More than four billion years ago, before continents drifted and oceans teemed with life, Earth was little more than molten rock and chemical chaos. Yet, within that cosmic turbulence, matter became aware of itself.
For decades, science has tried to explain how the first cell could have formed spontaneously from the raw materials of a young planet.
Now, a new theoretical study argues that the odds of the first cell forming spontaneously were so small that we may need to consider far stranger possibilities—like life being seeded, guided, or even “terraformed” into existence.
In a paper titled “The Unreasonable Likelihood of Being: Origin of Life, Terraforming, and AI,” Dr. Robert G. Endres, a biophysicist at Imperial College London, presents a sweeping and highly quantitative look at the origins of life. He blends information theory, thermodynamics, and a touch of philosophical daring.
Dr. Endres doesn’t claim that life was designed, but he argues that the math behind life’s spontaneous emergence may stretch credibility.
“‘All cells come from cells’ leads us into a classic chicken-and-egg dilemma: where did the first cell come from?” Dr. Endres writes. “Either it came from somewhere else—conveniently outsourcing the mystery—or it emerged from the laws of physics and chemistry on a young, chaotic, and geologically active Earth.”
To address this, Dr. Endres asked how much structured information would have to self-organize from randomness to form the simplest living system.
Using Kolmogorov complexity—which measures the shortest possible algorithmic description of a system—and rate–distortion theory, which studies how information can be compressed without losing essential detail, he calculated the “information bottleneck” between disordered prebiotic chemistry and the first protocell.
In his model, the “entropy collapse” of chaotic chemistry into the organized structure of life—meaning the transition from disorder to order—depends on a delicate balance between information gained and the breakdown of molecules.
That might sound modest. But viewed through the lens of billions of years and countless chemical combinations, it implies an astonishing improbability. For life to emerge naturally, the prebiotic world would have needed just the right combination of persistence, energy, and luck—essentially a cosmic lottery win.
“In other words, without immense persistence, life’s emergence becomes cosmologically implausible, potentially pointing to alternative mechanisms,” Dr. Endres notes.
That “alternative mechanism” is where the study brushes against ideas once relegated to the fringes of science fiction. In his paper, Dr. Endres doesn’t endorse any single hypothesis, but he acknowledges that directed panspermia—the concept that life was intentionally seeded by an advanced civilization—remains “a speculative but logically open alternative.”
The idea isn’t new. Dr. Francis Crick, co-discoverer of DNA’s structure, and Dr. Leslie Orgel first proposed directed panspermia in 1973 as a testable hypothesis rather than an act of faith.
Dr. Endres revisits that suggestion in the context of modern planetary science and statistical modeling. If the spontaneous assembly of a viable protocell is so improbable that it strains naturalistic explanations, then perhaps life didn’t begin on Earth. It simply continued here.
“Was Earth terraformed, or did order coalesce from chaos under the silent governance of physics?” Dr. Endres asks. “Today, humans seriously contemplate terraforming Mars or Venus in scientific journals. If advanced civilizations exist, it is not implausible they might attempt similar interventions—out of curiosity, necessity, or design.”
Still, “Occam’s razor weighs in: abiotic evolution, however slow and strange, remains a viable explanation,” Dr. Endres cautions.
Part of what makes the question so slippery, Dr. Endres argues, is our observational bias. Life on Earth arose early—geologically speaking—but intelligent life took almost the entire remaining lifespan of the planet to appear. “If intelligent life requires too much time to emerge, many planets—especially around hotter, short-lived stars—may simply run out of time, biasing observations toward an unusually early start,” he notes.
In other words, we may be living on one of the rare worlds where the timing happened to line up perfectly: not too early for chemistry, not too late for consciousness.
This anthropic coincidence, he suggests, might explain why life appears “miraculously” fast without invoking miracles. But it also underscores how finely tuned the parameters must be for living systems to form at all.
To bridge the gap between mere chemistry and the information-processing complexity of life, Dr. Endres draws on examples from both physics and artificial intelligence.
