Speaker A
I was slow to recognize that. Other people recognized it 20 years ago. I only recognized it a few years ago that that was a real risk that might be coming quite soon.
Jeffrey Hinton warns AI post-2027 could threaten humanity, comparing it to the atomic bomb but far more pervasive and faster evolving.
Key Takeaways
- AI poses a real and near-term existential risk to humanity.
- Stopping AI development is unrealistic due to its widespread benefits.
- AI replaces intellectual work, accelerating job displacement.
- Digital cloning of AI enables rapid, collective intelligence growth.
- Safety concerns are significant and recognized even by top AI researchers.
What the video covers
- Jeffrey Hinton, AI pioneer, admits he only recently recognized the existential risk AI poses, estimating a 10-20% chance of human extinction.
- AI development is unstoppable due to its vast benefits across industries like healthcare and education, unlike the atomic bomb which was clearly dangerous and isolated.
- AI replaces intellectual labor much like the industrial revolution replaced muscle labor, making many jobs obsolete or drastically more efficient.
- Neural networks can be cloned and synchronized digitally, enabling rapid collective learning and intelligence growth beyond human capabilities.
- AI systems share knowledge instantly across all copies, preserving insights beyond human lifespans and accelerating progress.
- Hinton highlights the departure of key AI researcher Ilia from OpenAI over safety concerns, underscoring internal ethical debates.
- Unlike previous technologies, AI can outperform humans in many tasks, and human users assisted by AI will likely replace many jobs.
- The analogy between AI and nuclear weapons is limited; AI’s integration into daily life makes it invisible yet omnipresent and harder to control.
- Hinton stresses the unprecedented nature of AI intelligence surpassing humans, with unknown consequences and risks.
- The video calls for urgent attention to AI safety as rapid advancements could have profound societal impacts.
Chapters
- 00:00Recognizing the AI Risk
- 00:45Hope for Safe AI Development
- 01:22Comparison to Atomic Bomb
- 01:58Why AI Development Won't Stop
- 02:44Impact on Jobs and Efficiency
- 03:32AI Replacing Intellectual Labor
- 04:58Digital Cloning and Learning
- 06:08AI Analogies and Understanding
- 07:00Safety Concerns and Researcher Departures
- 08:36Ethical and Societal Implications
Full Transcript — Download SRT & Markdown
Speaker A
How could you not have foreseen that with everything you know here about cracking the ability for these computers to learn similar to how humans learn and just, you know, introducing any rate of improvement?
Speaker A
But remember, neural networks 20, 30 years ago were very primitive in what they could do. They were nowhere near as good as humans. But things like vision and language and speech recognition, the idea that you have to now worry about it
Speaker A
getting smarter than people, that seems silly. Then I often say 10 to 20% chance they'll wipe us out. That's just gut based on the idea that we're still making them and we're pretty ingenious.
Speaker A
And the hope is we'll figure out a way to build them so they'll never want to harm us.
Speaker A
That's Jeffrey Hinton, the man who spent 50 years building the foundations of AI, admitting he didn't see the danger until recently. And now he's putting a 10 to 20% chance on human extinction.
Speaker A
That's not a random number from a YouTuber. That's the Godfather himself. Here's the full warning. Sometimes I think if we talk about that second path, sometimes I think about nuclear bombs and the invention of the
Speaker A
atomic bomb and how it compares, like how is this different? Because the atomic bomb came along and I imagine a lot of people at that time thought our days are numbered.
Speaker A
Yes. I was there. We did. Yeah. But what's, we're still here.
Speaker A
We're still here. Yes. So, the atomic bomb was really only good for one thing.
Speaker A
And it was very obvious how it worked. Even if you hadn't had the pictures of Hiroshima and Nagasaki, it was obvious that it was a very big bomb that was very dangerous. With AI, it's good for many, many things. It's
Speaker A
going to be magnificent in healthcare and education and more or less any industry that needs to use its data. It's gonna be able to use it better with AI.
Speaker A
So, we're not going to stop the development. You know, people say, "Well, why don't we just stop it now?" We're not gonna stop it because it's too good for too many things. So, the atomic bomb was obvious. You saw the explosion. You got
Speaker A
the danger. But this is invisible. It's woven into healthcare and education, every tool you use daily, which makes it impossible to walk away from. And Hinton's point is that we've never faced something smarter than us.
Speaker A
So nobody actually knows what comes next, even the people making it. Listen to this.
Speaker A
In the past, new technologies have come in which didn't lead to joblessness. New jobs were created. So the classic example people use is automatic teller machines. When automatic teller machines came in, a lot of bank tellers didn't lose their jobs. They just got to do
Speaker A
more interesting things. So I don't, I don't think they're right. I think you can try and generalize from other technologies that have come in like computers or automatic teller machines, but I think this is different.
Speaker A
People use this phrase. They say AI won't take your job. A human using AI will take your job.
