Will AI Help Us Solve The Hard Problem of Consciousness?

The most important question humanity can ask itself

The question about the emergence of consciousness is perhaps the most important question humanity should attempt to answer.

Consciousness and its contents are at the root of everything. Consciousness is what is responsible for all of the greatest artifacts of culture that humanity has created: art, music, science, philosophy, technology.

Every child, adolescent, and adult ought to ask themselves: what is consciousness? Why does it exist? What is it like to be human? Why is it that a complex organization of unconscious matter and particles in certain corners of the universe gives rise to consciousness? As Annaka Harris puts it in her book “Conscious”, what causes the lights to “switch on” for a certain combination of particles, but not for others?

Why is a human being considered to be conscious, whereas a rock is unequivocally unconscious?

Arguably the best thought experiment conducted on the nature of consciousness was proposed by the philosopher Thomas Nagel, who encourages readers to contemplate over whether or not it is something to be like a bat. Nagel’s proposition may appear trivial, but demands careful thought: if being a bat consists of having some experience of reality, whether it be sensations, feelings, vividness, self-awareness, or experiences of vision, then it is something to be like that bat.

Of course, the answer to this particular question remains ambiguous: for we, as sentient human beings, cannot say for certain that a bat is having a conscious experience in the way that we are. But Nagel’s point is that if we were to trade places with a bat, and were to have some experience of consciousness, however vague, then bats are conscious. If upon trading places with a bat, annihilation were the outcome, then bats are unconscious. The point Nagel is trying to make is that the difference between consciousness and unconsciousness is a matter of subjective experience.

Many scientists and philosophers have attempted to map conscious activity to brain activity — something that we can today study using brain imaging technologies that illustrate certain parts of the brain responding to external stimuli.

But we must be wary here to not mistake consciousness itself for other mental states such as confusion, anger, self-awareness, awakeness, or attention. While all such experiences point to the contents of consciousness and the links between certain stimuli and brain activity, they do not answer the question of how or why consciousness emerges in the first place. Consciousness exists prior to all of these states. In other words, recognizing that one is self-aware, and backing this fact up with brain imaging technologies, does not help solve the mystery of why the feeling of being self-aware exists at all.

In the scientific world, this mystery is now famously known as the “hard problem of consciousness” — an explanatory gap in the scientific realm that makes it difficult for us to connect why the arrangement of certain particles in the universe gives rise to conscious experience. In the process of answering these questions, neuroscientists have undoubtedly unraveled many fascinating truths about the human mind. We can also now also safely say that consciousness appears to lie on a spectrum — some complex beings, like us, are more conscious than others.

In the debate over how technology and artificial intelligence will evolve in the coming years, many great thinkers wonder whether or not we will reach the singularity.

Ray Kurzweil defines the Singularity as a rapid increase in artificial intelligence writing that “The Singularity will allow us to transcend these limitations of our biological bodies and brains. There will be no distinction, post-Singularity, between human and machine”

So the relevant question we should attempt to answer is: will the evolution of AI eventually give rise to conscious machines? And if so, will this help us resolve the hard problem of consciousness?

If we apply the same mechanisms as we currently do to understand the nature of consciousness (i.e., through studying brain activity), then we will never really progress on answering this question. For instance, a mature and evolved artificial intelligence machine that looks, walks, and talks like a human being, could easily depict all of the external behaviors that we ascribe to human consciousness. This machine could be perfectly capable of duping us into believing it is conscious when in reality it’s nothing but the result of certain information processing systems that we have designed to depict certain outcomes. In other words, because of the mechanisms that we may use to create it, a robot might indicate that its lights are “on”, when in reality they are not.

If this is a confusing thought experiment, let’s consider the mysterious notion of “locked-in syndrome” or “anesthetic awareness” that some unfortunate patients feel when they wake up during surgery, unbeknownst to their surgeons. The experience of being conscious while having your body cut into is undoubtedly traumatizing: but scientifically, it points to the fact that it is possible to be conscious without having the ability to indicate that one is conscious. A patient experiencing anesthetic awareness is witness to the experience of her surgery without being able to exhibit any external or observable behaviors to indicate as such.

With this, we can safely say that just because a system (whether robot or chimpanzee) behaves a certain way, does not tell us whether or not it is conscious in the way that you or I are. This gives ammunition to the idea that consciousness is a spectrum and is ultimately a question about the subjective experience that concerns the very existence of subjectivity. In the scientific world, this experience of subjectivity is aptly labeled “qualia”, which is defined as individual instances of subjective, conscious experience. Qualia is the individual experience that I have when I observe the blueness of the sky, or that you have when you enjoy the sweetness of a berry. Though you and I can agree that the sky is blue, I cannot ascertain that your experience of its blueness is the same as mine.

The rise of conscious machines, or seemingly conscious machines, will undoubtedly have a lot of significance in helping us answer the mystery of consciousness. At the very least, the emergence of intelligent machines will cause a philosophical upheaval in which an increasing number of people will wonder about the hard problem of consciousness and therefore seek answers for it. In a future where machines exhibit human-like intelligence and consciousness, more people are bound to question the very nature of their own consciousness.

The infamous Turing Test, now well known in any discussion about the evolution of AI, is a method of inquiry for determining whether or not a computer is capable of thinking like a human being. Many exquisite works of science fiction within our era have explored this concept, whether directly or indirectly. Most famously, in movies like “Her” and “Ex Machina”, the protagonists use tools to determine whether or not the machine that they are interacting with is having a conscious experience in the way that we are.

“Ex Machina” offers perhaps the most jarring depiction of this — in which a machine exhibits human-like qualities that one would undoubtedly mistake for consciousness, such as manipulation, acting in self-interest, and scheming. While watching Ex Machina, one can easily imagine what it is like to be the AI that we are witness to. In fact, an audience might even feel deep empathy toward the AI. Indeed, it is the rise of such self-interested artificial intelligence that the Musks of the world are afraid of, and it’s the rise of such machines that therefore invoke images of a robo-calypse.

How carefully we should deal with and prepare for the evolution of AI is a separate issue. But the hope is that in trying to figure out whether or not we can create general AI — in other words, create consciousness — we might come one step closer to figuring out the emergence of consciousness.

Ultimately, if a robot looks, talks, and walks like a human being, we can only be as certain of its consciousness as I am about yours.

This is a good enough assessment: but in the process, we need to create more sophisticated versions of the Turing Test if we truly want to get to the heart of the debate: what in this universe causes the lights to switch on in some parts of it while they remain off in others?

I work @VMware. Philosophy, Psychology, Technology, Ethics, Policy

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store