien civilization that visits Earth long after we are gone, much as we now contemplate reviving mammoths; and the simulation hypothesis itself — if we already live inside a simulation, the machine running it can record and restore any mind. And if a soul does exist, in practice that is not so different from a superintelligence running our world; either way the observational consequences would be the same.
Which Plan Is Yours?
A map is useless if you don’t know where you are standing on it. Different people should travel through this landscape differently.
The biggest factor is age. The younger you are, the better your odds of surviving until radical life extension, which makes Plan A the natural choice. Children today already have a life expectancy of around 80, so they may reach longevity escape velocity with no special measures at all, beyond avoiding risky activities and taking reasonable care of their health. For older people the calculus flips: the chance of surviving until the therapies arrive is small, so cryonics moves up the priority list, and it becomes time to write those memoirs for Plan C.
The second factor is money, though less than people think. Life expectancy rises only slowly with wealth — Indian miners live just a few years less than American billionaires. And despite the popular belief, life-extension therapies will not be reserved for the super-rich: to be reliable, they must be tested on thousands of people, which forces them to be relatively cheap. The rich have one real advantage — they can fund the research and buy early access — but they also face the field’s very high concentration of scammers. For an old and poor person, the most realistic plan is C: collecting data about yourself, which may need nothing more than a computer with a camera, or even just paper and a pen.
Where It All Converges
If I am eventually uploaded, I expect to live as an emulation — an “em” — inside a large ecosystem of other minds, overseen by a superintelligent AI. Such a mind could be loaded back into a robotic body to act in the real world, enjoying an effectively indestructible existence with a constantly updated backup that allows immediate resurrection if the body is destroyed.
That is the whole point of treating these as levels of a single defense rather than rival theories. No one plan is guaranteed. But by combining them in the right order — fighting aging while you can, cryonics if that fails, digital immortality beneath that, and the strange comforts of big-world immortality at the very bottom — and by preparing all of them at once, a person can give themselves the maximum chance at living, in some form, indefinitely. And none of it works alone: indefinite life extension cannot be reached without cooperation among many people. The fortress has many walls, and we have to defend every one of them — together.
Based on my article https://philpapers.org/rec/TURMSF-2
The roadmap: http://immortality-roadmap.com/IMMORTEN12.jpg
The idea of life extension is gaining traction. Several methods for reaching what I prefer to call potentially indefinite life extension — I use it as a synonym for “immortality,” which carries religious connotations for some — have been suggested: cryonics, mind uploading, digital immortality. But most proponents support just one of these methods, and are convinced that their chosen one is the best and most effective way to live forever.
I think that is a mistake. The future is very uncertain, and we don’t know what will work in life extension, or when. It is easy to be overoptimistic or over-pessimistic, and our ability to predict — let alone influence — the speed of technological progress is limited. There is personal uncertainty too: some people live naturally until 90, others die young, and we all start at different ages, which changes our chances of surviving until the bright future.
To counter this, I want to lay out a different approach: a multilevel “death-prevention” plan. We can think about death as an enemy that attacks the fortress of our survival from many different sides. So we need to defend all the sides at once.
The idea of “multilevel defense” is borrowed from engineering safety — the way we keep nuclear plants safe. There, a failure on one level of defense means the next level starts to act: normal regulating rods, then emergency shutdown rods, then the containment building, then the exclusion zone. My survival strategy works the same way. The plans are presented in their order of implementation. If I have to start Plan B, it means Plan A has failed. But — and this is the crucial part — all the plans must be prepared simultaneously, in advance. You cannot scramble for cryonics on your deathbed if you never signed up.
Each plan also has two faces: a personal one — the things I do to raise my own odds of survival — and a social one — the collective work of building these technologies and making them available, since none of this happens if nobody works on it. And remarkably, all four plans converge on the same destination at the end: indefinite survival as an uploaded mind inside an ecosystem created by a benevolent superintelligent AI.
Plan A: Surviving Until AI
The main idea of Plan A is to surf the wave of technological progress long enough to reach what Aubrey de Grey called longevity escape velocity (LEV) — the moment in history when personal life expectancy grows by more than one year for each calendar year that passes. Two exponential processes are racing each other here: the growth of my personal probability of death from aging, and the exponential growth of medical and computer technologies that could extend my life.
Aging is the main killer — it causes, directly and indirectly, up to 95 percent of deaths in developed countries. But stopping aging is not enough. Even if you froze a person at the age of minimal risk — around ten years old, with no aging, good parental care, and little risky behavior — accidents alone would still cap life expectancy at a few thousand years. To make the risk of death actually approach zero, we need to combine two things: stop aging, and create mind uploading technologies. Once a mind lives on a digital carrier, it is easy to copy and back up, and it can exist as long as some future analogue of the Internet exists — which could be billions of years.
