The Ceiling of Human Life — On Models, Records, and the Years That Actually Count

In 1965, the French lawyer André-François Raffray entered into an agreement concerning a flat in Arles. He was 47. The owner, Jeanne Calment, was 90. She retained the right to occupy the flat, and Raffray undertook to pay her a life annuity. [3] The age difference may have seemed to favour him. It did not, however, tell him for how many years he would be paying.

Raffray died thirty years later, never living to enjoy free use of the flat. The obligation to continue payments fell to his widow, and Jeanne Calment outlived him by nineteen months. When she died on 4 August 1997, she was 122 years and 164 days old. To this day, no one has been reliably confirmed to have lived longer. [4] This story is often told as a joke about an unlucky investment. What interests me is the risk concealed within it: the owner's age was known, the duration of the obligation remained unknown.

Today we ask about lifespan on a different scale. Mathematical models and DNA research are meant to help establish how long a human can live and what constrains us. Where is the ceiling of human life: just above Calment's record, at 125–130 years, or only somewhere around 150 or 200? Each of these figures says something different. It's worth checking what, before we choose the most appealing one.

A Record That Refuses to Age

Calment's record is exceptionally well documented: from civil registry records and censuses to family and notarial documents. It has been challenged, but re-analysis of the sources provided no grounds for rejecting it. [5] Whether one can live still longer, and by how much, is something we try to infer from demographic data and research on ageing.

The highest observed value is insufficient to establish a biological maximum. In the human case, the problem is particularly difficult: cases of extreme longevity are few, and a single error in a date of birth can materially alter the conclusions. Above 120 years, statistical analysis begins to resemble the meticulous examination of individual biographies.

Average life expectancy must also be distinguished from maximum lifespan. The first great longevity revolution consisted primarily in reducing infant mortality and premature deaths from infections, heart disease, and many other conditions. More people now reach old age. We have not, however, begun living to Jeanne Calment's age en masse, and her record remains unbroken. Increasing the chances of a long life and shifting its biological boundary are two different achievements.

A study published in 2024 in Nature Aging showed that in countries with the highest life expectancy, the rate of further increase has slowed markedly. The authors deemed radical life extension in this century unlikely without genuine slowing of biological ageing. More effective treatment of individual diseases need not remove all age-related constraints. [12]

One Word Too Far

The starting point for this text was a short popular-science reel by Konrad Skotnicki, known as "the TikTok doctor." [1] It prompted me to examine a thought experiment described in a scientific publication. Imagine that medicine has mastered nearly all the mechanisms of ageing: from telomere shortening — telomeres being the end sections of chromosomes — to mitochondrial dysfunction and the harmful effects of cellular senescence. Only somatic mutations remain, that is, changes in the DNA of our body's cells arising over the course of a life. How many years could we then live?

The authors of a paper published in 2026 in npj Aging sought the answer. Reading their results, one must be careful with the word "maximum." The range of 146–194 years given in the publication refers to median lifespan. The median describes the middle of a distribution: half the people in a hypothetical group would live shorter lives and half longer. Replacing it with a maximum changes the meaning of the entire story. And we are still talking about the output of a model, not a measurement of anyone's actual lifespan. [2]

As a modelled maximum, the authors took the age at which the probability of survival falls to one in a hundred thousand. Under various assumptions about the interdependence of organ failure, this ranged from 210 to 557 years. Survival beyond that threshold remains possible within the model, though extremely improbable. The limit here follows from the definition adopted. [2]

The figure 194 is easy to remember as the answer to the question of how long a human can live. The researchers, however, calculated something else: what still constrains us if we treat nearly all other obstacles as removed. In investing, too, I start with the assumptions. The most attractive figure in a spreadsheet does not exempt me from checking them.

The Organism According to an Engineer

At the heart of the paper are somatic mutations: changes in the DNA of body cells which, unlike germline mutations, are not passed on to offspring. Many have no significant consequences; some may damage a gene essential to a cell's survival. The authors treat these changes as irreversible and link them to entropy — in this framing, the loss of information about the original DNA sequence. The difficulty lies in restoring that sequence when no template remains. The assumption adopted does not determine whether future medicine will manage such a repair.