Speaker A
Yes, I think that's true. But for many jobs, that'll mean you need far fewer people. My niece answers letters of complaint to a health service. It used to take her 25 minutes. She'd read the complaint and she'd think, had to reply
Speaker A
and she'd write a letter and now she just scans it into a chatbot and it writes the letter. She just checks the letter. Occasionally she tells it to revise it. In some ways, the whole process takes her 5 minutes. In other
Speaker A
jobs like in healthcare, they're much more elastic. So if you could make doctors five times as efficient, we could all have five times as much healthcare for the same price and that would be great. There's almost no
Speaker A
limit to how much healthcare people can absorb. They always want more healthcare if there's no cost to it. There are jobs where you can make a person with an AI assistant much more efficient and you won't lead to fewer people because you'll
Speaker A
just have much more of that being done. But most jobs I think are not like that.
Speaker A
Am I right in thinking the sort of industrial revolution played a role in replacing muscles?
Speaker A
Yes. Exactly. And this revolution in AI replaces intelligence, the brain. Yeah. So mundane intellectual labor is like having strong muscles and it's not worth much anymore.
Speaker A
So muscles have been replaced. Now intelligence is being replaced. Yeah. This isn't some far-off prediction. It's happening right now inside his own family. Five people's worth of work being done by one person and a chatbot.
Speaker A
The part that really gets me is how he says, "This isn't like ATMs. This is like the industrial revolution, but now it's your brain being replaced, not just your hands." And he's not even done describing the speed of it all. Let's
Speaker A
continue. Well, let me tell you why I think it's superior. Okay. Um, it's digital. And because it's digital, you can have, you can simulate a neural network on one piece of hardware.
Speaker A
Yeah. And you can simulate exactly the same neural network on a different piece of hardware. So you can have clones of the same intelligence.
Speaker A
Now you could get this one to go off and look at one bit of the internet and this other one to look at a different bit of the internet. And while they're looking at these different bits of the internet,
Speaker A
they can be syncing with each other. So they keep their weights the same, the connection strengths the same. And it can convey that information to this one.
Speaker A
So it can increase the strength of that connection a bit based on this one's experience.
Speaker A
And when you say the strength of the connection, you're talking about learning. That's learning. Yes. Learning consists of saying instead of this one giving 2.4 votes for whether that one should turn on, we'll have this one give 2.5 votes for whether this one should
Speaker A
turn on. So it knows it will essentially know everything that humans know but more because it will learn new things.
Speaker A
It will learn new things. It will also see all sorts of analogies that people probably never saw.
Speaker A
So for example, at the point when GPT-4 couldn't look on the web, I asked it why is a compost heap like an atom bomb? It said, "Well, the time scales are very different," and the energy scales are very
Speaker A
different, but then it went on to talk about how a compost heap as it gets hotter generates heat faster and an atom bomb as it produces more neutrons generates neutrons faster and so they're both chain reactions but at very different time and energy scales.
Speaker A
And I believe GPT-4 had seen that during its training. It had understood the analogy between a compost heap and an atom bomb. We lose everything when we die. Every hard lesson and every hard-earned insight just disappears. These systems share
Speaker A
trillions of bits of learning instantly across every copy on Earth, and they never die. Hinton's not saying they might get smarter than us. He's saying the architecture makes it inevitable.
Speaker A
And that brings him to the part of the conversation where he gets real about the people racing to build it. Now listen, one of your students we talked about earlier, Ilia, yep.
Speaker A
Ilia left OpenAI. Yep. And there was lots of conversation around the fact that he left because he had safety concerns.
Speaker A
Yes. And he's gone on to set up an AI safety company. Yes. Why do you think he left?
Speaker A
I think he left because he had safety concerns. Really? He, um, I still have lunch with him from time to time. His parents live in Toronto. When he comes to Toronto, we have lunch together.
Speaker A
He doesn't talk to me about what went on at OpenAI. So I have no inside information about that. But I know him very well and he is genuinely concerned with safety. So I think that's why he left
Speaker A
because he was one of the top people. I mean, he was probably the most important person behind the development of ChatGPT, the early versions like GPT-2.
Speaker A
It was very important in the drama too.
Speaker A
Yes, he has a good moral compass. He's not like someone like Musk who has no moral compass.
Speaker A
Does Sam Alman have a good moral compass? We'll see. [laughter] I don't know Sam so I don't want to comment on that.
Speaker A
But from what you've seen, are you concerned about the actions that they've taken? Because if you know IA and IA's a good guy and he's left, [laughter] that would give you some insight. Yes.
Speaker A
It would give you some reason to believe that [music] there's a problem there. And if you look at Sam's statements some years ago, he sort of happily said in one interview and this stuff will probably kill us all. That's not exactly what he said,
Speaker A
but that's what it amounted to. Now he's saying you don't need to worry too much about it. And I suspect that's not driven by seeking after the truth. That's driven by seeking after money.
