Whether you think artificial intelligence will save the world or end it, you have Jeffrey Hinton to thank. Hinton has been called The Godfather of AI, a British computer scientist whose controversial ideas help make advanced artificial intelligence possible and so change the world. Hinton believes that AI will do enormous good, but tonight he has a warning. He says that AI systems may be more intelligent than we know, and there's a chance the machines could take over. Which made us ask the question: Does humanity know what it's doing?
"No, um, I think we're moving into a period when, for the first time ever, we may have things more intelligent than us."
You believe they can understand?
"Yes."
You believe they are intelligent?
"Yes."
You believe these systems have experiences of their own and can make decisions based on those experiences, in the same sense as people do?
"Yes."
Are they conscious?
"I think they probably don't have much self-awareness at present, so in that sense I don't think they're conscious."
Will they have self-awareness, consciousness?
"I, oh yes, I think they will in time, and so human beings will be the second most intelligent beings on the planet."
Yeah. Jeffrey Hinton told us the artificial intelligence he set in motion was an accident, born of a failure. In the 1970s at the University of Edinburgh, he dreamed of simulating a neural network on a computer, simply as a tool for what he was really studying: the human brain. But back then, almost no one thought software could mimic the brain. His PhD advisor told him to drop it before it ruined his career. Hinton says he failed to figure out the human mind, but the long pursuit led to an artificial version.
"It took much, much longer than I expected, it took like 50 years before it worked well, but in the end, it did work well."
At what point did you realize that you were right about neural networks, and most everyone else was wrong?
"I always thought I was right."
In 2019, Hinton and collaborators Yan Laon on the left and Yosua Beno won the Turing Award, the Nobel Prize of computing. To understand how their work on artificial neural networks helped machines learn to learn, let us take you to a game.
"Look at that, oh my goodness."
This is Google's AI lab in London, which we first showed you this past April. Jeffrey Hinton wasn't involved in this soccer project, but these robots are a great example of machine learning. The thing to understand is that the robots were not programmed to play soccer; they were told to score. They had to learn how on their own.
"Oh, go in."
In general, here's how AI does it. Hinton and his collaborators created software in layers, with each layer handling part of the problem. That's the so-called neural network. But this is the key: when, for example, the robot scores, a message is sent back down through all of the layers that says, "That pathway was right." Likewise, when an answer is wrong, that message goes down through the network. So, correct connections get stronger, wrong connections get weaker, and by trial and error, the machine teaches itself.
"You think these AI systems are better at learning than the human mind?"
"I think they may be, yes. And at present, they're quite a lot smaller. So even the biggest chatbots only have about a trillion connections in them. The human brain has about 100 trillion. And yet, in the trillion connections in a chatbot, it knows far more than you do in your 100 trillion connections, which suggests it's got a much better way of getting knowledge into those connections."
A much better way of getting knowledge that isn't fully understood.
"We have a very good idea of, sort of, roughly what it's doing, but as soon as it gets really complicated, we don't actually know what's going on anymore than we know what's going on in your brain."
What do you mean we don't know exactly how it works? It was designed by people.
"No, it wasn't. What we did was, we designed the learning algorithm. That's a bit like designing the principle of evolution. But when this learning algorithm then interacts with data, it produces complicated neural networks that are good at doing things, but we don't really understand exactly how they do those things."
What are the implications of these systems autonomously writing their own computer code and executing their own computer code?
"That's a serious worry, right? So, one of the ways in which these systems might escape control is by writing their own computer code to modify themselves. And that's something we need to seriously worry about."
What do you say to someone who might argue, if the systems become benevolent, just turn them off?
"They will be able to manipulate people, right? And these will be very good at convincing people because they'll have learned from all the novels that
were ever written, all the books by Makavelli, all the political connives. They'll know all that stuff, they'll know how to do it."
Know of the human kind runs in Jeffrey Hinton's family. His ancestors include mathematician George Buou, who invented the basis of computing, and George Everest, who surveyed India and got that mountain named after him. But as a boy, Hinton himself could never climb the peak of expectations raised by a domineering father.
"Every morning when I went to school, he'd actually say to me as I walked down the driveway, 'Get in their pitching, and maybe when you're twice as old as me, you'll be half as good.'"