The Immortality Warrior
On the personal side, a person first has to make a decision to actively fight for life extension. Some will survive a long time through pure luck and genetics, or through high-quality care. But as our understanding of aging grows, the power of personal choices grows with it.
I would suggest the term immortality warrior for a person whose primary goal is reaching indefinite life extension, balancing egoistic and altruistic behavior to get there. He does not try to survive by lowering anyone else’s life expectancy — but he also does not completely ignore his own needs. The logic is straight out of decision theory: if everybody “defects” and pursues only selfish survival, the global project collapses for lack of cooperation. But if everybody neglects their own health while nobly helping others, collective survival fails too.
There are many psychological obstacles to becoming a perfect immortality warrior — anyone who has tried to quit smoking or exercise every day knows this. A good warrior is rational, takes care of his mental health, and works on his own debiasing. And he should care above all about preventing damage to his own brain, because a brain that ages loses its ability to adapt and to rationally seek out the new technologies that will save him.
Low-Hanging Fruit, Then Aging
The warrior should start with the low-hanging fruit: getting rid of dangerous addictions — smoking, drinking, overeating — and investing in prophylaxis through regular check-ups, controlling chronic conditions like type-2 diabetes and hypertension, and vaccination. He can even calculate how many millimorts he prevents by installing a carbon-monoxide detector, not riding a motorbike, wearing reflective clothes at night, and fastening his seat belt.
Once those are handled, survival depends mostly on aging and genetics. As of writing, the available tools to slow aging include a fasting-mimicking diet, sport (slow running and dancing seem best), hot bathing, and a handful of generally safe geroprotectors that should be evaluated individually, like vitamin D and metformin. None of this is dramatic: all of these low-tech interventions together may buy maybe ten years. But even a small gain matters enormously, because it means hundreds of millions of people survive until the next technology arrives — and ultimately until immortality itself. We are not talking about merely healthy aging. We are talking about a lottery ticket where the prize is millions of years of future life.
Putting Yourself in the Best Place
The best chance of survival is to be an early adopter of new life-extension technologies the moment they appear and the risks look low. Several things help: knowledge of the field; actually creating the technology yourself, as a scientist or investor; participating in clinical trials, which may give a ten-year head start at the cost of some risk; living in a country with high life expectancy and advanced science; funding; and being part of a community of like-minded people for access, feedback, and motivation. One caution on biohacking: responsible self-experimentation should avoid self-replicating entities like gene-delivery viruses, which carry a real risk of causing a global catastrophe, and concentrate instead on data collection and non-replicating experiments.
The Technological Ladder
Behind the personal strategy sits a long technological ladder for the whole civilization, presented in logical sequence much like the space program went from satellite to animal to human to the Moon. The early rungs are simpler, cheaper, and safer but give smaller gains; later rungs are more powerful but harder. Briefly, the steps run from a weak slowdown of aging (geroprotectors, diet, removing risk factors), to a strong slowdown based on real understanding of its mechanisms (gene delivery, stem-cell and microbiome replacement, clearing senescent cells), to negligible senescence (a “vaccine against aging,” SENS-style repair), to genuine rejuvenation (Yamanaka factors, signals like Klotho), to a new body (a head on life support, a cyborg, a cloned biological body, or a nanotechnological one), to cyborgization of the brain, and finally to mind uploading itself. The social side, meanwhile, is mostly about money and law. I estimate that roughly $100 billion a year could be usefully spent on honest longevity research — which sounds enormous until you learn the world already spends around $300 billion a year on unproven anti-aging snake oil. The other obstacles are an FDA that still does not recognize aging as a disease, and a general human — often religious — unwillingness to live longer at all.
Plan B: Cryonics
The honest truth is that the chance of surviving until 2100 is not high, even without any global catastrophe, and it depends heavily on your current age. Most people alive today will probably not make it. But biological death is not information-theoretical death. We can preserve the human brain in the hope that future AI and nanotechnology will be able to bring the person back.
Cryonics — cooling the brain to very low temperature while preventing ice formation, in a process called vitrification — is currently the only commercially available option. And it is surprisingly unpopular. Only around 250 people are cryopreserved, across three facilities, two in the US and one in Russia. To put that unpopularity in perspective: more bodies are eaten by birds each year under the Zoroastrian funeral tradition in Mumbai. Because so few sign up, prices stay high, the legal status remains murky almost everywhere outside Arizona, and research stays thin.
The personal strategy is refreshingly straightforward. Sign up for cryonics — you can fund it through life insurance, so it is far cheaper than people assume. Make a will with clear instructions. Wear a cryotoken so that strangers know what to do if you die unexpectedly. Have a trusted relative ready to organize the process. And as you get older, live near a cryofacility and contact it immediately in any life-threatening situation. On the social side, the single biggest need is for legal stability: ideally, suspended bodies would have the same legal protection as a person in a coma, so that destroying them would count as murder.