They then describe organs using models drawn from reliability analysis. An organ functions as long as enough viable cells remain within it. Tissues capable of renewal fare surprisingly well in the simulation: the liver and bronchial epithelium have substantial regenerative capacity. A stronger constraint is posed by cerebral cortex neurons and cardiomyocytes, the cells of the heart muscle. Their capacity for renewal is limited, and many of them serve us for decades.

In an ideal world without ageing, with a constant all-cause mortality risk equivalent to that of a contemporary thirty-year-old in Swiss data, the model yielded a median of 1,759 years. A combined model of four cell types (neurons, cardiomyocytes, hepatocytes, and basal cells of the respiratory epithelium) lowered this to around 156 years. Other mechanisms of ageing and the age-related rise in cancer risk were omitted. Mutations leading directly to cell death were included. Neither organ replacement nor therapies inhibiting mutation formation was assumed. From these assumptions no forecast follows regarding the age we will actually reach. [2]

If the constraint turns out to be the loss of neurons, the path to a very long life may run through protecting, repairing, or replacing them. In the brain, however, cell count alone is not enough to judge success. What also counts is their function and their connections, on which memory and learned abilities depend.

Another Year Like a Coin Toss

Scientists have argued for years over whether an unsurpassable limit can be assigned to human life. A prominent 2016 publication in Nature pointed to around 115 years as a typical annual record and assessed the chance of anyone exceeding 125 years in a given year at less than one in ten thousand. The authors wrote of a natural, even absolute, limit that we have probably already reached. [6] In rebuttals published a year later in the same journal, critics questioned, among other things, the choice of trend break point, the small number of observations, and the method of analysing annual maxima. [7]

Another group of researchers analysed data on 3,836 Italians aged at least 105, with documented confirmation. In Science they described a mortality plateau: the risk of death ceased to rise markedly with age. For women born in 1904, the model yielded an annual mortality risk of around 47.5 percent; for later cohorts it was slightly lower. [8] It sounds almost optimistic until we translate it into plain language: the chance of surviving another year is barely greater than the chance of flipping heads. If the risk remained constant, of a thousand people celebrating their 110th birthday, one or two would reach 120. A 130th birthday, in turn, would be reached by approximately one in four hundred thousand of those who turned 110.

Not all data confirm such a plateau. An analysis of 3,789 French individuals of confirmed age, observed from the age of 105, indicated in 2023 that mortality risk continues to rise. [10] Other work, covering people over 108 from several countries, found no convincing evidence of an upper limit to lifespan. [9] Failing to detect a limit does not prove there isn't one. Still less does it mean immortality: even without a rigid limit, the chance of surviving further decades may be negligible.

A demographic forecast by Michael Pearce and Adrian Raftery gives over 99 percent probability that Calment's record will be broken by 2100, around 44 percent that someone will reach at least 128 years, and around 13 percent for reaching 130. The results rest on the assumption that after the age of 110 mortality risk is approximately constant, and that the number of people aged 110–114 will grow as forecast. [11] The record may therefore improve without anyone being rejuvenated. More people will reach a very advanced age, and with their number will grow the chance of another exceptionally long biography.

The research does not point to a common limit. The rising mortality risk in the French analysis calls for more caution toward forecasts based on a plateau. My supposition is that without a biological breakthrough, this century's records will be closer to 125–130 years than to 150. That is my interpretation, not an agreed species limit nor a confident prediction of the future.

What the Laboratory Can Actually Do

One of the most important contemporary classifications describes twelve interconnected hallmarks of ageing: from genomic instability and telomere shortening, through mitochondrial dysfunction, loss of proteostasis and cellular senescence, to chronic inflammation and disrupted intercellular communication. [13] Identifying these processes does not mean, however, that we can effectively halt each one.