Dad was an authority on Beatles.
"He knew a lot more about Beatles than he knew about people."
Did you feel that as a child?
"A bit, yes."
When he died, we went to his study at the University, and the walls were lined with boxes of papers on different kinds of beetle, and just near the door, there was a slightly smaller box that simply said "not insects." And that's where he had all the things about the family.
Today, at 75, Hinton recently retired after what he calls 10 happy years at Google. Now, he's professor emeritus at the University of Toronto, and he happened to mention he has more academic citations than his father. Some of his research led to chatbots like Google's Bard, which we met last spring.
"Confounding, absolutely confounding."
We asked Bard to write a story from six words: "For sale, baby shoes, never worn."
"Holy cow."
"The shoes were a gift from my wife, but we never had a baby."
Bard created a deeply human tale of a man whose wife could not conceive, and a stranger who accepted the shoes to heal the pain after her miscarriage.
"I am rarely speechless. I don't know what to make of this."
Chatbots are said to be language models that just predict the next most likely word based on probability.
"You'll hear people saying things like, 'They're just doing autocomplete, they're just trying to pred the next word,' and 'they're just using statistics.' Well, it's true they're just trying to predict the next word, but if you think about it, to predict the next word, you have to understand the sentences. So the idea they're just predicting the next word so they're not intelligent is crazy. You have to be really intelligent to predict the next word really accurately."
To prove it, Hinton showed us a test he devised for Chat GPT-4, the chatbot from a company called OpenAI. It was sort of reassuring to see a Turing Award winner mistype and blame the computer.
"Oh, damn, this thing, we're going to go back and start again."
That's okay. Hinton's test was a riddle about house painting. An answer would demand reasoning and planning. This is what he typed into Chat GPT-4:
"The rooms in my house are painted white, or blue, or yellow, and yellow paint fades to white within a year. In two years' time, I'd like all the rooms to be white. What should I do?"
The answer began in one second. GPT-4 advised: The rooms painted in blue need to be repainted. The rooms painted in yellow don't need to be repainted because they would fade to white before the deadline.
"And oh, I didn't even think of that."
It warned: If you paint the yellow rooms white, there's a risk the color might be off when the yellow fades. Besides, it advised, you'd be wasting resources painting rooms that were going to fade to white anyway.
"You believe that Chat GPT-4 understands?"
"I believe it definitely understands, yes. And in five years' time, I think in 5 years' time, it may well be able to reason better than us."
Reasoning that he says is leading to AI's risks and great benefits.
"So, an obvious area where there's huge benefits is healthcare. AI is already comparable with radiologists at understanding what's going on in medical images. It's going to be very good at designing drugs; it already is designing drugs. So that's an area where it's almost entirely going to do good. I like that area."
The risks are what?
"Well, the risks are having a whole class of people who are unemployed and not valued much because what they used to do is now done by machines."
Other immediate risks he worries about include fake news, unintended bias in employment and policing, and autonomous battlefield robots.
"What is a path forward that ensures safety?"
I don't know. I can't see a path that guarantees safety. We're entering a period of great uncertainty where we're dealing with things we've never dealt with before. Normally, the first time you deal with something totally novel, you get it wrong, and we can't afford to get it wrong with these things. Can't afford to get it wrong. Why? Well, because they might take over.
Take over from humanity? Yes, that's a possibility. Why would they? Saying it will happen if we could stop them ever wanting to, that would be great. But it's not clear we can stop them ever wanting to. Jeffrey Hinton told us he has no regrets because of AI's potential for good. But he says now is the moment to run experiments, to understand AI, for governments to impose regulations, and for a world treaty to ban the use of military robots.
He reminded us of Robert Oppenheimer who, after inventing the atomic bomb, campaigned against the hydrogen bomb. A man who changed the world and found the world beyond his control. It may be we look back and see this as a kind of turning point, when humanity had to make the decision about whether to develop these things further and what to do to protect themselves if they did.
Um, I don't know. I think my main message is there's enormous uncertainty about what's going to happen next. These things do understand, and because they understand, we need to think hard about what's going to happen next. And we just don't know.