How many separate interventions will be needed to solve aging? And what does evolution imply regarding that question?

The complications of aging

By some estimates, the human body contains 12 different biological systems, 78 different organs, 135 metabolic pathways, almost 200 different types of cell, and over 17,000 different proteins.

That’s a lot that can go wrong!

Indeed, as we age, more and more of these biological units develop all sorts of problems. Our minds and bodies increasingly lose the vitality and resilience of our youth. Dysfunction in lower-level biological units leads to visible sicknesses and decline in higher level units, and, ultimately, to death.

Whenever we humans reflected on this sorry trend of general deterioration in our bodies as we age, it was natural for us to wonder, also, what it would take to halt that trend.

In bygone days, some fanciful tales imagined that a single command from a powerful deity could put an end to our aging. Or that a life lived in some kind of cosmic harmony or absolute moral purity would allow a yogi or saint to escape the normal ravages of the passage of time.

In more recent times, the pendulum has swung to the other extreme. Forget any ideas that a single intervention might reverse all aspects of aging, say these critics. Moreover, forget any ideas that a series of different interventions might combine to do the trick. That’s because aging is overwhelmingly complicated.

Want to modify a metabolic pathway to reverse a given type of biological damage? That will surely have adverse side-effects, say the doubters. Each gene in our DNA is involved in multiple different activities; a genetic change to avoid one kind of damage in one subsystem will surely cause new kinds of damage in other subsystems. And so on.

In this view, the comprehensive solution to aging lies beyond human ability. There are far too many moving parts for any series of interventions to have a lasting beneficial effect. Therefore, we should give up any hope of curing aging.

1, n or ∞

Given the above context, I propose a three-way classification of all theories for how aging can be solved. The three categories of theory can be called singular, plural, and infinite.

Theories in the infinite category despair about the possibility of humans being able to subdue all the many areas of damage that occur throughout the body over time. In this view, if some types of damage are reduced, it will be at the cost of harming other aspects of human biology. Trying to avoid both sets of harm simultaneously will just result in yet other aspects of our metabolism being pushed over the edge. And so on.

These theories split further into two:

• Fans of the idea of a forthcoming beneficial artificial superintelligence (ASI) expect that, what humans are unable to accomplish, will nevertheless be within the extraordinary capabilities of an ASI. In this case, the solution to aging requires accelerating the advent of such an ASI
• There’s an even more pessimistic subclass, which holds that aging is so difficult that not even an ASI could solve it; it would be like expecting an ASI to move a spaceship faster than the speed of light.

Theories in the plural category maintain, instead, that a careful mixture of different damage repair interventions will in due course succeed in undoing biological damage throughout the entirety of the human body – before that damage causes serious harm – without resulting in unchecked new damage elsewhere in the body.

Plural theories are sometimes described as an engineering approach. It’s similar to the task of building a large suspension bridge: you need to get lots of separate things right, or the bridge will fall down, or sink under its own weight, or sway catastrophically when certain winds blow. It’s similar to the task of heavier-than-air powered flight: you need to solve the subtasks of take-off, steering, landing, and a sufficiently light engine delivering adequate thrust. Yes, you need to get lots of separate things right, with wise choices between trade-offs, but that’s what the discipline of engineering enables.

Thus, you may hear about a grand project to “solve the seven deadly sins of aging”, referring to seven named types of biological damage at the molecular, cellular, and extracellular levels. Or about projects to solve each of the 12 ‘hallmarks of aging’.

Theories in the singular category claim there is a greater unity in the phenomenon of aging. Despite the appearance of a large variety of different sorts of age-related biological damage, these theories propose that there is a single underlying mechanism (for example, an epigenetic clock), which, if altered, would be sufficient to reverse and solve aging.

If that is true, it sounds like good news, for those of us who wish our loved ones (and ourselves) to be able to keep on living with high vitality for many decades (perhaps even centuries) to come.

But what grounds are there for believing such an idea? That question takes us to a study of comparative evolution.

Biology and indefinite youth

We’re used to seeing creatures grow old and frail. That’s what happens to our pet cats and dogs, to horses, to mice and rats in laboratories, and, yes, to us humans.

