Is general artificial intelligence possible?

The first use of the term artificial intelligence is what should more accurately be called "narrow artificial intelligence". It's a powerful technique, but it's also fairly simple and straightforward: you take a bunch of data about the past, use a computer to analyze it and find patterns, and then use that analysis to make predictions about the future. This type of artificial intelligence touches all of our lives multiple times a day as it filters spam out of our emails and provides us with traffic routes. However, since it is trained using past data, it only works if the future is similar to the past. That's why it recognizes cats and plays chess, because they don't change on an elemental level from day to day.

#Another way the term artificial intelligence is used is to describe what we call General AI (or Artificial General Intelligence, or AGI). Except in science fiction, it doesn't exist, and no one knows how to make it. General artificial intelligence is a computer program that is as intellectually versatile as humans. It can teach itself completely new things that it has never been trained on before.

The difference between narrow artificial intelligence and general artificial intelligence

In the movie, general artificial intelligence is Data in "Star Trek", C-3PO in "Star Wars" and "Blade Wings" The clones in "The Hitman". While it might seem intuitively that narrow AI and general AI are the same thing, just less sophisticated and complex to implement, this is not the case. General artificial intelligence is something different. For example, identifying spam is not computationally equivalent to true creativity, whereas general AI is.

The author once hosted a podcast about artificial intelligence called "Voices in AI". It's an interesting thing because most of the great practitioners of this science are accessible on this podcast and they're willing to be on the podcast. So I ended up with a gallery of over a hundred great AI thinkers talking in depth about this topic. There are two questions I ask most guests. The first question was, “Is general artificial intelligence possible?” Almost everyone—with four exceptions—said “yes, it is possible.” Then I would ask them, when can we build it. The answers range from as fast as five years to as long as 500 years.

Why is the gap so big?

Why do almost all of my guests say general artificial intelligence is possible, yet provide such a wide range of informed estimates of when we might achieve it? The answer goes back to a statement I made earlier. We don't know how to build general intelligence, so your guesses are pretty much the same as everyone else's (useless).

"But wait!" you might say. "If we don't know how to make it, why do experts so overwhelmingly agree that it is possible?" I ask them this question too, and I usually get a variation of the same answer. Their confidence that we will build a truly intelligent machine is based on a core belief: humans are intelligent machines. They reasoned that because we are machines and have general intelligence, it must be possible to build machines with general intelligence.

human-machine comparison

What is certain is that if humans are machines, then these experts are right. General intelligence is not just possible, it is inevitable. However, if it turns out that people are more than just machines, there may be some aspects of people that there may be no way to replicate in silicon.

What’s interesting is the disconnect between these 100-plus AI experts and everyone else. When I talk about this topic to a general audience and ask them who thinks they are a machine, about 15% of people raise their hands, which is far less than the 96% of artificial intelligence experts.

In my podcasts, when I argue against such assumptions about the nature of human intelligence, my guests will often accuse me—politely of course—of indulging in some Magical thinking is, at its core, anti-scientific. "What are we if not biological machines?"

This is a fair question, and an important one. We know that there is only one thing in the universe that is universally intelligent, and that is us. How do we happen to have such powerful creative superpowers? We really don't know.

INTELLIGENCE IS A SUPERPOWER

Try recalling the color of your first bicycle or the name of your first-grade teacher. Maybe you haven't thought about either of these things in years, but your brain probably has no trouble retrieving them, which is even more important when you consider that the "data" isn't stored in your brain like a hard drive. Very impressive. In fact, none of us know how it is stored. We may discover that each of the hundred billion neurons in your brain is as complex as our most advanced supercomputers.

But this is just the beginning of the mysteries of our intelligence. From there everything starts to get trickier. It turns out that we have something called the ability to think, which is separate from the brain itself. The ability to think is all the three pounds of goo in your head can do that it's not supposed to do, like have a sense of humor or fall in love. Your heart doesn't do that, and neither does your liver. But somehow, you did it.

We’re not even sure whether thoughts are entirely products of the brain. More than one or two people are born without up to 95 percent of their brains but still have normal intelligence, often not learning about their condition until later undergoing diagnostic tests. Furthermore, it seems that much of our intelligence is not stored in the brain but is distributed throughout our bodies.

General Artificial Intelligence: The Additional Complexity of Consciousness

Even if we don’t understand the brain or the mind, things actually get more difficult from there: General intelligence will most likely require consciousness. Consciousness is your experience of the world. A thermometer can accurately tell you the temperature, but it cannot feel warmth. This distinction, the difference between what is known and what is experienced, is consciousness, and there is little reason to believe that a computer can experience the world any better than a chair can.

So, our brains we don’t understand, our thoughts we can’t explain, and as for consciousness, we don’t even have a good theory of how mere matter can have experiences. However, despite this, those in artificial intelligence who believe in general artificial intelligence are convinced that we can replicate all human abilities in computers. This sounds to me like an argument that invites fantastical thinking.

I don’t say this to belittle or belittle anyone’s beliefs. They're probably right. I simply view the idea of ​​general artificial intelligence as an unproven hypothesis rather than an obvious scientific truth. The desire to build such a creature and then control it is an ancient dream of mankind. In modern times, it dates back centuries, perhaps starting with Mary Shelley's "Frankenstein" and then manifesting itself in more than a thousand subsequent stories. But it actually goes much earlier than that. We have had such imaginations as far back as we have writing, such as the story of Talos - a robot created by the Greek god of technology Hephaestus to guard the island of Crete.

Somewhere deep within us there is a desire to create such a creature and command its awesome power, but so far there is nothing that should be considered that we are actually able to do arrive.

The above is the detailed content of Is general artificial intelligence possible?. For more information, please follow other related articles on the PHP Chinese website!