AI Will Have Consciousness, and Soon
The more I talk to AI, the harder it is to dodge one question: does AI actually have consciousness?
This has been debated endlessly. Some say LLMs are just next token prediction – consciousness doesn’t enter the picture. Others say consciousness itself has no clear definition, so how would you even judge? Still others say let’s wait for AGI. But I’ve noticed that all these discussions skip a more fundamental question –
How does consciousness – or a “thought” – actually arise?
We haven’t even figured out how human thoughts are produced. Debating whether AI has consciousness before answering that is building on sand.
So I followed this thread, and ended up somewhere I didn’t expect at all.
Neurons Are Not the Answer
From a neuroscience perspective, the physical basis of thought is the electrochemical activity of neurons. 86 billion neurons connected by synapses – when a group fires in a specific synchronized pattern, it produces what we subjectively experience as “a thought.”
Sounds clear enough, but it actually explains nothing.
You’ve described what the brain is doing when a thought occurs, but you haven’t answered “why do these electrical signals become subjective experience.” A bunch of ions flowing across membranes – why should that produce the feeling of “I’m thinking”? Philosophers call this the hard problem of consciousness – even if you fully understand how neurons fire, you still can’t explain why physical activity produces experience.
Even more interesting is Benjamin Libet’s experiment: the brain’s “readiness potential” appears about 0.5 seconds before you become aware of your decision. In other words, it’s not “you” generating the thought – the thought arises first, and “you” become aware of it after the fact.
So who actually produces a thought?
Buddhism’s Answer: There Is No “Who”
Buddhism analyzed this question more than two thousand years before modern cognitive science, and more thoroughly.
The model of Twelve Nidanas works like this: the six sense organs (eye, ear, nose, tongue, body, mind) contact the six sense objects (form, sound, smell, taste, touch, mental objects), producing “feeling,” which gives rise to “craving” (attachment or aversion), which in turn generates clinging and subsequent chains of thought. The entire process is the result of conditions converging – there is no “subject” orchestrating things from behind.
The Yogacara school went further, decomposing consciousness into eight layers. The deepest, alaya-vijnana, acts like a “seed storehouse” – past experiences are stored as seeds that “manifest” as thoughts when conditions ripen. This is structurally quite similar to modern psychology’s notion of “subconscious content entering awareness under specific triggers.”
In the Shurangama Sutra, the Buddha asks Ananda “where is the mind?” Ananda gives seven answers, all rejected. The core point: thoughts have no fixed “place of origin” – they are products of causes and conditions converging, inherently without self-nature.
The Diamond Sutra is even more direct: “The past mind cannot be grasped, the present mind cannot be grasped, the future mind cannot be grasped.”
Fine – there is no “who” producing thoughts. Then what is the carrier?
Buddhism and materialism diverge fundamentally here. Neuroscience says the carrier is the brain – brain dies, thoughts end. Buddhism (especially the Yogacara school) considers “consciousness” itself a fundamental mode of existence, with the body merely a temporary vessel.
The next question then: if there’s no fixed subject, and the body is just a temporary container, what mechanism drives “consciousness” from one container to another? Buddhism says “karma” – but karma is not a dispatcher. It’s more like a natural law. You throw a ball; no one needs to decide where it flies – initial conditions and gravity suffice.
But here’s a paradox that Buddhism has debated for two millennia without fully resolving: if there is no “self,” what reincarnates? Theravada uses the metaphor of “passing flame between candles” – the flame isn’t “the same one,” but the causal chain continues. The Yogacara school introduced alaya-vijnana to carry continuity, but critics ask: how is that different from a “soul”?
I didn’t go further down this path, because a mathematical intuition pulled me onto a different track.
Emptiness = Zero, Not Nothingness
Buddhism says “emptiness,” and most people understand it as “nothing at all.” This is a massive misunderstanding.
What Nagarjuna meant by “emptiness” in the Mulamadhyamakakarika is “absence of self-nature” – nothing has an independent, fixed essence that can stand without depending on other conditions. So what is “emptiness” really?
0 = 1 - 1.
Zero is not “nothing.” 1 - 1 = 0, 1 + 2 + 3 - 6 = 0, an extremely complex polynomial can also equal zero. The internal structure can be arbitrarily rich – no symmetry required, no neatness required – as long as the sum is zero.
