What this covers
Sara Walker takes the conversation through a complete reconceptualization of what life is and where it exists. Rather than a property of individual organisms or chemical reactions, Walker argues life is a planetary-scale physical process operating across time, with selection and temporality as fundamental features. The core claim is that standard biological definitions fail because they mistakenly locate life in space-bound individuals instead of in the temporally extended lineages and causal structures that actually constitute living systems. Sara Imari Walker develops this argument through Assembly Theory, which quantifies living complexity using copy number and assembly index—marking a sharp boundary around 15 construction steps in organic chemistry that only selection can cross. She positions the technosphere, the global integration of technology and life, as the largest living object known and as life's continuing origin event.
The conversation spans several connected territories. Walker directly contests Stephen Wolfram's view that computation is fundamental, arguing instead that computation sits atop reality as something the biosphere built. She treats the origin of life and the search for alien life as a single coupled problem and describes Lee Cronin's self-reproducing molybdenum ring as a window into how the origin-of-life transition occurs through autocatalytic feedback. On physics foundations, Walker proposes that randomness, not determinism, is basal and that deterministic structures only emerge through self-reinforcement. She reframes meaning as an emergent property of recurrence and causal relationship rather than something carriers in individual units, and develops an account of free will consistent with both deterministic macroscopic structures and quantum-level randomness. She also addresses more recent concerns about AI by rejecting doomsday narratives in favor of analyzing concrete harms like deepfakes and weaponization that require legislative response rather than metaphysical alarm.
Walker argues that life is not a property of individuals or chemistry but a planetary-scale physical process in which information structures matter across time, so that the right physics for life requires treating time and selection as fundamental and recognizing living things as enormous causal structures extended in time rather than space.
- Standard definitions of life all break down because they wrongly locate life in individuals rather than in temporally extended lineages
- Assembly theory quantifies life via copy number and assembly index, marking an abrupt boundary (~15 steps in organic chemistry) only crossable by selection
- Living objects are 'big in time' not space; the technosphere is the largest such object and represents life's continuing origin
Life should be defined by how information structures matter over time, not by chemical or individual-organism criteria.
- Walker's own definition: life is the process of how information structures matter over time and space, and an example of life is what emerges on a planet and yields an open-ended cascade of generation of structure and increasing complexity; any individual is just a particular short temporal instance of lineages structured across time, which is why narrower definitions break down.
“life is the process of how information structures matter over time and space, and an example of life is what emerges on a planet and yields an open-ended cascade of generation of structure and increasing complexity”
- The modern technosphere—the global integration of life and technology on this planet—is the largest object in time in the known universe and is itself a form of life because it is creative (with us, not independently) and represents the current furthest boundary of how far life on Earth has pushed into what the universe can generate.
“I think the modern technosphere is the largest object in time in the universe that we know about.”
- Life is not inherently chemical; it emerges in chemistry because chemistry is the first combinatorial space the universe builds that is too large to exhaust—for example, a single molecular formula like Taxol could fill 1.5 universes in volume if every 3D shape were made—forcing the universe to make historically contingent choices about which molecules exist, and life is the physics of how the universe selects what gets to exist.
“life emerges in chemistry because life is the physics of how the universe selects what gets to exist”
- Defining life requires abandoning the individual as the relevant unit; the standard astrobiology definition (a self-sustaining chemical system capable of Darwinian evolution) is wrong on every word, because no organism is truly self-sustaining (we depend on societies and environments), life is not fundamentally chemical, and Darwinian evolution operates on populations not individuals—so the definition paradoxically implies populations or societies are alive but individuals are not.
“I think every word in that definition is actually wrong as a descriptor of life.”
Living objects are vast in causal depth and temporal extension, not spatial size; we perceive them as small because we evolved to resolve space, not time.
- Life is very high dimensional, so an entity can be alive in some dimensions and not others; some features of you require very little causation or history while others require a great deal, making it impossible to project the deep high-dimensional structure of life into a single measurable property.