Rapamycin extends the lifespan of genetically heterogeneous mice, even when administration begins late. [15] Human studies, however, are small or short and assess mainly safety, selected bodily functions, and laboratory markers. They yield ambiguous results, and the drug can cause adverse effects. It has not been shown to extend the life of a healthy person. [16] Senolytics — substances intended to clear cells in a state of senescence — also perform promisingly in many animal models. Senescence is a form of cellular ageing in which cells can harm surrounding tissue. In a phase 2 trial involving 60 postmenopausal women, however, a combination of dasatinib and quercetin produced no significant improvement in the primary marker of bone tissue breakdown. [17]

In the two-year CALERIE experiment, 220 healthy non-obese individuals were randomly assigned to either a calorie-restriction group or a control group. Participants in the first reduced their calorie intake by an average of 11.9 percent. In an analysis of data from 197 people, the DunedinPACE indicator suggested a slowing of the estimated pace of ageing by around 2–3 percent relative to the control group. It is calculated on the basis of chemical markers in DNA. No significant change was recorded in the results of two other biological clocks, PhenoAge and GrimAge. [14] Will the participants live longer as a result? We don't know. A change in an indicator does not settle the question of extended life.

On 9 June 2026, the trial sponsor announced the first dose of ER-100. [19] This experimental gene therapy is designed to trigger, in a controlled manner, the production of three proteins known collectively as OSK, which regulate gene activity. Researchers are attempting in this way to restore to cells certain characteristics of a younger organism. The phase 1 trial assesses primarily the safety of the therapy in people with glaucoma or a particular type of ischaemic optic nerve damage. It is an attempt to treat a specific eye condition. Announcing that the first patient has been dosed is not yet evidence of efficacy, nor a harbinger of whole-body rejuvenation.

The studies cited have not demonstrated that an intervention targeting biological ageing extends the lives of healthy people. In a consumer warning about health fraud, the FDA notes that no drug has so far been proven to slow or reverse the ageing process. [18]

Nine Years of Difference

The figure closest to everyday life is not 122, 156, or 194. It is 9.1. That was the median gap in 2019 between life expectancy and HALE — health-adjusted life expectancy — in an analysis of 183 countries based on WHO data. [20] This does not mean nine final years spent continuously in illness. The measure accounts for the duration and severity of health problems at various stages of life. It does not describe an inevitable old age for every one of us.

Would shifting serious illness from the age of 72 to 82 not matter more to millions of people than one person's spectacular record? The most tangible goal of longevity research today lies closer to hand: preserving health, capability, and independence for as long as possible.

This shifts the business perspective. The longevity market gives sellers a particular advantage: some promises can only be settled after several decades. A conveniently distant deadline. Meanwhile, a biological age test can show a change after just a few months. It is easy to present an improved score as evidence of a patient's future health. Investors and entrepreneurs should therefore ask about clinical outcomes, the study population, trial quality, and the durability of the effect. What benefit will a person experience, and how will we recognise it?

Value may emerge in less spectacular areas: earlier disease detection, rehabilitation, preventing loss of function, designing housing and services for older people, better care, and technologies supporting independence. One need not promise immortality to create a needed and durable service. A small gap between life expectancy and HALE is likewise not enough to declare success. It may occur in a country where people simply die early. The goal should be more years in health and fewer limitations on capability.

Who Will Get the Extra Years?

Let us assume, though, that a second longevity revolution arrives and healthy life is extended by thirty or fifty years. Then far more than medicine will change. The rules of retirement, education, career planning, and business succession will all need rethinking. With a lifespan of 150 years, the span between reaching independence and late old age could accommodate several entirely different professional and family choices.

A hundred years of activity probably won't fit into one career. Should an owner run a business for eight decades if they remain capable while their successors have themselves grown old? Such prolonged coexistence of several adult generations would also change the housing market and the structure of families. Capability alone, after all, does not determine how responsibility should be shared. A founder's longer activity need not mean an equally long tenure in power.

Then there is the question of access. If a costly therapy reaches the wealthiest first, the additional years may deepen their advantage. They would provide more time for compounding capital, maintaining influence, and deferring succession. Inequalities would begin to accumulate like capital at compound interest: an existing advantage generating the next, over an even longer period. That is reason to think about access, funding, and rules before a therapy proves effective. Once it has become someone's advantage, agreeing those rules will be harder.

Many technologies, however, began as expensive solutions available to a few and later became cheaper. If partial cellular reprogramming proves effective and safe, its first applications may serve to treat specific conditions — protecting eyesight, for instance. There will be no need to wait for someone's 200th birthday to assess the value of such a therapy. This seems to me a more probable route to a longer healthy life than a single "cure for old age." We do not yet know whether it will also allow us to raise the biological maximum.

