Philosophizing Consciousness
In conversations about tech, we often hear the term “consciousness.” Why, though, is it such a popular idea that’s so frequently referenced? Furthermore, what can it help us understand about the nature of our own human experience, and what, moreover, can be taken away from these discussions to inform the technology we’re creating?
To investigate this topic, first, it is important to consider the unique characteristics humans possess. If we can figure out what separates us from our technological creations, we can discover hints of what uniquely defines the human experience. I propose that we have an experience of the world paired with information processings.
This is a challenging concept to think about, and an even harder one to explain, but it certainly appears to be a factor that distinguishes natural life from artificial technology. There doesn’t seem to be a definitive point at which we would consider it conscious. Consciousness is a binary state, but at what point in the progression from a lifeless rock to a human being does it arise? After adding a thousand transistors to a circuit board? A million? From a strictly materialist point of view, it seems illogical that any quantity of material substance could give rise to immaterial experiences. The philosopher David Chalmers explains the absurdity of this existence:
“Even after we have fully explained the performance of all the relevant cognitive and behavioral functions (discrimination, integration, internal access, verbal report, and so on), a further question remains open: Why is the performance of these functions accompanied by experience? Why doesn’t the information processing simply go on “in the dark,” free of any inner feel? When electromagnetic waves hit a retina and are processed by a visual system, why is that processing experienced as the sensation of vivid red rather than occurring without any subjective quality at all?”
Consider for a second how odd it is that we have these inner experiences, especially when they’re totally unnecessary for our function. Puzzling, right? Nearly every other feature we possess is adapted with some objective in mind (usually sustaining our life or continuing our species).
When it comes to actually creating another conscious being (like an artificial intelligence), the distinction between appearances and reality must be considered. It’s highly plausible that an advanced AI could externally simulate the experience of a conscious being through behavior, mannerisms, and claim, without even having these convictions stem from an internal experience. Likely, these “convictions” are imitations of our human convictions.
This idea of imitation is often connected to the “Chinese Room Experiment,” a thought experiment from the 80’s by John Searle, which demonstrates the possibility that a computer program could execute a task perfectly without ever understanding the reasoning underlying the task. Searle talks about a person in a room, who has no understanding of any language besides English, and follows translation instructions from a book written in English. He argues that if the book is comprehensive enough, the person could translate without even understanding Chinese:
Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a database) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test for understanding Chinese but he does not understand a word of Chinese.
I was thinking about this experiment, and while it seems to be an accurate representation of our current AI situation on a surface-level glance, I don’t think it models what we consider to be “understanding” in a way that applies to modern AI. The issue ultimately depends on whether an AI model could configure its settings in such a way to produce consciousness (which is a possibility for how it happened for us humans). If an AI were capable of this,
As we develop technology, we’ll be necessitated to develop a deeper understanding of ourselves and questions like, “What does it mean to be human?” The issue of consciousness seems like a good place to start this journey.