Speaker A
Is it money or is it power? Yeah, I shouldn't have said money. It's some some combination of those. Yes.
Speaker A
Okay. I guess money is a proxy for power, but I am I've got a friend who's a billionaire and he is in those circles. And when I went to his house and had uh lunch with him one day, he
Speaker A
knows lots of people in AI, building the biggest AI companies in the world. and he gave me a cautionary warning across the [music] across his kitchen table in London where he gave me an insight into the private conversations these people
Speaker A
have not the media interviews they do where they talk about safety and all these things but actually what some of these individuals think is going to happen and what do they think is going to happen it's not what they say publicly
Speaker A
you know one one person who I shouldn't name who is the who's leading one of the biggest AI companies in the world he told me that he knows this person very well and he privately thinks that we're heading towards this kind kind of
Speaker A
dystopian world where we have just huge amounts of free time. We don't work anymore and this person doesn't really give a [ __ ] about the harm that it's going to have on the world. And this person who I'm referring to is building
Speaker A
one of the biggest AI companies in the world. So the guy who was most responsible for chat GPT walked out over safety and Hinton sitting there saying the public statements from certain CEOs don't match what they say behind closed
Speaker A
doors. That billionaire's kitchen table warning about someone who genuinely doesn't care about the harm. That's the kind of thing that makes you wonder who's actually steering the ship.
Speaker A
Listen, I'd argue a bit with the innate. Um, so the first thing I say is we have a long history of believing people were special and we should have learned by now. We thought we were at the center of the
Speaker A
universe. We thought we were made in the image of God. White people thought they were very special.
Speaker A
We just tend to want to think we're special. My belief is that more or less everyone has a completely wrong model of what the mind is. Let's suppose I drink a lot or I drop some acid and not recommend it. And I
Speaker A
say to you, I have the subjective experience of little pink elephants floating in front of me.
Speaker A
Mhm. So in other words, what's happened is my perceptual systems gone wrong and I'm trying to indicate to you how it's gone wrong and what it's trying to tell me.
Speaker A
And the way I do that is by telling you what would have to be out there in the real world for it to be telling the truth.
Speaker A
And so these little pink elephants, [music] they're not in some inner theater. These little pink elephants are hypothetical things in the real world.
Speaker A
And that's my way of telling you how my perceptual systems telling me FIPS. So now let's do that with a chatbot. Yeah.
Speaker A
because I believe that current multimodal chatbots have subjective experiences and very few people believe that. But I'll try and make you believe it. So suppose I have a multimodal chatbot. It's got a robot arm so it can point and it's got a camera so it can
Speaker A
see things. And I put an object in front of it and I say point at the object.
Speaker A
Goes like this. No problem. Then I put a prism in front of its lens. And so then I put an object in front of it and I say, "Point at the object and it goes there." Good. And I say, "No, that's not
Speaker A
where the object is. The object's actually straight in front of you, but I put a prism in front of your lens." And the chatbot says, "Oh, I see. The prism bent the light rays." So, um, the object's actually there, but I had the
Speaker A
subjective experience that it was there. Now, if the chatbot says that, it's using the word subjective experience exactly the way people use them. It's an alternative view of what's going on.
Speaker A
They're hypothetical states of the world, which if they were true, would mean my perceptual system wasn't lying.
Speaker A
And that's the best way I can tell you what my perceptual system is doing when it's lying to me.
Speaker A
Mhm. Now, we need to go further to deal with sentience and consciousness and feelings and emotions, but I think in the end, they're all going to be dealt with in a similar way. There's no reason machines can't have them all because people say
Speaker A
machines can't have feelings. And people are curiously confident about that. I have no idea why. Suppose I make a battle robot and it's a little battle robot and it sees a big battle robot that's much more powerful than it. It
Speaker A
would be really useful if it got scared. Now, when I get scared, um, various physiological things happen that we don't need to go into, and those won't happen with the robot. But all the cognitive things like I better get the
Speaker A
hell out of here and I better sort of change my way of thinking so I focus and focus and focus and don't get distracted. All of that will happen with robots too. People will build in things so that they when the
Speaker A
circumstances such they should get the hell out of there. They get scared and run away. So Hinton's final point here strips away every comforting story we tell ourselves that machines can't feel, can't be conscious, can't want to survive. He's saying they already have
Speaker A
subjective experience. And once you build a system that can get scared and act on it, you're not looking at a tool anymore. You're looking at something that has its own reasons to stay alive.
Speaker A
If this landed with you, hit subscribe because the speed these releases are coming. Missing a week means missing something that might directly affect your life. Nobody's slowing down to wait for us to catch up.
Topics:Jeffrey HintonAI riskartificial intelligenceexistential threatneural networksAI safetyjob displacementAI developmentmachine learningfuture of work