Not Everything Can Be Counted in Years

So how long can a human live? The documented record is 122 years and 164 days. Some demographic models allow for records of 125–130 years this century without a biological breakthrough. The data cited, however, provide no grounds for expecting a 150th birthday. We do not know of an unsurpassable limit. Nor is one established by a model in which we assume mastery of nearly all the mechanisms of ageing.

The somatic mutation model points to one possible obstacle. Demography describes today's probabilities, and research on ageing tests whether its mechanisms can be influenced effectively and safely.

Jeanne Calment outlived the man who tried to estimate how long he would be paying her annuity. She did not thereby define the limit of the species. What she did was remind us that even a reasonable forecast does not remove risk. For Raffray, each successive year carried a measurable price. For Calment, it was the continuation of a life whose length neither party knew at the moment of signing.

We can therefore go on asking about 150 or 200 years, provided that waiting for a breakthrough does not obscure matters far closer at hand: effective treatment, access to care, and the ability to decide for ourselves even when we need help.

Sources

1. Konrad Skotnicki ("the TikTok doctor"), popular-science reel cited by the author as inspiration.
2. Evgeniy Efimov, Vlad Fedotov, Leonid Malaev, Ekaterina E. Khrameeva, Dmitrii Kriukov, "Somatic mutations impose an entropic upper bound on human lifespan," npj Aging, 2026.
3. Pierre-Yves Glass, "It Was A Great Deal But Not For Him: 120-Year-Old Turns Tables In Real Estate Gamble," Associated Press / The Spokesman-Review, 29 December 1995.
4. Guinness World Records, "Oldest person ever," accessed 27 September 2026.
5. Jean-Marie Robine et al., "The Real Facts Supporting Jeanne Calment as the Oldest Ever Human," The Journals of Gerontology: Series A, 2019.
6. Xiao Dong, Brandon Milholland, Jan Vijg, "Evidence for a limit to human lifespan," Nature, 2016.
7. Nicholas J. L. Brown, Casper J. Albers, Stuart J. Ritchie, "Contesting the evidence for limited human lifespan," Nature, 2017.
8. Elisabetta Barbi et al., "The plateau of human mortality: Demography of longevity pioneers," Science, 2018.
9. Léo R. Belzile et al., "Human mortality at extreme age," Royal Society Open Science, 2021.
10. Linh Hoang Khanh Dang et al., "The question of the human mortality plateau: Contrasting insights by longevity pioneers," Demographic Research, 2023.
11. Michael Pearce, Adrian E. Raftery, "Probabilistic forecasting of maximum human lifespan by 2100 using Bayesian population projections," Demographic Research, 2021.
12. S. Jay Olshansky et al., "Implausibility of radical life extension in humans in the twenty-first century," Nature Aging, 2024.
13. Carlos López-Otín et al., "Hallmarks of Aging: An Expanding Universe," Cell, 2023.
14. R. Waziry, C. P. Ryan, D. L. Corcoran et al., "Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial," Nature Aging, 2023.
15. David E. Harrison et al., "Rapamycin fed late in life extends lifespan in genetically heterogeneous mice," Nature, 2009.
16. Deborah J. W. Lee, Ajla Hodzic Kuerec, Andrea B. Maier, "Targeting ageing with rapamycin and its derivatives in humans: a systematic review," The Lancet Healthy Longevity, 2024.
17. Joshua N. Farr et al., "Effects of intermittent senolytic therapy on bone metabolism in postmenopausal women: a phase 2 randomized controlled trial," Nature Medicine, 2024.
18. U.S. Food and Drug Administration, "Medication Health Fraud for Specific Diseases and Conditions," accessed 27 September 2026.
19. Life Biosciences, "Life Biosciences Announces First Patient Dosed in Phase 1 Trial of ER-100 for Optic Neuropathies," 9 June 2026.
20. Armin Garmany, Andre Terzic, "Healthspan-lifespan gap differs in magnitude and disease contribution across world regions," Communications Medicine, 2025.

‍