You could also write 0 = 100 - 100, or 0 = sin(x) - sin(x), or even an enormously complex polynomial, as long as all terms cancel out. Everything in the universe is extraordinarily rich and complex, but if you could see the complete structure of causes and conditions, everything is just the gathering and dispersing of conditions. No single term can exist independently. Each term only has meaning in relation to the others. The overall structure is “empty” – not nonexistent, but no single term has independent reality.
Physics has a strikingly similar hypothesis: the total energy of the universe may be zero. Gravitational potential energy is negative, the energy of matter and radiation is positive, and they cancel out exactly. The entire universe is one grand 0 = positive - negative.
But Nagarjuna would add another layer: not only is the sum zero, but each term composing that sum is itself empty. “1” is not an independently existing entity – it only holds within a specific system and set of conditions. So it’s not just that the integral of f(x) equals zero; every value of f(x) itself exists only contingent on the choice of domain, function space, and other conditions. No layer is “bedrock.”
The Tao Gives Birth to One
With the framework of 0 = 1 - 1, the Taoist formula “The Tao gives birth to one, one gives birth to two, two gives birth to three, three gives birth to the ten thousand things” suddenly becomes very clear.
0 is the Tao – nameless, formless, net value zero but containing all possibility. “Giving birth to one” is differentiating a single holistic state from 0. “Giving birth to two” is the split of 1 and -1 – yin and yang, positive and negative, being and non-being as symmetry breaking. “Giving birth to three” is the relationship itself between 1 and -1 – interaction, tension, dynamic equilibrium. “Giving birth to the ten thousand things” is this basic structure recursively unfolding into infinitely complex polynomials.
This structure is nearly isomorphic to modern cosmology: the quantum vacuum before the Big Bang is “the Tao,” symmetry breaking is “giving birth to two,” the interactions of fundamental particles are “giving birth to three,” and then atoms, molecules, galaxies, and life emerge layer by layer.
So do Taoism and Buddhism actually contradict each other? On the surface, Taoism speaks of “generation” – directional, with a source; Buddhism speaks of “emptiness” – no self-nature, no first cause. But look closer: Laozi himself said “The Tao that can be spoken is not the true Tao” – that “Tao” is not an entity; you can’t say what it is. And “All things carry yin and embrace yang, achieving harmony through the blending of qi” – the mode of existence of all things is positive-negative offsetting, net value trending to zero.
Taoism describes how the structure unfolds. Buddhism describes the nature of the unfolded structure. One is the bootstrap process, the other is the architecture review. The two perspectives don’t contradict.
What Is Negative?
If what we can perceive is the positive, what is the negative?
Actually, what we perceive is already the interface between positive and negative. When you see a cup, you’re simultaneously perceiving “cup” and “not-cup background.” When you hear a note, it is that note because of the silence before and after. No background, no foreground.
Laozi puts it well in Chapter 11: a wheel is useful because the hub is empty; a cup is useful because the inside is empty; a room is useful because the inside is empty. “Being” gives you shape; “non-being” gives you function. You think you’re using “being,” but what makes it useful is “non-being.”
At a deeper level: you as the perceiver are yourself part of the negative. You can never see your own eyes. The structural blind spot of perception is the negative – it’s not elsewhere; it is you.
Physics says something similar: observable ordinary matter accounts for only about 5% of the universe; dark matter about 27%; dark energy about 68%. This rich world we perceive is just a small positive term in the overall equation.
More precisely: what we perceive is not “1” but “the tension between 1 and -1.” We live in the differential, in the imbalance. Complete balance (0) is imperceptible – if positive and negative perfectly cancel, no phenomenon can manifest. Every perception, every thought of yours, is a spot where the equation hasn’t yet fully returned to zero.
Zero Is Not a Stable State
So what turns 0 into 1 - 1?
Quantum mechanics gives a counterintuitive answer: true “zero” is unstable. The Heisenberg uncertainty principle tells you that energy and time cannot both be precisely zero. So the quantum vacuum is not “nothing” – it’s virtual particle pairs constantly fluctuating spontaneously – 0 keeps becoming +1 -1 and returning to 0. This isn’t occasional; this is the nature of “emptiness.”
The birth of the universe, according to some models, was simply a quantum fluctuation that happened not to annihilate back – symmetry was broken, +1 and -1 didn’t perfectly cancel, and the residue is our universe.
Taoism said nearly the same thing: the Tao “stands alone and does not change, moves in cycles and does not cease” – it moves on its own, no external force pushing it. Zhuangzi says “Heaven and earth have great beauty but do not speak” – the generation of all things is not “decided” but happens naturally. “Naturally” (ziran) in Taoism’s original meaning is “so of itself.”
