#453 — AI and the New Face of Antisemitism

Welcome to the Making Sense Podcast. This is Sam Harris. Just a note to say that if you're hearing this, you're not currently on our subscriber feed and will only be hearing the first part of this conversation. In order to access full episodes of the Making Sense Podcast, you'll need to subscribe at samharris.org. We don't run ads on the podcast, and therefore it's made possible entirely through the support of our subscribers. So if you enjoy what we're doing here, please consider becoming one.

Well, I'm here with Judea Pearl. Judea, thanks for coming into the studio. Great to see you. It's the second time, isn't it? Uh yeah, I came to you last time. Yeah. Yeah, yeah. Yeah, I was in your office. And I I actually didn't look to see when that was, but that's a a few years ago, certainly. Uh that might be.

That was for your book, uh, The Book of Why. The Book of Why. Which uh kinda wraps up for a a popular audience all of your work on causality and uh logic of that, which we'll which we'll touch briefly because I wanna t I have to ask you about AI given that you're one of the fathers of the the field, but that's not really our agenda today, but we'll we'll start near there. But I I wanna talk to you about your um your new book. You have a new book. Coexistence and other fighting words which I um

I'm sorry to say I have not yet read, but that will um give you the uh ability to uh say anything uh to a naive audience on this topic. It r I'm sure it covers much of the ground I wanna cover with you because I'm like you, I think, uh very concerned about cultural issues and um the way that uh we've seen a rise of anti Semitism on both the left and the right and we're now seeing the condition of Israel as a near pariah state, you know, on the world stage.

Briefly, let's start with your background. Where where were you born and and what did your parents do? Well, I was born in a little town called Bnebruck, which is seven and a half miles north of Tel Aviv. It was established in nineteen twenty four. by my grandfather, Heimperl, with twenty five other Hasidic families who came from Poland and decided uh that it's time to g go back to where they belong.

So when did they move to Israel, your parents? In nineteen twenty four. My father okay. My father family came in nineteen twenty four. And when were you born? In nineteen thirty six. Okay. So and what did your parents do? Well my father was the secretary of the um Leibrak municipality. Okay. But that only later. Learning later.

F he came in and became a farmer. You come to Israel in nineteen twenty four, you buy a piece of land and you slap uh water for miles away and you grow redish. That's what he did. Yeah. Yeah, that had to be hard work. Uh it probably still is hard work, but that was Farming was had to be the first order of business. First order, yeah. The idea was to establish a biblical town with religious orientation and make it into uh agricultural success.

Do you know about much about your parents' state of mind when they left? Yes. I mean what was that? Yes, I know. Did did they see w were they w witnessing uh Weimar and its uh No, no, no. My father didn't see. No. No. Uh that was nineteen twenty four. And uh well the legend says At least the family law says that my grandfather came home one day, he was accusted uh

by a Polish peasant and called a dirty Jew, and he came home bloody and he said to his wife and four children, Start packing. We are going to where we belong. Okay. That's a family law, but it has some truth in it, yeah. And what were your principal intellectual influences as a kid and as uh I mean, how did you find your path to computer science as a young person? First I had a very, very good education in I It's um In in in Tel Aviv or I went to a high school in Tel Aviv, yes.

I grew up in Venebrak, but the municipality of Tel Aviv gave a quota to its peripherals, to its suburbs. And Bneibrach was one of its suburbs. So from our town they chose four people. I was chosen among them. It was a privilege at the time to go to a hotel. We were high school. And we had a beautiful education. You know why? Because my high school teachers were professors in Heidelberg and Berlin. Yeah. Yeah. That uh were pushed out by uh uh Hitler.

And when they came to Israel, they couldn't find an academic job, so they taught high school. And we were just privileged and lucky to be part of this unique educational experiment. Yeah, yeah. So and your first language is Hebrew? My first language is Hebrew. So all the studies were in Hebrew.

So um but the people who had just come from Heidelberg, they they were your your professors were speaking Hebrew at that point or what Hebrew. Huh. Interesting. They had to to struggle. Some of them still had uh Yekish accent. Uhhuh. Yeah.