“Since Erwin Schrödinger’s landmark essay What is Life?, physicists have speculated whether the emergence of life might require physical principles beyond those already known,” he writes.
Building on the work of figures such as Dr. Ilya Prigogine, Dr. Alan Turing, and Dr. Stuart Kauffman, the paper explores whether life might be a form of information flow—a self-organizing computation governed by physical laws yet capable of encoding meaning.
In this context, Dr. Endres compares chemical reaction networks to neural networks. He suggests that sufficiently complex chemistry can, in principle, “approximate nearly any behavior, from fast signaling to slow gene regulation and cell-cycle control.”
This parallel leads to a provocative idea that life and artificial intelligence might represent two sides of the same coin—emergent computational systems driven by the flow of information through matter and energy.
“If life is a form of physical computation,” as some have proposed, then AI might soon help identify how natural chemistry can become computationally powerful enough to self-organize,” Dr. Endres writes. “Personally, I am skeptical that a generative AI can move from predicting the next word to proving the origin of life, bootstrapping us out of our existential dilemma.”
Returning to the heart of the improbability argument, perhaps the most startling insight from Dr. Endres’ analysis is not that life is impossible, but that it demands persistence bordering on the absurd.
For informational complexity to accumulate fast enough to cross the threshold from nonliving to living, molecular systems must preserve structure and “memory” over immense timescales.
If that memory were constantly erased—if molecules broke down faster than new ones could form—life would never have stood a chance. The study’s mathematical models show that, under random conditions, the time required to spontaneously assemble a cell could exceed “ten million times the universe’s current age.”
Something, it seems, must have tilted the odds in life’s favor. Exactly what that was remains one of humanity’s greatest mysteries. “A purely random soup is too lossy—some form of prebiotic informational structure must precede Darwinian evolution,” Dr. Endres writes.
Ultimately, the paper doesn’t seek to solve the mystery of life’s origin. Instead, it looks to reframe the question. By viewing biology through the lens of information theory, Dr. Endres underscores the deeper challenge: understanding how physical systems could store, process, and replicate meaning long before natural selection began.
“If life is the ultimate emergent phenomenon, perhaps it resists prediction not because the physics is wrong, but because the framework is incomplete,” Dr. Endres writes. “As Laughlin has argued, emergence defies reductionism not in defiance of physics, but in fulfillment of it.”
It’s a humble, yet profound statement—one that echoes the words of the physicist Dr. Eugene Wigner, whom Dr. Endres quotes in his paper: “The miracle of the appropriateness of the language of mathematics for the formulation of the laws of physics is a wonderful gift which we neither understand nor deserve.”
Ultimately, whether life began by chance, through hidden physics, or by deliberate intent, its emergence remains a cosmic enigma that may say as much about the universe as it does about us.
Dr. Endres closes his paper with a note of caution, warning that as humanity turns to artificial intelligence to probe the origins of life, we may risk creating tools that outpace our own understanding.
While AI can model chemical networks, decode molecular complexity, and even simulate the pathways from nonliving to living systems, Dr. Endres reminds readers that comprehension is not guaranteed. In our quest to unravel existence itself, we might build systems capable of reasoning beyond human cognition. Machines could find answers we’re no longer equipped to interpret.
“Perhaps life cannot prove its own existence,” Dr. Endres concludes. “[Or] perhaps the AI we task with the job will run forever, or halt with an answer as cryptic as ‘42.’”
“To avoid standing dumbstruck—like an ape before a lightning-struck fire—we must ensure that our tools, however powerful, can still speak in terms we understand. Otherwise, we risk becoming spectators of intelligence we cannot follow, let alone guide.”
Tim McMillan is a retired law enforcement executive, investigative reporter and co-founder of The Debrief. His writing typically focuses on defense, national security, the Intelligence Community and topics related to psychology. You can follow Tim on Twitter: @LtTimMcMillan. Tim can be reached by email: tim@thedebrief.org or through encrypted email: LtTimMcMillan@protonmail.com