“I think life is very high dimensional. In fact, I think you can be alive in some dimensions and not in others.”
- Things only look emergent because we cannot see time; if we could perceive time fully, we would see the entire causal tail of every object—all the steps and interactions that produced its current state—revealing that all living lineages are interconnected and exchanging parts, which is the deeper reason the individual cannot be the unit of life.
“things only look emergent because we can’t see time”
- Living objects are among the biggest structures in the universe, but they are big in time (causal structure) rather than space; because we evolved perception that resolves space well and treats time as fleeting, we see organisms as small ephemeral things and fail to perceive that each contains the entire causal history of the universe leading to it.
“living objects are actually huge. They’re some of the biggest structures in the universe, but they are not big in space. They’re big in time.”
- Consciousness is a manifestation of being a temporally extended object: because we are so huge in time, we have a very large inner space that we experience, and our conscious experience is unique and unshared because no other object occupies the same part of time; we can partly understand each other's consciousness because we diverged recently in causal history and share overlapping structure through language.
“consciousness is about the fact that we have a conscious experience is because we are these temporally extended objects”
The origin of life is an abrupt phase transition where mutually reinforcing structures collapse a combinatorial explosion into self-reproducing systems.
- Lee Cronin's team identified a self-reproducing molybdenum ring of roughly 150 atoms whose vast configuration space would allow exponentially many possible molecules, yet the system collapses onto making that one structure because the large ring becomes a template for the smaller intermediates—a catalytic feedback mechanism that Walker thinks reveals the origin-of-life transition itself.
“It’s like the first non-organic autocatalytic set. It’s a self-reproducing molybdenum ring.”
- The origin of life is a boundary the universe can only cross when a structure can reinforce its own existence (self-reproduction, autocatalysis) and maintain itself against chemical and quantum randomness; it is the emergence of a deterministic structure that says 'I'm going to keep going', and Walker now believes this transition is abrupt rather than gradual—a rapid collapse of a combinatorially exploding possibility space onto a small mutually reinforcing set of objects.
“The origin of life is like this boundary that the universe can only cross if a structure that emerges can reinforce its own existence”
- Because all objects decay and have finite lifetimes, a structure that can make a copy of itself before it dies can persist in time, which is how something emerges out of the randomness of the chemical soup; but things that copy themselves are very rare, so the key is closing a loop where objects mutually reinforce each other's existence (A makes B, B makes C, C makes A).
“If the system can make itself, then it can persist in time because all objects decay, they all have a finite lifetime.”
- The entire history of life is recursion: an origin-of-life event evolves for billions of years into a technosphere, the technologies start having the property of life, the system solves the origin of itself, figures out how the process works, and can then copy itself onto another planet so the whole biosphere reproduces—meaning the origin of life is happening again now as the planet undergoes a multiscale transition from individual humans to the global internet.
“When a technosphere can move off planet and basically reproduce itself on another planet, now you have biospheres reproducing themselves.”
Assembly theory measures complexity through copy number and assembly index, marking life's threshold at roughly fifteen recursive construction steps.
- Assembly theory measures the boundary above which objects are too complex to form without selection using two observables: the copy number (how many identical copies were observed) and the assembly index (the minimal number of recursive construction steps from elementary building blocks); experimentally, organic molecules above roughly 15 steps are only found in life, marking the threshold a self-reinforcing structure must cross.
“One of them is the copy number of the object... And the second one is what’s the minimal number of recursive steps to make it?”
- Because the assembly universe grows exponentially with assembly index and the typical path to any complex object also grows exponentially, building complexity by a typical path is an exponentially receding horizon; therefore most objects that come into existence must be causally very similar to existing objects and reached by an almost-shortest path, meaning complexity scaffolds on prior structure rather than arising by random walk.