In fact there’s a regular pattern to the growth of this frailty. If frailty is measured by the likelihood of an animal dying from any cause within the next fixed period of time (for example, in the next year), this frailty rises exponentially as the animal becomes older. Thus humans aged 60 are about ten times as likely to die in the next 12 months as they were at age 35. And humans at age 85 are ten times more likely again. (These three probabilities are roughly one in a thousand, one in a hundred, and one in ten.)

This pattern has a name: the Gompertz law of mortality.

Intuitively, the law seems to make sense:

• As damage spreads and deepens throughout the body, part of what breaks down are the mechanisms such as the immune system and stem cells that would normally help repair other types of damage
• The greater the damage throughout the body, the more vulnerable the body becomes to external shocks and strains – such as infections and injuries.

But wait. What’s true for cats, dogs, horses, mice, rats, and humans – as well as large numbers of other species – is not true for every species.

One valuable source of information on that point is the 2014 Nature article by Owen Jones, James Vaupel, and collaborators, ‘Diversity of ageing across the tree of life’. The article reviewed how mortality (frailty) increased with age, and also how fertility changed. This is from the article’s abstract:

Although it has been predicted that evolution should inevitably lead to increasing mortality and declining fertility with age after maturity, there is great variation among these species, including increasing, constant, decreasing, humped and bowed trajectories for both long- and short-lived species.

The different curves can be seen in an image included in their article. In each graph, the red lines show mortality at different ages, whereas the blue lines show fertility rates:

Some individual birds illustrate the same possibility. One example is the albatross known as “Wisdom”, who was given a tag on her leg in 1956 when she was already estimated as being around five years old. She has been observed many times in recent decades, with no apparent drop in her fitness, nor in her fertility. Here’s a photo and an extract from a Facebook posting by the Pacific Islands U.S. Fish and Wildlife Service dated 27th November 2024:

“Wisdom returns to Midway Atoll National Wildlife Refuge bringing more Thanksgiving joy to the Midway Atoll staff who celebrate witnessing Wisdom reaching at least 74 years old this coming winter. Wisdom, a mōlī (Laysan albatross), is the world’s oldest known, banded bird.”

Note the tag “Z333” on her leg in the photo. And note the egg that she has laid. (There’s a video here, taken by service volunteer Dan Rapp.)

Colonies of naked mole-rats display the same phenomenon: no sign of any increase in mortality with age, or of any decline in fertility. This was the headline in an article by a number of researchers at Calico: “Naked mole-rat mortality rates defy Gompertzian laws by not increasing with age”.

This graphic from the article compares the mortality curves for four different species: horses, mice, humans, and naked mole-rats:

The conclusion – which could be bolstered by references to many other animal species – is that nature itself appears to have found ways to avoid any exponential increase in mortality.

All that we humans need to do, therefore, is to find out the biological secrets of these species, and use that knowledge to create interventions that have similar effects in our own bodies.

The reversal of aging is nigh. Right?

Four criticisms and two answers

Not so fast, respond advocates of the ‘infinite’ schools of thought. Reversing aging isn’t that easy.

Four objections can be placed against the line of reasoning in the previous section:

1. Animals such as naked mole-rats and albatrosses may eventually manifest an exponential increase in mortality, but so far, the experiments have only revealed the early (apparently linear, or even flat) portion of the curve
2. Even if their mortality doesn’t increase exponentially, it is likely to increase linearly, due to aspects such as an observed epigenetic drift within the genome
3. It may not be possible to transfer the damage-repair aspects of the biology of these species to humans, without losing aspects of human biology that are fundamentally important to us
4. Even if such a transfer is possible in theory, in practice it may require an endless sequence of refinements, adjustments, and iterations – taking us back into the territory of infinite difficulty.

There are two ways to counter these criticisms. The first way leads to the singular group of theories of aging, and the second to the plural group.

The main difference between the two answers is in the assessment of the capabilities of evolution:

• The singular view is that evolution could have created humans that don’t age, but didn’t do so, because evolution was optimizing for outcomes other than extreme individual longevity
• The plural view is that evolution was fundamentally constrained; it wasn’t able to create humans that don’t age
• The singular view implies that biological systems might be coaxed into comprehensively regenerating themselves via their intrinsic capabilities
• The plural view is that the intrinsic regenerative capabilities of biological systems are limited: they will likely need to be augmented by a plurality of different damage-repair interventions.