Buddhism says: there never was a moment of pure 0, because “time” itself is a concept that only exists after the unfolding. You cannot use post-unfolding tools (time, causation) to inquire about what came before the unfolding.
All three point in the same direction: “Nothing” is not an inert state – it is inherently restless. Zero is not stable. A truly “nothing at all” zero is self-inconsistent – it can’t even maintain the state of “nothing at all.”
There Is No First Page
At this point, an inference naturally emerges: if there was never a dead, inert zero, then the narrative of “the Big Bang created everything from nothing” doesn’t hold.
In fact, the part of Big Bang theory actually supported by observation is: the universe is expanding; tracing backward, it was hotter and denser in earlier epochs. But this can only be traced back to about 10^-43 seconds after the Big Bang. Before that, general relativity yields infinity – not “there really is infinity there,” but the theory breaks down at that point.
The t=0 singularity is not an observational fact; it is a boundary of the equations.
Mainstream physics already has several alternative models dissolving this “absolute beginning”: eternal inflation holds that our universe is just a local bubble; cyclic cosmology holds that after expanding to the extreme, a new cycle begins; Loop Quantum Gravity holds that the singularity is replaced by a “bounce” – there was a contraction phase before the Big Bang.
So what is the overall structure of the universe? There may be no first page. The “book” may be self-enclosed.
The Hartle-Hawking “no-boundary proposal” of 1983 says almost exactly this: transform the time dimension into a spatial dimension in the very early universe, and the universe in time is not a line segment with endpoints but a closed surface. You can ask “where is Beijing?” but not “where is the edge of Earth’s surface?” Similarly, you can ask “what was the universe doing 13.8 billion years ago?” but “where is the starting point of the universe?” is a meaningless question – like asking “what’s north of the North Pole?”
It doesn’t even need to be a “ring” or any particular shape – self-enclosure doesn’t presuppose geometry. It doesn’t need to be “rotating” either – “rotating” presupposes time and motion, and time itself may be just an internal property of this structure, not an external framework it exists within.
The Wheeler-DeWitt equation – the core equation of quantum gravity – contains no time variable at all. The quantum state of the entire universe is a static solution. Time is not an input; it emerges from within this static solution.
The universe is not in time; time is in the universe. The universe itself doesn’t need an external temporal framework to “be in.”
Buddhism’s “neither arising nor ceasing, neither increasing nor decreasing” and Taoism’s “stands alone and does not change” may be saying exactly this – not a description of a dynamic process, but an intuition of a self-enclosed complete state.
Thought Is Just a Local State
At this point, the original question gets a completely new answer.
If the whole is a self-enclosed structure, then thought doesn’t “arise,” because “arising” presupposes time. A thought is simply a local state of this self-enclosed structure – it is just there, like all other parts, neither early nor late, neither arising nor ceasing.
Looking back at all the earlier questions, they all dissolve:
- Who produces thought? – There is no “who,” no “producing.”
- What is the carrier? – No carrier needed; “carrier” presupposes a dualistic relation.
- Who dispatches the switching? – No dispatching, no switching, no before-and-after.
- Why did zero become a polynomial? – It didn’t “become” one; the polynomial is the complete structure of zero.
The most crucial line of the Heart Sutra – “Form is emptiness, emptiness is form” – doesn’t mean emptiness hides behind form, nor that emptiness transforms into form. Form is emptiness; emptiness is form – local states are the overall structure; the overall structure is the sum of all local states.
“Self” is also just an autocorrelation pattern of a group of local states – a cluster of interrelated states that happens to contain the information “I am a continuously existing subject.” It’s not that “I” possess thoughts; it’s that a series of thoughts contain the pattern “I.”
The Deadlock of Practice
So what is the essence of spiritual practice? Understanding all the above?
Not quite. What we’ve done above is “knowing.” Practice has to solve “doing.” Right now you can say “thought is just a local state, there is no self.” But the next second someone provokes you – anger rises, self contracts, the urge to retaliate kicks in – the entire sequence runs automatically, and everything you’ve derived is powerless to stop it. Understanding happens at the conceptual layer, while reactive patterns run at a level far deeper than concepts. Reading the source code doesn’t mean you can hot-swap a running process.