Plan C: Digital Immortality
Cryonics can fail too. Maybe I won’t be preserved in time, maybe my brain will thaw, maybe crucial information will be lost. Given all the uncertainty, I put the chance of cryonics working at around 10 percent. So there needs to be a further layer of defense.
By digital immortality I mean something specific: collecting data about a person now, in the hope that a future superintelligent AI will be able to reconstruct a working model of that person from it. This runs straight into the notoriously hard “personal identity problem” — we don’t know how much fidelity such a model needs, or whether the reconstruction would really be me even if it were conscious. Since we cannot solve that now, I take a deliberately conservative approach: save as much information as possible, and hope that something in it turns out to be enough.
Not all data is equal, though. The information worth saving should be unique (the fact that I have two legs says nothing; my name is unique), valuable (connected to my actual preferences — my credit-card PIN is important but is not part of my personality), and predictive (it should actually shape my future behavior). There are two ways to gather it: passively, by recording everything you do, and actively — spending a focused month or two writing your life story, archiving your social networks, photographing your documents and home, running psychological tests, recording EEG, and capturing yourself in different modes of behavior, from public talks to fun with friends.
Digital immortality is Plan C rather than Plan B because being reconstructed from a real cryopreserved brain is better than being rebuilt from a mere model. But the two help each other — information lost in cryopreservation might be restored from the digital record. And the endpoint is identical to the other plans: a human mind uploaded into a computer overseen by an advanced AI.
Plan D: Big World Immortality
Plan D is a collection of ideas in which immortality appears with no conscious action on our part — but which all have very low probability. For religious people, this is actually Plan A: they are already certain that an afterlife exists. For me it cannot be, because the evidence is thin and the concept has real theoretical problems. Still, there are a few more rational versions worth keeping in reserve, none of which require a soul or a God.
The best known is “quantum immortality,” built on the many-worlds interpretation of quantum mechanics: at the moment of death there is always a branch in which I did not die. It is deeply controversial — we don’t know if the universe is truly infinite, whether distant copies count as me, or what to do about the shrinking “share of existence.” Worse, raw quantum immortality could mean surviving as an infinitely old, barely conscious, suffering being. But it pairs surprisingly well with cryonics, which tilts the most likely surviving branch toward a good outcome rather than a horrible one.
Other Plan D ideas include resurrection by a future superintelligent AI — perhaps reconstructing the dead from traces of the past, or via quantum randomness in a vast simulation; resurrection by an alien civilization that visits Earth long after we are gone, much as we now contemplate reviving mammoths; and the simulation hypothesis itself — if we already live inside a simulation, the machine running it can record and restore any mind. And if a soul does exist, in practice that is not so different from a superintelligence running our world; either way the observational consequences would be the same.
Which Plan Is Yours?
A map is useless if you don’t know where you are standing on it. Different people should travel through this landscape differently.
The biggest factor is age. The younger you are, the better your odds of surviving until radical life extension, which makes Plan A the natural choice. Children today already have a life expectancy of around 80, so they may reach longevity escape velocity with no special measures at all, beyond avoiding risky activities and taking reasonable care of their health. For older people the calculus flips: the chance of surviving until the therapies arrive is small, so cryonics moves up the priority list, and it becomes time to write those memoirs for Plan C.
The second factor is money, though less than people think. Life expectancy rises only slowly with wealth — Indian miners live just a few years less than American billionaires. And despite the popular belief, life-extension therapies will not be reserved for the super-rich: to be reliable, they must be tested on thousands of people, which forces them to be relatively cheap. The rich have one real advantage — they can fund the research and buy early access — but they also face the field’s very high concentration of scammers. For an old and poor person, the most realistic plan is C: collecting data about yourself, which may need nothing more than a computer with a camera, or even just paper and a pen.
Where It All Converges
If I am eventually uploaded, I expect to live as an emulation — an “em” — inside a large ecosystem of other minds, overseen by a superintelligent AI. Such a mind could be loaded back into a robotic body to act in the real world, enjoying an effectively indestructible existence with a constantly updated backup that allows immediate resurrection if the body is destroyed.
That is the whole point of treating these as levels of a single defense rather than rival theories. No one plan is guaranteed. But by combining them in the right order — fighting aging while you can, cryonics if that fails, digital immortality beneath that, and the strange comforts of big-world immortality at the very bottom — and by preparing all of them at once, a person can give themselves the maximum chance at living, in some form, indefinitely. And none of it works alone: indefinite life extension cannot be reached without cooperation among many people. The fortress has many walls, and we have to defend every one of them — together.