 Let’s look more closely at this difference of opinion.

The capabilities of evolution

What creatures do we see around us in nature? Those which have inherited attributes that made their ancestors sufficiently fit to survive for at least a period of time in their environment, so that they were able to pass on their characteristics to a new generation which could in turn reach maturity and repeat the process.

The more suited a creature is to these tasks – survival and reproduction – the greater the likelihood that its descendants will grow in number in subsequent generations, out-competing other creatures which are less fit.

What does this imply about the likelihood of creatures that have longer lifespans?

Other things being equal, longer lifespans mean more chances to have offspring, and therefore greater numbers of descendants which are similarly long-lived. This suggests that evolution should produce ever greater numbers of increasingly longer-lived creatures.

However, other things are not equal. A creature that continues to have new offspring at a fast pace will deplete the resources it could apply to other tasks:

• Looking after earlier offspring that are still young and would benefit from parental support
• Spending energy to repair damage that has accumulated in its own body.

This in turn suggests that evolution should produce ever greater numbers of creatures that are long-lived and which space out their offspring, so that they can attend to other tasks as well as creating and taking care of offspring.

Indeed, we do see examples of species with these properties, including elephants and whales. Humans living in conditions similar to those in which we evolved typically space children 2-3 years apart. Birds that lay eggs only once every 1-2 years, such as albatrosses like Wisdom, follow this pattern as well.

But not all species follow this pattern of long lives and large birth spacing. Far from it! Other factors affect the survive-and-reproduce game:

• A large set of descendants can survive without their patriarch or matriarch having a long life
• Species which cut corners on damage repair mechanisms inside their own bodies can pour the spare energy into fecundity; this may well give them more descendants than species which reproduce at a slower pace
• Even if long-lived members of a species remain healthy as they age, they still die from time to time from causes unrelated to aging, such as predation, starvation, accident, or a deadly infection
• Evolution needs to provide species with the ability to adapt to new types of predation, accidents, climate, and so on, as environmental conditions change
• Rather than all creatures in a species being near-identical clones of a single ancestor, with limited adaptability, there are advantages to the species in having a mixed repertoire of biological capabilities – that is, without the species being dominated by long-lived ancestors.

This leads to a famous result, first stated in a 1951 lecture by biologist Peter Medawar: evolution has less concern over a creature once it has already been able to reproduce and create some offspring. At this point, for the long-term thriving of that line of animals, it may be better to apply resources to strengthen young animals, even if this increases the frailty of older animals.

This argument is a straightforward consequence of the principle that biological resources are limited. There are many aspects of these limitations:

• Genes that have beneficial effects early in life (for example, to accelerate growth) may have detrimental effects later in life (when growth hormones are no longer needed, and excessive growth can form cancerous tumours)
• Cells that carry numerous regulatory genes, that disable other genes at different points in a lifecycle, are taking space from genes that could be more useful in other ways
• Cells with more capabilities need larger genomes and therefore more effort to copy and divide themselves
• Food that is eaten by older members of a population reduces the amount of food available to younger, more varied members of that population
• Generalists aren’t specialists: animals that are capable of surviving in a wide variety of different environments are likely to be less suited to particular environments than ones that are specially optimized for those environments.

Accordingly, evolution is likely to favour species that produce sufficient variety in each new generation, and where older members of the population die off, leaving resources for the most capable members of new generations.

To summarise: producing very-long lived animals isn’t free; it is often a better evolutionary strategy to promote variety and to pass the torch from each generation to the next.

Intrinsic rejuvenation capabilities

Despite what I’ve just outlined, it seems that evolution has provided biological systems with a number of intrinsic rejuvenation capabilities:

• Some creatures, such as the axolotl (an amphibian) and the zebra fish, have the ability to regrow many parts of their body if they are damaged; planarian worms can even regenerate their entire body from a small portion
• The cells that will form the next generation – so-called germ cells – are specially protected against damage, and cells in the embryo have their epigenetic age reset to zero
• The telomeres at the ends of chromosomes shorten with age, but this can be reversed by an enzyme called telomerase
• Transcription factors discovered by Nobel Prize winner Shinya Yamanaka are able to reduce the epigenetic age of cells; other transcription factors with similar properties have been identified more recently
• The so-called ‘immortal jellyfish – turritopsis dohrnii – is capable of reverting to a completely healthy earlier stage of its life, as if a butterfly could revert to being a caterpillar
• Worker bees, which typically have much shorter lives than the queen of the hive (despite having the same genome), can have their remaining lifespan significantly increased in the event that the hive becomes queenless and a worker bee starts to lay eggs in place of the queen.