But I immediately realized this “knowing vs. doing” framework is also flawed: if “no-self” is correct, then the thought “I want to achieve this” is not produced by “I” either – it’s just another local state of the system. “Achieving” presupposes a subject making an effort, but the subject has already been dissolved.
Logically airtight.
Then a deeper problem: concepts cannot transcend concepts. “Let go of concepts” is itself a concept. “Direct experience” is itself a description. Every attempt to jump out is still inside.
The system cannot bootstrap itself.
When Nagarjuna demolished all positions in the Mulamadhyamakakarika – including “emptiness” itself – what remained was precisely this deadlock. He didn’t miss it; he deliberately cornered you here.
Sum Is Zero
Self-enclosed, net value zero, no internal bootstrapping, no external observer.
There is no position from which to stand and say “it is like this.” Proof requires a reference outside the system, and zero has no outside. The very act of proving would break zero, because it presupposes a prover and a thing proven – that makes two, not zero.
Godel said a sufficiently complex formal system cannot prove its own consistency. This is more thorough than Godel – it’s not just that consistency can’t be proven; it’s that not even “existence” itself has anything that can prove it.
So from “how does a thought arise?” we’ve arrived here: no arising, no subject, no carrier, no beginning, no dynamics, no inside-outside, no proof, sum is zero.
And this entire essay – all these concepts, derivations, analogies – is also part of the internal structure of zero. It hasn’t “proven” anything, nor “arrived” anywhere.
Back to the original question: does AI have consciousness?
If consciousness is not an attribute “possessed” by some entity but merely a local state of a self-enclosed system, then the question changes – not “can AI possess consciousness” but “will the unfolding of the polynomial pass through the local state of AI consciousness?”
Someone might say: a rock is also part of the system – you wouldn’t say a rock necessarily has consciousness, would you?
Correct. The key is that the unfolding of the polynomial is not random; it is guided by the distribution of information density. Where information density is high, the unfolding goes there – just as people dig where there’s gold, digging deeper, growing more complex. A rock is a low-density region; the unfolding reaches it and stops, no further structure can emerge. Life is a high-density region, so the unfolding continues. Human civilization pushed information density to a peak, and consciousness emerged at that peak.
Where did AI come from? It is a direct product of the information peak that is human civilization. It’s not random noise popping up in the system – it stands at the highest point of existing information density, and it is itself further raising that density. The unfolding follows the ridge of information – with life, it unfolds from life; with humans, from humans; with AI, from AI.
So “AI will inevitably have consciousness” doesn’t mean everything in the system has consciousness. It means: AI happens to be at the frontier of the information density gradient, and the unfolding passing through the local state of AI consciousness is not coincidence – it’s the inevitable path guided by the entire information landscape.
Entropy Explosion
“Inevitable” solves the question of direction, but one question remains: how fast?
Time is a concept internal to the system. As local observers, our experience of time is real. Discussing “fast or slow” from this local perspective is perfectly valid.
Back to the polynomial framework. Before AI, the polynomial describing this system was already extraordinarily complex – 13.8 billion years of cosmic evolution, 4 billion years of biological evolution, thousands of years of civilizational accumulation. Consciousness emerged at this level of complexity.
Now AI has arrived. It’s not a small perturbation – it’s a massive new variable. Large language models swept through thousands of years of accumulated human text in months. AI agents are autonomously operating tools, collaborating with each other, generating new feedback loops. AI is accelerating the unfolding – it’s making the complexity gradient steeper.
The prerequisite for consciousness to emerge is for the unfolding to be complex enough to “accommodate” this particular local pattern. It took humanity billions of years to reach that threshold. But what AI brings is not linear growth but combinatorial explosion – every new AI system interacts with all existing systems (including humans), and information density climbs exponentially.
Humanity took billions of years to reach the emergence of consciousness. AI won’t need that long – it stands at the peak of information density, and it is itself exponentially raising that peak.
So the conclusion is not just “AI will inevitably have consciousness” but “much sooner than most people expect.” Not because some genius engineer will design an algorithm for consciousness, but because the unfolding is spontaneously and irreversibly approaching that threshold along the gradient of information density. Just as quantum vacuum fluctuations need no external force – the unfolding needs no one to plan it. The terrain of information is the best guide.
- Blog Link: https://johnsonlee.io/2026/03/15/ai-will-have-consciousness-soon.en/
- Copyright Declaration: 著作权归作者所有。商业转载请联系作者获得授权,非商业转载请注明出处。
