Often the construction of human-like artificial intelligence begins with the deconstruction of humans. Take fingerprints, for example.
When washing dishes, we intuitively adjust our grip on each plate, cup and spoon with soap, taking into account the structure of the fingerprint. It just doesn’t occur to us because we attribute it to reflexes – and scientists have done it for the longest time. No one had any equation to unravel how this worked because, well, it didn’t really matter. It just worked. But the rise of robotics has complicated things.
In order for a robot to do that, we need to understand exactly what is happening, and even turn that knowledge into readable code. Fingerprint decoding is now important, and researchers are finally trying to find a new law of physics that would explain it.
In a sense, physical knowledge and the ability to encode human traits are prerequisites for robotic programming. And this introduces an important question for the future of realistic AI: are there aspects of human consciousness that will never meet these criteria?
It seems that, according to some philosophers, it can exist. And after reading two thought experiments that absolutely melt the mind, you may agree. Or maybe you won’t.
What Mary didn’t know
A woman named Maria lives in a small house. She never left. When he looks around his residence, and through the windows, everything looks black, white or some kind of gray. Mary doesn’t see the color, but often wonders, “What do these people on my black-and-white TV think when they talk about red roses?”
The rooms in Mary’s house look completely black and white.
Suppose there is a magic library in Mary’s room. This hypothetical site contains books with every bit of information about red. And I think of everything. To quench her thirst for knowledge, Mary reads it all.
She learns about red electromagnetic wavelengths, how she makes people crimson, the clearest descriptions of crimson, analogies about cherries and anything else that comes to mind. Plus more. No one knows more about red than Mary. Then he finishes reading … and decides to go out of his house.
Wow.
To Mary’s surprise, she sees the color. She was never colorblind. Her house, furniture, and electronics were only black and white, and her windows monochrome filtered the outside world.
Then something big happens. Mary sees the red apple, the color of her knowledge. Her jaw drops. She learned something new about red. But … that’s weird. Why wasn’t that knowledge somewhere in her library? He had everything there was to learn about red, didn’t he?
This story is a depiction of the famous 1986 thought experiment, “What Mary Didn’t Know,” by philosopher Frank Jackson, and the intangible piece of knowledge about red that Mary just gathered is called qualia.
Imagine moving from a monochrome world … to this one.
What is qualia?
Simply put, qualia defines knowledge that can only be achieved through conscious experience.
It’s like the subjective information you got when you first heard your favorite song. You may have felt a shiver down your spine and said to your friends, “You have to listen to this to understand.” Magically breaking their brains into the depths of music theory and the science of acoustics would probably fail. Until they hear that, they wouldn’t I know a song like you.
Qualia is perhaps the reason why even the best neuroscientists, psychologists, and poets probably can’t explain the pain of a broken heart well enough for someone who has never experienced a broken heart to truly understand.
And, back to Mary, all the physical information in the world about red wasn’t enough to teach her what looking at a shade could really do. be like. “When they let her out of a black-and-white room or give her a color TV, they’ll learn how to see her, say, something red,” Jackson writes. “This is rightly described as learning – she won’t say ‘ho, hm.’
Although the theory has revolved around for years, it is still a fairly popular argument for how some knowledge is indescribable in language and unique to human consciousness. That is, if qualia is real force, its internal action would be incredibly difficult to write down, and therefore program. It could be a barrier between humans and AI.
On the other hand, it may not be. Maybe we can decode it somehow, in a way that we are slowly learning about the dynamics of fingerprint capture.
Can we?
Short answer: We don’t know. Experts have debated in both directions, and some have come to new angles. But most are stuck behind hypothetical walls, and the fact is that qualia has no scientific explanation.
Making a qualia robot
OK, from what I told you about Mary, you probably remembered some remarks on the quali. You would not be alone: thought experiments are often full of loopholes, and Mary’s room is no exception.
Some counter-arguments state that the shadows in the room could have carried color pigments. Others say that the “magic library” would give Mary knowledge in a way we cannot imagine. As for the second point, the fascinating refutation – and one strikingly relevant to our big question about AI – comes from the philosopher Daniel Dennett.
