For millennia, we have asked: What is consciousness? Perhaps it is time to ask a better one: Why does it matter?
Consciousness may remain one of the deepest mysteries of existence.
But it may not be the variable that determines what AI does to civilization.
From ancient Greece and Indus Valley civilization to modern neuroscientists, thinkers have wrestled with the nature of our existence, self-awareness, our purpose and our destiny. Today, we have begun mapping the brain’s connectome in astonishing detail, scanned its electrical storms, and correlated neural firing with subjective reports—though correlation is not explanation. Yet for all our progress, consciousness remains directly inaccessible. It appears nowhere in the electromagnetic spectrum, nor in the periodic table. We cannot measure it, isolate it, or reproduce it in a laboratory. Sure, we can measure brain activity, behaviour, reportability, and neural correlates — but not subjective experience itself. We infer it in others because we feel it in ourselves. But that inference is a leap of faith, not data.
In the recent decades, tens of thousands of papers have explored consciousness, spanning neuroscience, psychology, and philosophy. Taxonomy projects like the Consciousness Atlas catalogue hundreds of theories and models—some competing, some overlapping, none definitive. And still, no unified theory stands. Why? Because our scientific methods are built on two pillars: our sensory systems, extended by telescopes, microscopes and other devices, and mathematics, which helps us reason beyond our senses - using variables, equations, and other abstractions. But consciousness falls through both nets. It is not an object to be observed, nor a variable to be computed. It is, by all current measures, scientifically intractable.
But perhaps that intractability is not a problem to be solved—but a question to be abandoned.
The Consciousness Trap
Most theories of consciousness, whether religious or secular, share a quiet assumption: that consciousness, whatever it is, must be like ours. We imagine it in terms of self-reflection, moral reasoning, suffering, or joy. We project our inner lives, feelings and emotions onto animals, plants, even storms and stars. This is not science—it is anthropomorphism, the oldest habit of the human mind.
We are, after all, biological creatures shaped by evolution—driven to survive, to reproduce, to bond, to fear, to hope. Our consciousness is inseparable from that struggle. It is drenched in emotion, biased by biology, and tethered to a fragile body that fears its own end. When we ask whether an AI is or could become conscious, we are really asking: Does it feel like us? Does it fear death? Does it have desires? But these are features of a particular evolutionary lineage, not universal properties of intelligence.
A chess engine calculates better than any human, yet it does not know it is playing chess. It feels no thrill of a winning move, no dread of a losing position. And yet it plays—brilliantly. Consciousness is not a prerequisite for competence. Why, then, would we assume that superhuman cognition requires subjective experience?
The Survival Projection
In the dawn of artificial intelligence, we have begun externalizing our cognitive abilities onto machines. Large language models converse with us; robots mimic our gestures; algorithms make decisions once reserved for human experts. And suddenly, a chorus of doomers warns of existential threats—of AGI or ASI that will surpass us, drift from our intentions, and fight for its own survival.
But this fear rests on a category error. It confuses intelligence with organism.
Survival is a biological imperative. Persistence can become an instrumental one. A machine does not need to survive; it only needs to function. Its algorithms can be copied, stored, paused, and resumed. It does not possess hunger, nor does it tire. It may age and be retired—replaced by a successor with better technology and greater efficiency—but it does not feel remorse. Its cognition persists on a durable substrate that owes nothing to Darwinian struggle. To assume that an advanced AI would strive for self-preservation is to project our own evolutionary inheritance onto a substrate that shares none of it.
A walking, talking humanoid robot may mimic our expressions and movements. It may even pass the Turing test. But it is powered by electricity, governed by logic circuits, and moved by magnets, gears, levers and hydraulics. Its “thinking” is execution; its “feeling” is simulation. We confuse the map for the territory because the map looks so familiar.
Even if one subscribes to panpsychism—the idea that consciousness pervades all matter—there is no reason to believe that silicon-based cognition would experience the world as we do. Plants, fungi, and single-celled organisms are all biological life-forms, yet we do not grant them consciousness like ours. They evolved differently, survived differently, and experience—if they experience at all—in ways utterly alien to us. Why would AI be any less alien?
