In discussions about longevity science, aging is sometimes described as an inevitable consequence of entropy: biological systems degrade over time, disorder increases, and the hallmarks of aging are merely downstream consequences of that physical decline.
That framing sounds intuitive, but it is incomplete. If aging were simply the result of random entropy, then targeted intervention would appear futile. Why focus on telomeres, senescent cells, mitochondrial dysfunction, or genomic instability if all of these are merely symptoms of unavoidable disorder?
The problem is that living systems are not passive objects gradually falling apart in isolation. Life maintains order by continuously exchanging matter and energy with its environment. Biological organisms build, repair, regulate, recycle, and adapt. If they did not, life could not exist at all.1
So the better question is not whether entropy exists. Of course it does. The better question is: why do the biological systems that maintain order begin to fail in predictable ways over time?
Aging Shows Patterns, Not Random Collapse
Several observations challenge the idea that biological aging is driven primarily by random entropy.
First, entropy is a property of closed or isolated systems, while living organisms are open systems. Biological systems continuously exchange both matter and energy with their environment. They maintain structure through metabolism, repair, and regulation. This constant input allows organisms to resist disorder for extended periods of time.1 Aging, therefore, cannot be explained simply as the passive accumulation of entropy in a closed system.
Importantly, the second law of thermodynamics still applies to living systems. However, organisms maintain local order by exporting entropy to their environment. Aging, therefore, is not the inevitable accumulation of disorder in isolation, but a gradual failure of the biological processes that maintain that order over time.
Second, entropy is inherently random, but the failures associated with aging are not. If biological decline were truly random, we would expect all age-related diseases to occur with roughly equal probability. In practice, that is not what we observe. Cardiovascular disease, cancer, and neurodegenerative disorders dominate mortality patterns, while other forms of organ failure are comparatively rare. These distributions are highly structured, not uniform.
Third, aging follows recognizable and reproducible patterns of failure. Organisms within a species tend to decline in similar ways over time. In engineering, this type of behavior is often described as a pattern of failure—where reliability decreases in predictable ways based on underlying system constraints.2 Biological systems exhibit similar behavior, with recurring pathways of decline that can be observed across populations.
These patterns suggest that aging is not a random process, but a structured one—driven by specific mechanisms that become increasingly dysregulated over time.
For example, telomere shortening follows a well-characterized pattern across cell divisions. As telomeres reach critically short lengths, cells enter senescence or undergo apoptosis—contributing to tissue dysfunction over time. This is not a random event, but a measurable and reproducible biological process.
If aging follows identifiable patterns, then those patterns can, in principle, be measured, modeled, and ultimately modified.
Statistically, human biological failures follow these patterns. There is an entire discipline—actuarial science—built around predicting them. A relatively small number of diseases account for the majority of deaths, and they do so in predictable proportions and over fairly well-defined time frames.
With enough data, it becomes possible to go even further. By comparing medical history, lifestyle, and biological markers over time, we can begin to estimate not just if someone is at risk, but what they are at risk for, and roughly when. That level of predictability would be completely impossible if aging were truly driven by random entropy.
This is exactly why the “Hallmarks of Aging” framework was developed—to identify these recurring patterns of failure and understand the mechanisms behind them.3 Once those mechanisms are understood, they can be targeted. That is the entire point.
I agree with the general idea that combining multiple interventions will likely prove more effective than relying on any single one. That makes sense intuitively and will almost certainly be borne out experimentally. However, the claim that targeting a single mechanism has “no significant impact” is not supported by the data.
Source: Ora Biomedical Mouse Model Consortium (MMC) dataset.4
Individual interventions in mouse models have demonstrated measurable effects on lifespan—evidence that targeting specific mechanisms can produce meaningful outcomes.
Looking at results like these, it becomes difficult to argue that individual interventions have no meaningful impact. We may debate how large those effects are, or how they translate to humans, but the direction is clear: targeting specific biological processes can change outcomes.
That does not mean any one intervention is the complete solution. It means the system is responsive. And if it is responsive, then it is not governed by randomness—it is governed by mechanisms.
I agree that reductionist approaches alone will not be sufficient to solve aging. A systems-level understanding will be required. But that is very different from concluding that aging is driven by randomness and therefore cannot be meaningfully addressed.
There are already well-developed models describing how and why these recurring patterns of biological failure occur, and importantly, these models are supported by experimental data. The relationship between theory and observation in this field is not forced—the data aligns with the framework.
As the evidence continues to show, biological aging is not the result of random entropy. It is driven by identifiable processes that follow recognizable patterns over time. And where there are patterns, there is the potential for intervention.
The key question is not simply whether these processes can be influenced, but how they can be influenced safely—how much, for how long, and in which cells. Control, not just activation, becomes the central challenge.
Much remains to be understood, and translating these insights into safe, effective therapies will require careful validation. But the presence of structure and predictability makes that effort both rational and necessary.
Understanding those mechanisms—and learning how to influence them in a controlled and measurable way—is the path forward.
This perspective underpins our approach at Exponential Therapeutics: aging is not an inevitability to be observed, but a system to be understood—and ultimately, to be influenced.
References:
1. Schrödinger E. What Is Life? The Physical Aspect of the Living Cell. Cambridge University Press; 1944.
2. Ebeling CE. An Introduction to Reliability and Maintainability Engineering. McGraw-Hill; 1997.
3. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The Hallmarks of Aging. Cell. 2013;153(6):1194–1217.
4. Ora Biomedical. Mouse Model Consortium (MMC) lifespan intervention data. Available at: https://orabiomedical.com/sponsor-mmc/