This prompts the question: why are these capabilities used only in limited cases? Why can’t more creatures regenerate limbs or organs? Why isn’t the epigenetic age of somatic (body) cells reversed from time to time, rather than allowing damage to accumulate there? Why isn’t the enzyme telomerase applied more regularly, to prevent telomeres shrinking to the point where cell division is no longer possible?

As before, there are two answers:

• The singular view is that evolution chose (in a meaningful sense of that word) to be sparing in its use of these intrinsic mechanisms, optimizing the success of the collective set of descendants, rather than the longevity of individuals
• The plural view is that evolution was fundamentally constrained by a cascade of trade-off considerations; for example, applying telomerase more widely could result in more cancer, and likewise in the case of reversing the epigenetic ages of cells
• The singular view is that humans can now choose differently from evolution, and can safely trigger these innate rejuvenation mechanisms
• The plural view is that each such mechanism may well be part of the solution to aging, but is unlikely to provide a complete solution.

The singular view: for and against

It’s time to recap. Creatures in most animal species age, usually with an accelerated rate of frailty/mortality, but there are exceptions – species that manifest what has been called ‘negligible senescence’. Moreover, biology has a bag of tools that partially or completely reverse damage – but it makes surprisingly rare use of these tools.

What I am calling the singular view asserts that it should be relatively easy to use these tools much more widely, so that humans can have negligible senescence too. In this view, the triggers for these tools lie fairly close to the surface of the existing network of biological pathways that operate in humans. We won’t need to extensively re-engineer these pathways.

There are many different theories within the singular view. They all believe there is one key trigger that starts the required regenerative processes which rejuvenate the entire body into a youthful state. They differ on what they think this trigger is. Examples include:

• Reprogramming the “signalome” – the set of biochemicals which convey information between different parts of the body – possibly by introducing new exosomes
• Reinvigorating the mitochondria within the body, allowing the body to use more energy on other tasks of repair and regeneration
• Spreading more telomerase around the body, thereby enabling more cells to multiply, as needed for various repair or regeneration tasks
• Strengthening the body’s CAP – the Cholinergic Anti-Inflammatory Pathway
• Applying the Yamanaka transcription factors (or a different set of transcription factors) throughout the body, to reduce the epigenetic age of cells, giving them a new lease of life.

These theories all have the advantage of a degree of conceptual simplicity, even though the practical details of implementing the desired triggers may require a lot of research. This conceptual simplicity may help attract funding for these research projects.

However, any suggestion that these mechanisms are already within the reach of biology has to answer a strong objection: why didn’t random mutations in the biology of individuals take place, to push these individuals into this state of negligible senescence? These mutants and their descendants would have a comparative advantage over other members of their population. Over time, the negligible senescence would have spread throughout the entire population.

A supporter of the singular view may reply: that sequence of events may have happened on some occasions, but the excessively long-lived animals would have crowded out younger variants, thereby limiting the emergence of the greater genetic diversity needed for the long-term survival of any species. Accordingly, those species would have tended to become extinct on account of lack of variety and adaptability.

Indeed, consider the possibility that, deep in the biological past, evolution stumbled upon a mechanism that effectively programmed an increase in mortality into individuals as they age. Poetically, that could be called “the original curse”. Some scientists call it “programmed aging”.

Programmed aging would keep new generations turning over, and prevent the kind of species extinction scenario described above. Although programmed aging would have caused many individuals to die at an earlier age, it would have made the species more likely to continue in existence.

This analysis provides a different perspective on the singular view. In this perspective, the most important thing is to discover the mechanism for this programmed aging, and to offer people the ability to turn it off.

What is this mechanism? A popular candidate for this mechanism is epigenetic modifications to chromosomes around the body, which happen at a fairly constant rate. Additional methyl (CH₃) groups become attached to the DNA, altering how proteins are made in that cell. Interestingly, this epigenetic drift occurs even in species such as the naked mole-rat that display negligible senescence, although the rate of that drift is slower in these species than in others.

