Yuval Noah Harari - “Nexus” & Threat of AI in the Information Age | The Daily Show

You are a popular writer. Your books have sold over 45 million copies. Whoa.

[APPLAUSE] The Atlantic referred to some of your writing style as since the dawn of time style books. You go way back, and you bring us into the future. These are big, important tomes.

Simultaneously, I heard you meditate for two hours every single day. Yes. How the [BLEEP] do you make all this happen?

[LAUGHTER] I don't have kids. You don't have kids? [LAUGHTER] What have I done?

[APPLAUSE] Why don't you write a pamphlet that says just that? You want to get shit done, don't have kids. Some people manage to do both.

But, you know. But you have time to dive into this. Yes.

I'm curious. This book is about information. Yeah.

And you reject the notion that more information is a good thing, that it leads to truth and wisdom. Is this you being jaded by the Trump administration and the time we're in, or does this thought process go back? It's basically like thinking that more food is always good for you.

You know, there is a limit to how much food the body needs. And in a similar way, there is a limit to how much food for thought, food for the mind, the mind needs, which is information. And the same way that most-- there is so much junk food out there, there is also so much junk information out there.

And we basically need to go on an information diet. Yes, this-- [CHEERING] But I need my sweet, sweet Twitter snacks, Yuval, I need it. I need it.

It's exactly that. The same way that over the last few generations, they learned-- the industries learned how to produce artificial food, which is pumped full of fat and sugar and salt and is addictive and not good for us, they've also learned how to manufacture this artificial information, which is pumped full of greed and hate and fear and is addictive to our mind and isn't good for it. Now, I totally agree, and I feel stuffed on all of it.

But I also have this feeling that when you step outside of this information mainstream, that's just-- this pipeline of BS that is out there-- that you suddenly step out of the conversation. It feels like we don't have the luxury of going on a diet if we want to be part of the conversation around us. Because the conversation is increasingly managed not by human beings but by algorithms.

And algorithms function in a completely different way than us. They are not organic. For instance, human beings, as organic animals, we run by cycles.

Sometimes we need to be very active. Sometimes we need to rest. But algorithms never rest.

They are tireless, and they expect us to be the same. So we now live in this new cycle which never rests, and the same thing happens in politics, in finance. You know, previously if you think about Wall Street, so even Wall Street takes risks.

The market is open from Mondays to Fridays, 9:30 in the morning to 4:00 in the afternoon. That's it. If a new war erupts in the Middle East, an unlikely event, but let's say a new war erupts in the Middle East on Friday at five minutes past four, Wall Street will react only on Monday morning.

It is on weekend vacation. And this is actually a good thing, because if you force organic entities to be on all the time, they eventually collapse and die, which is really what is happening to us as individuals and as societies. I think maybe the most misunderstood and abused word in the English language today is the word excited.

People think that excited means happy. Like, I meet you, and I say, I'm so excited to meet you. Yeah, that's what happened with us backstage.

[LAUGHTER] - Yeah, yeah, yeah. But excited-- Keep going, Yuval. Excited doesn't mean happy.

JORDAN KLEPPER: It doesn't mean happy Excited means that all your nervous system and your brain is like fuzzing. It's on. And it's good to be excited sometime.

But if you keep an organic being, an animal, excited all the time, it eventually collapses and dies. So you're saying beforehand I should have said, Yuval-- I'm so relaxed to meet you. I'm relaxed to meet you.

I apologize. I'm dead inside, but that's not your problem. [LAUGHTER] [APPLAUSE] Well, think, for instance, about the election cycles and US politics.

Wouldn't it be better if it was a bit more boring? I would love it if it were boring. I would love it if it were boring.

And we see what happens in Europe where it's shorter. It's boring, but everything, everything is pulling us to maximalize, right? - Yes.

The idea that-- the fact that if we had a news cycle that could end on Friday and then we pick it back up on Monday would be fantastic. But it doesn't seem like the algorithms, it doesn't seem like the financial benefits are pushing us in that direction at all. Where do you see a path like that going through?

If you keep kind of increasing the pace all the time, we can't handle it. So the algorithms can, so they take over. But it's not good news for humanity.

We need to slow down, basically. And, you know, we are facing now these non-organic entities which work and think in a completely different way from us. And the question is, who is going to adapt to whom?