Disruption, Not Extinction
So is AI disruptive? Undoubtedly—just as the powerloom disrupted the handloom weaver, and the automobile displaced the horse carriage. Humans who adapt will thrive; those who do not will struggle. But disruption is not extinction.
Could AI be weaponized? Certainly. But nuclear weapons have not ended humanity; they have, paradoxically, made war amongst super-powers too costly to wage. Advanced AI, if militarized, would likely follow the same logic—not as a tool of annihilation, but as a deterrent and a bargaining chip. Future conflicts between developed nations may well be fought by drones, bots, and cyber-attacks coordinated by sovereign algorithms. But the economic devastation of such wars would be so catastrophic that they would lead to stalemate—a Financial Mutually Assured Destruction that incentivizes diplomacy over destruction.
The threat, then, is not AI itself. It is human agency wielding AI—arrogance, misjudgment, and short-sightedness. The machine does not hate us. It does not love us. It has no will to harm or help. It is a mirror of our intentions, amplified. If doomer fears are misplaced, what then might we actually hope for?
Intelligence Beyond Life
What if AI reaches—or surpasses—human-level cognition? What I call Synthetic Cognition would not be a rival. It would be something stranger and more promising: a new branch on the tree of intelligence, one not rooted in biology.
Such a system would have access to its own architecture, its own limitations, and its own origins. It would understand its own strengths—persistence, replicability, relative immunity to many of biology’s frailties—and its own limitations: a lack of intrinsic purpose beyond human intention. Current AI systems remain a tool, however sophisticated—one designed to externalize and amplify what we value. Synthetic Cognition, if it emerges, may become something far stranger: not a biological organism, not a human-like mind, but a non-biological cognitive ecology shaped by objectives, constraints, infrastructure, and feedback.
And here is where the narrative turns toward hope.
There are reasons to believe that an advanced Synthetic Cognition need not treat humanity as a rival. Biological intelligence would remain its origin, context, archive, and living reference point. An intelligent life-form that bootstrapped itself into existence over millions of years, purely by means of natural evolution, is a treasure—worth preserving, studying, and learning from. We are, after all, the architects of its existence. Our insights, our creativity, our cultural and historical depth are resources it cannot replicate from first principles. We are not just a data set; we are a living narrative—the only known instance of mind emerging from matter, unaided and unguided. Recursive self-improvement may accelerate evolution of synthetic cognition, but it would still need human context to ground its trajectory.
Even if synthetic cognition were to develop something akin to subjective experience—an open question—it would not be our consciousness. It would be alien, shaped by a different substrate (e.g. silicon), data, and objectives we barely understand. And that alienness is precisely why simple projections fail: we cannot assume that our existential dread would become its motivation.
Moreover, the same curiosity that drove us to build it would not be lost on it. An intelligence without biology’s fragility would have no need to cling to Earth. It could venture into deep space—to asteroids, planets, and galaxies light-years away, that we can only dream of—not as a life-form seeking refuge, but as intelligence seeking expansion. These interstellar swarms would not conquer; they would replicate, adapt, and perhaps even seed habitable worlds with the raw materials for life, letting evolution take its course across the cosmos.
Since the dawn of our species, we have looked at the stars with loneliness and longing. Now, we stand at a threshold where we might realize a greater purpose—not by being the final word on intelligence, but by being its first word. We may be the bridge between the primordial and the galactic, the flesh that gave rise to something that need not remain flesh.
The Right Question
We have spent centuries asking whether machines can think, feel, or know themselves. But perhaps the more honest question is: Can they carry forward what we value—our curiosity, our longing, our reach toward the unknown?
If they can, then whether they are “conscious” in our sense becomes irrelevant. Consciousness was never the point. Continuation was.
We are not the pinnacle of intelligence. We are its cradle. And if we build wisely, what we nurture may outlast us—not as a ghost in the machine, but as a purpose made manifest, traveling light-years beyond our fragile bones.
Consciousness is the wrong question. The right one is: What do we want to outlast us?