The conclusion of this line of thought is that the initiative to reverse epigenetic drift may have a bigger result than simply reversing this one hallmark of aging. The initiative may cause the reversal of all the hallmarks of aging.

Several research institutions have versions of this battle-plan, including David Sinclair’s labs at Harvard, and the well-funded Altos Labs. These research programmes are still at an early stage. They’ve already produced some interesting results. However, the plural point of view expects that, by themselves, these singular approaches won’t be sufficient to reverse all aspects of aging. Let’s look more closely into this.

The plural view: for and against

It’s time for me to put my own cards on the table. I see the singular view as being too metaphysical. It has too high a regard for the capabilities of evolution. It imagines that evolution has acted to hinder most species from achieving negligible senescence, so that animals in most species become weaker as they grow older. It further imagines that a solution to this situation – a solution that would avoid animals becoming progressively weaker as they age – is almost hidden in plain sight.

It’s almost as if a deity had placed a secret code in ancient scripture, that pious students of that religion could detect, providing them sure knowledge of a forthcoming cosmic transformation.

To be clear, I’m not opposed to any of the biological research programmes coming from the singular view – such as introducing exosomes to transform the signalome, or reinvigorating mitochondria, or spreading telomerase around the body, or strengthening the CAP, or reversing epigenetic drift. Some – perhaps all – of these projects may have very positive implications for delaying or even curing various age-related diseases.

It’s just that I expect no single intervention to be decisive. I’m persuaded by the argument that any such single intervention, if it significantly extended expected lifespan and had no associated drawback, would already have been adopted by normal processes of evolutionary selection.

I have some sympathy for the counter-argument – that any such species would have become weak to the point of extinction, on account of lack of variance in the population from new generations. But I find it more plausible that any such single intervention would only provide an incremental boost to longevity. That’s because I see the different types of age-related damage, throughout all the body’s subsystems, as being substantially independent from each other.

The task of anti-aging researchers, therefore, is:

• To identify a list of potential anti-aging interventions, and to verify how effective each one is on its own
• To explore applying combinations of these interventions in parallel
• To analyse the interactions (both positive and negative) and the side-effects of these interventions when they are applied in combination.

Importantly, I see no reason to restrict the set of potential interventions to those that already exist in nature. Evolution has produced wonders, but new tools available to human engineers can achieve new types of results.

People who hold the plural view should pay close attention to the research of those with the singular view – since the interventions explored by singular advocates may well be good candidates to include in combination experiments performed by plural advocates.

Unfortunately, experiments involving combinations of interventions will generally be more expensive (and more complicated) than those involving single interventions. So it is understandable that some investors or donors may prefer to support projects in the singular camp.

But let’s be clear. A plurality of different engineering problems needed to be solved before heavier-than-air powered flight became a daily practicality. Achieving a different kind of take-off – longevity escape velocity – will likely be similar.

Therefore I say: directing all funding toward research of singular interventions would be unwise. In the real world, it’s combinations of technologies that have the biggest impact.

How AI changes the discussion

There’s one more card to lay on the table: the AI card.

As AI improves throughout 2025 – and as we humans become more attuned to taking good advantage of AI – it’s likely that many aspects of the argument in this article will be refined:

• Bringing more data points into the discussion
• Identifying more salient aspects of these data points
• Identifying which parts of the argument are weak – and which are strong
• Suggesting alternative sets of experiments to conduct
• Proposing new hypotheses, that could make better sense of all the information and ideas collected together.

I eagerly look forward to these improvements. But I caution against over-reliance on AI. That was the conclusion of my previous Mindplex article, Solving aging – is AI all we need? For the time being, we humans need to remain in control of this conversation. So, I look forward to reading and responding to comments!

Acknowledgments

I’ve had many of the ideas in this essay in my mind for a long time, but recently, these ideas progressed as I reviewed:

• A video of a conversation between Josh Mitteldorf and Aubrey de Grey on the subject “Programmed or non-programmed aging”
• A thread on X/Twitter started by Peter Lidsky
• Conversations on “mobilized vitalists” Telegram channels with members of Vitalism.

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.