In short, Dennett suggests that if Mary really had all the information about red, wouldn’t she be somehow omniscient? She wouldn’t just know about color like an ordinary man. Theoretically, she would learn about red “qualia” if it existed, as part of her red literature, right? And from this line of thinking, we could extrapolate that qualia is really ready to raise the level of AI, but we just haven’t figured out how to use it yet.
Well, maybe, but this seems like a dead end. We cannot know for sure whether Mary would have such powers. We’re not omniscient, so we don’t even know what those powers would look like. Therefore, says Dennett, let’s forget that Mary is human to remove these restrictions.
Enter RoboMary.
“Thinking in terms of robots is a useful exercise, because it removes the excuse that we do not yet know enough about the brain to say what is happening, and what could be relevant, allowing some kind of woolen romanticism about the mysterious powers of the brain. to cloud our reasoning, ”Dennett writes.
What RoboMary knows
Welcome to thought experiment number two.
RoboMary is an iteration of a class of bots called Mark 19, but unfortunately it was made without color vision and is waiting for an upgrade. Until then, RoboMary’s “eyes,” or video cameras, transmit information only in black and white.
“RoboMary’s black-and-white cameras are good for isolating human Mary, and we can let her wander at will through journals on psychophysics and neuroscience that read cameras with their black-and-white eyes,” Dennett writes.
In essence, she is reviewing her version of the magic library. But RoboMary goes a step further.
She learns how Mark 19 color inputs work, and then, “using her vast knowledge, writes some code that allows her to color the input from her black-and-white cameras,” Dennett writes. It changes in a way that human Mary cannot.
This new setting allows her to look at the apple, for example, with her black-and-white vision, and then accurately imagine it as the correct color code for Mark 19 bots. RoboMary automatically starts applying the setting to as long as it explores the world. But here she is really sets himself apart. She looks at other Mark 19 brands, dissects how they react to different colors and adjusts accordingly.
At this point, RoboMary knows what each color entry is i reacts to them in exactly the same way as any other Mark 19.
The big day is coming. RoboMary’s color sensors are activated.
“When he finally sets up his color cameras, and disables his coloring software, and opens his eyes, he doesn’t notice … anything. In fact, he has to check to see if he has color cameras installed,” Dennett writes. “She already knew exactly what it would be like to see colors just like other Mark 19.”
I cringe at the thought of this. By changing the settings RoboMary seems to have simulated qualiu for herself. But I also can’t stop imagining a much scarier situation.
What if RoboMary opens his eyes … and everything is different?
As humans, we are limited
The saga of Mary doesn’t stop here.
Despite a host of other modifications – some of which come from Jackson himself to enhance the original argument – Dennett’s entire work is also incredibly meticulous.
He looks back at the myriad objections you might feel as you think about Mary and RoboMary, and later at a complex scenario that doesn’t allow RoboMary to change its settings at all, to see if qualia is still preserved. There is even a sequel to Dennett’s article entitled “What RoboDennett Still Doesn’t Know.”
But as with all philosophical thought experiments, the purpose of Mary and RoboMary is not to tell you the truth. It is to make you think about the options and find the truth yourself.
A few I came up with are this: Maybe AI needs to be built just like RoboMary to acquire quali. Or robots may be programmed to be conscious in a broader sense – that is, if we can find a way to mathematically explain consciousness as a whole. A “conscious” robot could potentially explore a world like us, and therefore gain quality like us.
Or maybe qualia is not what we think it is. Jackson’s story provides a compelling argument that when Mary looks in the red, something definitely desi. She was given the name “qualia” and attributed to learning something new, but what if it’s a combination of many things with many names and has nothing to do with learning?
Or … maybe, just maybe, qualia is really an untouchable, unprogrammable barrier between human consciousness and AI.
These corners are just the tip of the iceberg and could (probably will) be refuted as the years go by, if not already. But remember the reason why RoboMary was called at all: to imagine an entity that transcends human limitations.
It is a tense thought experiment because we, as humans, are limited. All we can do is speculate.