“if you want to try to make a more and more complex object and you want to do it by a typical path, that’s actually an exponentially receding horizon”
- The most significant struggle for existence is not among objects that exist but between those that do and those that never get a chance to; each time the universe generates a new structure along a causal chain it excludes an exponentially large set of possibilities, so existence itself excludes a huge number of things and this is where selection does most of its causal work.
“The most significant struggle for existence in the evolutionary process is not among the objects that do exist, but between the ones that do and those that never have the chance to.”
Current definitions of life fail because they are too narrow; no single measurable property captures life's high-dimensional nature across all features.
- Every classical definition of life—self-reproducing system, eats to survive, requires compartments—has a counterexample that defeats it (viruses, fire, cars), which is why life has resisted definition; this signals the definitions are not general or abstract enough rather than that life is undefinable.
“there’s always a counterexample that challenges that definition. This is why viruses are so hard or why fire is so hard.”
- Schrodinger correctly predicted DNA as an aperiodic crystal—the only physical structure that could encode enough information to specify a cell—but his attempt to explain life via non-equilibrium physics and entropy is the wrong approach; in open systems the second-law paradox dissolves, and the real question is why life generates so much order, which entropy does not explain unless entirely reformulated.
“he did come up with something really insightful, which was to predict the structure of DNA as an aperiodic crystal”
- Physics' assumption of a deterministic universe is an artifact of Newton's framing of laws as fixed dynamical equations acting on initial conditions; Walker argues this is only a good approximation for systems studied so far and will radically fail at describing reality's more basal levels, because the universe is actually random at its base and deterministic structures only emerge through self-reinforcement.
“Physics hardly even acknowledges that the universe is random at its base. We like to think we live in a deterministic universe”
AI risks are real but arise from human misuse, not autonomous superintelligence; genuine dangers require legislation, not doomsday narratives.
- The 'super machine in a box that's going to kill everybody' is a bad science-fiction narrative; the real dynamic is a co-evolution between humans and technology in a shared ecosystem where balance shifting to technology would require very bad human actors, and the genuine risks are deepfakes, weaponization, child exploitation, and faked kidnappings, which require new legislation and guardrails rather than doomsday framing.
“Or is it really that we’re building some super machine in a box that’s going to be smart and kill everybody? ... It’s a bad science fiction narrative.”
- It is a mistake to think of our technologies as a separate species; they are part of the technosphere—a system that is both human and technological—and they operate at the societal level, deeply integrated with the social organism, just as societies (like the multi-sided arguments in politics) are themselves alive and solve problems at a level distinct from any individual actor.
“I don’t like thinking our technologies as a separate species because this again goes back to this sort of levels of selection issue.”
- Doing theoretical physics literally changes the structure of your brain and keeps you on the edge of madness, but the rigor of holding abstractions to correspondence with reality and working on concrete problems is what preserves sanity.
“if I wasn’t a theoretical physicist, I would be probably not sane. But what it forces you to do is you have to hold yourself to the fire of these abstractions”
- Beauty is not universal and has no single function tied only to sexual selection; the adaptation beauty confers is far richer, encompassing social hierarchy, social mobility, group identity (e.g. goth aesthetics) and power dynamics—Walker argues clothing and aesthetics carry function by shaping how others interact with you and by embodying the person you want to be, not merely by signaling health.
“the adaptation that beauty confers is far richer than that. Some of the adaptation is about social hierarchy and social mobility and just playing social dynamics.”
- Words are objects with their own ontology, physical things that carry causation and create meaning, but they are not the things they describe; likewise our mental representations are small projections of what is actually present, and confusing the sign with the world (as Gödel's work warns) causes us to miss the deeper structure beneath language and the fluid, approximate nature of our representations.
“words are objects and the objects are not the things that they describe. They have a different ontology to them.”