Now, you're pointing at AI. Yeah. Now, is AI, do you see it as an existential threat?

Like, I've seen some of these Shrimp Jesuses, and I don't like it. These weird images that pop up online, but I don't necessarily connect that with the end of conversation. I think the most important thing to understand about AI is that AI is not a tool.

It is an agent. It's the first technology in history that can make decisions and invent new ideas by itself. Even something as powerful as the atom bomb could not decide anything by itself.

All the decisions were made by humans. Now we've created something which potentially can take power away from us. At present, it starts with very small things.

Like, for instance, there was an experiment when OpenAI developed GPT-4 like two years ago. They wanted to test what can this thing do? So they gave it a task to solve captcha puzzles, the captcha puzzles like when you go online and you want to access your bank or whatever, and they have this riddle that you have to solve, an image that you have to say what are the twisted words and letters to make sure you are not a robot.

That's tough. I've taken the test. It's tough.

- Yeah. Is that a street light? Is that a bicycle wheel?

I don't know. Let me do it again. Refresh.

And it's really difficult for GPT-4. GPT-4 could not solve the captcha. But what GPT-4 did, it accessed Task Rabbit, which is an online site where you can hire humans to do different things for you.

And it asked a human to solve the captcha for it. Now, the human got suspicious. The human asked, why do you need somebody to solve captcha for you?

Are you a robot? It asked directly are you a robot? And GPT-4 answered no.

I'm not a robot. I have a vision impairment, which is why I can't solve the captcha. So I need your help.

So the fully-- so the truly evolved human is not somebody who's smarter. It's just somebody who gets somebody else to do the work for them. [LAUGHTER] Smart.

- Yeah. - Scary. Very scary.

It's curious, you talk a little bit about-- there's a portion here you talk about the artist's role in the community of whether it's comedy or writers or filmmakers. People talk about is AI coming for our jobs? Part of what you articulate right here is that it's an artist's job to sort of paint these fears, let us understand the dynamics of human interaction.

You break things down into what these social networks need. And I'm paraphrasing, but like both stories of mythology that lift us up and also articulations of the bureaucracy. Bureaucracy is very important.

It's very important, which I think is-- explain that to me a little bit. But I also feel it's very difficult also for artists to articulate bureaucracy. That's the problem.

We are very good at articulating mythology. We love mythological stories, and mythology is very important. But ultimately, our world, the modern world, is built on bureaucracies.

And this is also where AI fits into the picture because we are now going to see millions and millions of AI bureaucrats. The kind of existential threat we are facing is not this Hollywood scenario of a single computer trying to take over the world. It's millions of AI bureaucrats in the banks, in the governments, in the armies, in the schools, making decisions about us.

Like, you apply to a bank to get a loan, and it's an AI bureaucrat deciding whether to give you a loan or not. You apply for the job, for a place in college, it's the same thing. Now, the thing with bureaucracy, it's boring.

It's boring. JORDAN KLEPPER: Yeah. It's very difficult for artists to write good stories about bureaucracies.

But if the function of art is help us understand reality, this is much more important than telling mythological stories. And, you know, when was the last time you saw a really good TV show about bureaucracy, let's say, about the budget? Like, how does the budget-- I'm binging a 12-part series on the budget right now that is-- [CHEF'S KISS] Well, I think about it.

I think of like movies like The Big Short. But for every The Big Short, you have 1,000 Marvel movies. Exactly.

That live in the world of mythology. Yeah, so superheroes, this is mythology. This is no how-- not how the budget works.

You don't have a super accountant fighting against I don't know what. [LAUGHTER] Yeah, let's-- we can workshop this. Yeah, yeah.

[LAUGHTER] But, you know, what shapes your life is these accountants with the budgets far more than the superheroes. And it's really a challenge to do a good TV series about the budget. And even if we try, it will end up, again, like a love story between somebody in account and somebody in another department.

And the budget will be pushed to the side. JORDAN KLEPPER: Yeah. But we need to really understand how these things work.

I think what I love about a lot of your work is it does explore the stories that we tell and how important that is to just humankind and the way that we create societies and build off one another. And the danger of not telling those stories or not bringing people in together. I think when I fear about our future and our democracies and our ability to hold these conversations, I think about things like AI.

