Many people take it for granted that humanity could one day create conscious, thinking machines. This kind of Artificial Intelligence has been explored in media since the dawn of science fiction (think I, Robot or Terminator). Let’s be clear though — we’re not talking here about the baby AI that lets your smartphone send you invasive ads. This is something more: a fully aware being created artificially from non-biological components, a robot that can fall in love, that can feel pain, that has its own hopes and dreams.
But, as with many things that people take for granted, the idea that such a thing could be created involves a number of serious philosophical commitments. If we are to create consciousness out of metal and wires, then we have to accept that consciousness can be created out of metal and wires. This often means accepting a certain view about ourselves as well; namely, that our own consciousness is merely the product of physical operations (presumably in our brains), and can’t be attributed to anything non-physical like a soul or a spirit. Indeed, if you are religious, the entire idea of AI in this strong sense should strike you as deeply implausible and problematic — maybe we can create a machine that acts like a person, but can we really create a soul in a lab? Probably not.
The idea that consciousness can arise out of purely material components like brains or wires is part of a larger philosophical worldview called materialism. According to that view, the most basic elements of the world are all material things. It follows then that consciousness must somehow be constructed on a purely material foundation. That might sound about right to you, but materialism is just one viewpoint among many.
The fundamental question here is “is the world physical or nonphysical?” Materialism says it is all basically physical. For my part, I’m not going to say much here about whether or not materialism is true or not. There are a lot of questions that materialism has a hard time answering, but then that’s true of most philosophical positions. Instead, I’m going to explore some of the other answers to this fundamental question. Materialism says it’s all physical. Most religions, by contrast, are dualistic. That is, they say that the world is partly physical and partly nonphysical; there are rocks and trees but also souls and god(s). These two positions are relatively well-understood. But there are two other positions that are less common. Idealism, for example, says the world is entirely nonphysical. And there is even a position that says that it’s actually something else that is at the bottom of both the physical and non-physical parts of the world.
Let’s take idealism first. Have you ever wondered what things are like when you’re not looking? According to idealism, when no one is looking, there is literally just nothing there. Or, to be more precise, there is nothing to what a thing is aside from our perceptions of it. This idea may sound kind of insane, but it actually makes a lot of sense. Imagine a fish, for example. Now take away the smell of it, the sight of it, the sound of it, the touch of it, the taste of it… what’s left? Nothing. If you take all the sensory elements out of it, the word ‘fish’ is just an empty sound, signifying nothing. Follow this train of thought to its conclusion and you realize that when we talk about material objects, we are really just talking about sets of commonly associated perceptions. The word ‘fish’ is just a convenient shorthand for a certain pattern of smells, sights, and sounds. And in fact, for an idealist the word ‘material’ itself just denotes the fact that certain perceptions are ones that everyone else can also have (where ‘mental’ denotes more private perceptions — hence why hallucinations are called mental).
In this way, idealism is essentially showing how what we take to be the material world is really a purely perceptual, immaterial, universe. This doesn’t mean that there is no such thing as rocks or trees, but it does mean that rocks and trees don’t have any existence independent of perception. In other words, if all the perceivers in the world died, the rocks and trees would go with them.
Now then, what would an idealist say about AI? Simply put, an idealist would find the entire idea of AI laughable. To them, consciousness is the basis of all existence; nothing exists that is not at least a possible object for consciousness. How then could consciousness itself be based on those objects? This would be like saying first that a foundation supports a building and then turning around and saying that well, the building also supports the foundation. For an idealist, trying to create AI is an exercise in self-deception and futility. We could create something that acts human, but it could never have the capacity for consciousness. You can fall in love with a robot because you have consciousness, but a robot can never fall in love with you — after all, the robot is just a pattern of perceptions with no inner life to speak of.
So that’s idealism. Idealism in this philosophical sense has mostly fallen out of fashion in the western world, but there were periods of history where it was the dominant viewpoint. Idealism can count among its adherents such influential figures as Berkeley and Kant, and that’s only the idealists in the western philosophical tradition. In the east, Buddhism and Hinduism are both popular religions that take the entire material world to be merely illusory, a mental projection like a dream, based on delusions and past sins. So while idealism may not feel natural to a scientifically-oriented person raised in the US, it is by no means a position to be dismissed out of hand.
We turn now to a fourth position. This position says that the world is neither fundamentally material nor fundamentally immaterial. Instead, it posits something else, a third kind of thing, and tries to show how both physicality and consciousness can develop out of it. There are a few different versions of this perspective, but here I’ll be presenting one that is sometimes called panlogicism. Panlogicism, often attributed to the philosopher G. W. F. Hegel, is the view that all things (‘pan’) are fundamentally logical in nature (‘logicism’).
Ok, well that probably sounds like a bunch of gibberish. Let’s take a second to see what this could actually mean. Have you ever used logic to think about the world? Maybe you’ve thought something like “if I jump in the pool, I’ll get wet” or even just something as basic as “that chair is blue.” Now here’s a question that might blow your mind: why do you think that the logical rules you use to think should actually line up with how the world is organized? Why should the world be set up to suit your ‘if-then’ statements? Why should the world be divided up into things with properties that do stuff just like our language is divided up into nouns and adjectives and verbs? Is it all just a happy coincidence?
This question is an important one and there are, as I’m sure you’ve guessed by now, a few different answers. Idealists could say logic and the world line up because the laws of thought (logic) are projected onto objects as we collect experiences and label them. Materialists could say that the world has these structures already and then our language and thought evolved to match them because not matching was maladaptive.
But panlogicists say otherwise. For them, these logical structures aren’t static organizational devices we just apply when we think. Instead, they are independent organizing principles of the universe, dynamically evolving and changing, albeit not in space and time. Just as a train of thought leads logically from one idea to the next, logic itself leads from one concept to another. At some point the “concept” of a material world is generated and boom! that’s how the material world comes to exist. A similar story of conceptual development can be told about how consciousness came about. The details of these are quite complicated, but the bottom line is that this purely logical development leads to both the material and the immaterial universes. In this way, logic is said to be the fundamental structure of the world.
What bearing does this have on AI? Well, like an idealist, the panlogicist would be a bit wary of AI. After all, they don’t really believe that the right assemblage of material components can lead to consciousness. That said, it is possible that AI could come about in the course of the world’s logical development. And that logical development is manifested (in part) through the actions of human beings, so maybe our attempts to create AI out of metal and wires is actually part of that logical development. But if we do create AI, it won’t really be because of some physical laws governing consciousness; it will be because AI is part of the conceptual progression of reality.
So there you have it. It turns out that there are a lot of ways to think about what reality really is, and depending how you think about it, you’ll have quite different ideas about whether AI is really possible. If you’re a materialist, you’ll just be looking to assemble the right material components in the right order. If you’re a dualist like most religious people, you’ll probably dismiss the idea out of hand — it’s souls that think and feel, and we can’t create those, so we can’t create a machine that thinks or feels. At best we could create something that does a perfect impression of humanity while still being hollow inside. If you’re an idealist, you’ll also throw out the idea of AI, since building consciousness out of the material world would be about as likely as building an egg out of omelettes. Last, if you’re a panlogicist (which you probably aren’t, let’s be real), you’ll accept the possibility of AI, but not on the basis of the physical sciences, but rather on the basis of the logical development of reality.
All these positions have any number of additional details and considerations; whole books have been written on a single argument for one of these positions. But while this article may not have gone into great detail on them, I hope it has given you at least a taste of each. Hopefully, the next time you see someone fall in love with a robot in a movie you can take a step back and think about what’s really going on there; is their love a self-indulgent fantasy or does the robot really love them back?