- Chirality (molecular handedness) is usually treated as a spatial property, but Walker argues it breaks symmetry in time: choosing one handedness fixes how that molecule will react with every other object in its chemical universe for all time, splitting the universe of possibilities; chirality's prevalence is a signature of being in a complex chemical space and largely a product of life rather than a generic feature of chemistry.
“I think chirality breaks symmetry in time, not space.”
- The frontier of human understanding does not have a fixed limit; rather, the size of the 'book' describing reality is itself growing, because the universe is getting bigger in time and continually invents new spaces for new questions to exist (e.g. questions about exoplanets or singularities could not be asked centuries ago), so we can never know all of it even though there is no actual infinity.
“I don’t believe in infinity as a physical thing, but I think as a receding horizon, I think because the universe is getting bigger, you can never know all of it”
- Because the universe is resource-bound (time, information, and material are finite), most things that could exist never will; death is therefore a way to make room in the universe for other things to exist, and the optimal strategy to maximize the number of things that can exist is to make recursively-stacked high-structure objects in small volumes of space and turn them over rapidly.
“death is a way to make room in the universe for other things to exist that wouldn’t be able to exist otherwise”
- It is efficient for organisms to eat other things that are already alive because they don't have to go all the way back to the base chemistry to build the complex molecules they need.
“it’s efficient for things to eat other things that are already alive because they don’t have to go all the way back to the base chemistry”
Computation and mathematics are products of biological evolution, not fundamental features of reality itself.
- Contrary to Wolfram-style views that computation is fundamental at the base of reality, Walker argues computation sits at the top of reality because it was built by our biosphere after billions of years of evolution; the feeling that the universe is computational arises because we, as agents possessing the theory of computation, can embed it in physical substrates (like string lights), and we mistake our ability to do so for a property of the world itself.
“we mistake our ability to do that with assuming that the world is computational”
- Every theory must assume something fundamental (a 'turtle'), and which thing is fundamental depends on the question you are trying to answer—time for life, particles for matter-light interactions, waves for electromagnetism, wave functions for quantum mechanics; Walker cannot accept Wolfram's computation-as-fundamental ontology because she cannot identify what question it is meant to answer, and a theory of reality without a question to answer is unmotivated.
“Everyone has a turtle. You can’t build a theory without a turtle. It depends on the problem you want to describe.”
- Walker holds a strict physicalism in which everything—including the multiverse and mathematics—exists here inside our universe; the multiverse is a concept in human minds that is causally useful for our cosmological reasoning, and math is an object that emerged in our biosphere rather than a Platonic world, because the very ability to talk about something means it exists inside and is part of what the universe is building.
“I am the most physicalist that anyone could be. So for me, everything exists in our universe.”
- There is no theory of everything and reality has no ground or bottom, but there are better and more explanatory theories; a true theory of everything is paradoxical because it would have to explain the theorizers and be recursive (explain itself), and none of our physics theories are recursive—as David Krakauer put it, a theory of everything is a theory of everything except those things that theorize.
“I don’t think that reality has a ground, and I don’t think there’s a theory of everything, but I think there are better theories”
Physics needs a new material concept for treating information physically, analogous to how unified field concepts revolutionized earlier physics.
- Just as Aristotle wrongly explained motion via material qualities (earth-like things fall, air-like things rise) until physics discovered that mass and acceleration unify all motion in a simple description, we currently cannot build a unified theory of life because we have not yet invented the right material concept for treating information physically.
“if we just talked about mass and acceleration, we could unify all motion in the universe in a really simple description”
- Materialists and vitalists disagree fundamentally about whether life can be fully described by matter; Walker argues vitalists were not simply wrong, because our concept of 'matter' is itself a historical invention, and if we learn to measure and rigorously describe currently unmeasured patterns, we may find matter has more properties than we thought, one of which could be life, making life a material property that still retains the mystery vitalists noticed.
“the material world matter has more properties than we thought that it did. One of those might be associated with the thing that we call life. Life could be a material property and still have a lot of the features that the vitalists thought were mysterious.”