But I also very much think about these mediums that our conversations are taking place in, whether it's on Twitter or cable news or TikTok. Like, none of these mediums are pointing towards or value any type of conversation that is helpful in a way that is beneficial. And so I'm afraid of the AI in the way that we're tracking.

But I don't see a platform or a place where the conversations that need to happen can happen. I think the number one question to ask to the Zuckerbergs and the Elon Musks of the world and so forth-- - Do you have their number? I can text them right now.

[LAUGHTER] So if you have their number, this is the question. How is it that we have the most sophisticated information technology in history, and we can no longer hold the conversation? We can no longer talk with each other.

That's the big question. And you see it in democracies all over the world. You see it here in the US.

You see it in my home country, in Israel. You see it in Brazil, in the Philippines, in France. The conversation is breaking down.

So what is happening? This extremely sophisticated information technology, it is not helping the conversation. It is destroying it.

Yeah, 100%. [CHEERING] I talk to older people on the road who go to-- like people at rallies, at MAGA rallies, who will go to Facebook as a place to converse with friends. And frankly, if you're in your 60s, that's the place to talk to, friends, to connect.

But in order to be a person on Facebook, it's not enough for you just to converse with the friends you have there. You have to publish news sources to get people to pay attention to you. And I feel like the Zuckerbergs and the Facebooks and these media sites that we have right now, we promised this idea of conversation or that you can connect with friends.

But we ask people to be publishers of ideas and stories and promoters of things that are outside the realm of what makes a healthy conversation and more so, muddy up the ability to have that honest conversation. You know, traditionally-- and we've been in this place every time a new information technology was invented-- we faced the same difficulties. For instance, when the printing revolution swept Europe in the early modern period, it did not lead directly, as many people think, to the scientific revolution.

The best sellers of the early print era were not Copernicus and Galileo Galilei and Newton. Hardly anybody read those books. The big bestsellers were religious tracts and were witch hunting manuals.

The big witch hunts, they were not a medieval phenomenon. Medieval people didn't care very much about witches. The really big witch hunts, they began after the print revolution.

One of the biggest bestsellers was a book called The Hammer of the Witches, which was a do-it-yourself manual to identifying and killing witches. The hammer of-- The Hammer of the Witches. And it was full of these stories about cannibalistic orgies and gatherings of-- and this was far more interesting than Copernicus with all his mathematics.

I've got to tell you, I'm writing it down-- Hammer of the Witches. That sounds good. Hammer of the Witches.

Also, my favorite Led Zeppelin album. [LAUGHTER] For instance. If you want to really understand like QAnon today, it's basically the same story.

There is a conspiracy of Satan worshipping witches that is trying to destroy the world. And good Christians need the ability to identify and destroy these witches. It's not a new thing on Facebook or Twitter.

It goes back to the print revolution in the 15th and 16th century. Are there any examples, looking back at history, though, where we faced these technological watershed moments where we are given new technology and that humanity has decided to revert and say no to it and move beyond it? It feels like a foregone conclusion that we are heading into this AI revolution, and we're not writing the rules.

A couple rich folks in Silicon Valley are. The-- you can't go back in history. That's impossible.

But the answer is always the same. You need institutions, and institutions, they are not heroic. They are not superheroes.

They don't-- they are not kind of the main theme of Marvel movies, but they are always people reached the conclusion they are the answer. Because if you want to-- you know, in the ocean of fake and junk information, if you want to know the truth, you need institutions like newspapers, like academic associations, like courts that develop mechanisms to sift through the evidence and decide what is reliable information and what is unreliable. Again, it's not heroic, but this is always the answer.

And we need to do it again with the current information revolution. So as long as newspapers stay strong as a business model, perhaps VHS machines can get in there, too, and fight the good fight. You know, you actually-- you signed a book for me backstage.

And one of the comments you made within it was to not lose hope. Help me. Help me do that.

Where do you see those little glimmers of hope when you look at this, this uncertain and perhaps scary future that we're walking into? You know, I think that AI is nowhere near its full potential. But humans also, we are nowhere near our full potential.

If we-- if for every dollar and every minute that we invest in developing artificial intelligence, we also invest in exploring and developing our own minds, we will be OK. But if we put all our bets on the technology, on the AI, and neglect to develop ourselves, this is very bad news for humanity. All right.

So I'm going to get that gym membership, and I'm going to cut out the sweets. Nexus is available now. Yuval Noah Harari.