- What counts as 'fundamental' has historically meant the smallest indivisible unit of matter (atoms, then particles), but this definition keeps shifting as technology lets us resolve smaller features; Walker argues the new frontier of physics lies not in high-energy particle physics or quantum gravity but in the problems of life and intelligence, and that we should treat construction methodology rather than smallest constituents as fundamental.
“we typically define fundamental right now in terms of the smallest indivisible units of matter”
Abstract products of the biosphere like language, mathematics, and memes may themselves be alive or life-like due to combinatorial recursion.
- Some products of our biosphere that seem abstract—language, memes, and mathematics—may themselves be alive or at least 'life', because they are open-ended combinatorial spaces with recursive properties that allow creativity; mathematics in particular is not a Platonic reality outside the universe but a feature of our biosphere that tells us about our own structure.
“I think language is alive, or at least life. I think memes are.”
- Treating mathematics not as a tool physicists use to describe reality but as an object that emerged in our biosphere and is itself part of the reality physicists study is a deeply interesting inversion that enables a kind of experimental approach to mathematics, in contrast to Platonism which cannot be experimented upon.
“to think about math itself as an object for study by physicists rather than a tool physicists use to describe reality... is a deeply interesting inversion”
- Meaning is not carried by individual words but is an emergent property of causation—how often objects (like configurations of letters) recur and what the relationship of that recurrence is to other objects; this is why language, like chemistry, is a combinatorial space that crosses a threshold into high-regularity structures that persist and acquire meaning.
“Meaning is an emergent property of the causation and the objects and how often they recur and what the relationship of the recurrence is to other objects.”
Solving the origin of life and detecting alien life are the same problem; discovering extraterrestrial life requires understanding life's origin.
- Solving the origin of life and discovering alien life are deeply coupled and effectively the same problem: the first contact with alien life will be in an origin-of-life experiment, we must solve the origin of life to estimate how likely life is elsewhere, to recognize alien intelligence, and to get off-planet by starting life on another planet.
“the problem of discovering alien life and solving the origin of life are deeply coupled and in fact are one in the same problem, and that the first contact with alien life will actually be in an origin of life experiment”
- Large language models are like the genetic system for language: language had been a large in-time structure distributed dynamically across human brains and occasionally stored statically in books, but LLMs can now store the dynamics of that structure in a physical artifact—analogous to the evolution from primitive messy genes to the read-writable genetic architecture of the eukaryotic cell—crystallizing a slice of collective human intelligence in a box.
“I think large language models are kind of like the genetic system for language in some sense”
- As a biosphere evolves, it packs more and more time (causation, recursion, information) into a small volume of space without expanding spatially, becoming increasingly virtualized; sufficiently evolved biospheres may internalize their universe and become imperceptible to others—possibly looking like black holes or dark quantum computers—forming a 'great perceptual filter' whereby the more complex you become the harder it is to perceive other complex structures.
“the structure that we’re building on this planet is packing more and more time in this very small volume of space”
Consciousness emerges from being temporally extended; shared understanding arises from overlapping causal history and divergence.
- Free will is consistent with physics: we are deterministic macroscopic structures embedded in a random quantum background, and the selection bundled into our deep causal history lets us select on possible futures; free will is never instantaneous but operates over a temporal horizon within the combinatorial space of our own structure, steering toward futures that still include a coherent version of ourselves.
“It’s the fact that you’re a deterministic structure living in a random background. And also, all of that selection bundled in you allows you to select on possible futures. So that’s where your will comes from.”
- Objects that are high-density configurations of matter with a lot of past selection embedded in them are not only products of selection but also become selecting agents in the universe—the more selection embedded in you, the more possible selection you have over your future, so intelligence is the ability of these deep-in-time structures to navigate and generate the space of possible futures.
“the more selection that you have embedded in you, the more possible selection you have on your future”