Yeah. Okay, so as I said, we spoke about your Book of Why last time where you um talk about the i the importance of causal reasoning. What's your current view of AI? What has surprised you in recent years? What is we how close to causal reasoning are we uh achieving in the current

crop of LLMs and uh I'm just wondering what you what uh how how you view progress at this point. In causal reasoning or in toward the I guess toward I mean I guess toward a AGI in general. If that is a goal that I don't think We are much closer. We have been deflected by the effect of L Ms where you have a low flying fruits and everybody is excited, which is fine. I mean they're doing

tremendously impressive job. But I don't think they take us toward AGI. So you think you think the the framework, the L L M Deep Learning Framework, is a dead end with respect to AGI? No, it's a step. But it w does it require a fundamental breakthrough that of a sort that we haven't Absolutely, yes. So it's not just more data and more compute. No, no, no, no, no, no. More data and

Can you articulate the reason why, you know, in terms that a a lay person can understand? I mean, if someone asked you, why, why is this? insurmountable by virtue of just throwing more data and compute at it? There are certain limitations.

mathematical limitations that are not crossable by scaling up. I show it clearly mathematically in my book. And what LLMs do right now is they summarize world models authored by people like you and me available on the web and they do some sort of mysterious summary of it. rather than discovering those world models directly from the data. To give you an example, if you have data coming from hospitals about the effect of treatment.

okay you don't fit it directly into the LLMs today you get The input is interpretation of that data authored by doctors, physicians, and people who already have world model body disease and what it does. But couldn't we just put the data itself in as well? Here you have a limitation. You have the limitation defined by the uh ladder of causation. That there is something that you cannot do if you don't have a certain interest. For instance, you cannot get causation from correlation. that is,

Well established, okay? No one would deny it, even satisfaction. by that. And you cannot get interpretation from in intervention. Interpretation means looking backward and doing introspection. But i but intervention is Just uh remind me, but it's it's intervention is what will happen if I do. Right. So it's a kind of an experiment or or a thought experiment. Yeah.

And also doesn't it imply a kind of counterfactual condition where you're saying, well, you know, what would have happened if we didn't intervene? No. No. Here you have a barrier. Uh you have to have additional information to cross from the intervention level to the interpretation level.

And you you put you'd put counterfactuals on the side of interpretation. Yes, correct. Because you go you say, Look what I've seen that David killed Goliath and uh what would have happened had the wind been differently, okay? So who among the other um patriarchs in the field fundamentally disagrees with you? I mean d to people like Jeffrey Hinton or others who have hadress I haven't, well, Jeff Hinton came up with the statement that we are facing a deadlock.

Okay. Oh, I I hadn't heard that. Yeah. But he didn't elaborate on uh the causal component. So I can't

recall if we spoke about this last time, but uh where are you on concerns around alignment and and an intelligence explosion? I mean, I know it sounds like you're not worried that that L L Ms will produce such a thing, but in principle, are you worried do you take IJ Good's uh and others uh early fears seriously that's a good idea.

Once we build AGI on whatever uh on the basis of whatever platform we're in the presence of something that can become recursively self improving and get away from us. Absolutely, yes. I don't see any computational impediments. to that horrifying dream and of course we are already seeing dangers of LLM when they fall into the hands of bad actors. But that's not what we are worried about. We are worried about a truly AGI system that will take over and may be a danger to humanity.

definitely foresee that possible. I can see how it can acquire free will and consciousness and a desire to play around with people. Yeah. That is quite feasible. It doesn't mean that I'm not gonna I'm gonna con stop uh working or uh understanding uh uh AI and its capability simply because I want to understand myself.

Yeah. Yeah. Are you worried that the field is um operating under a kind of a system of incentives, uh essentially an arms race that is uh going to select for reckless behavior? I mean just that we're we I if there is this potential failure mode of building something that destroys us, it seems, at least from the um the statements of the people who are doing this work, you know, the people who are running the major companies

You know, the the probability of such a encountering such existential risk is in their minds at le pretty high. I mean we're we're not hearing people like Sam Altman say Oh yeah, I think the chances are, you know, one in a million that we're going to destroy the future with this technology. They they're putting in the chances at like twenty percent, and yet they're still going as fast as possible. Does doesn't an arms race seem like the worst condition to do this? Carefully.

There are many other people that are worried about it, like uh Stuart Russell's and other Yeah. And the problem is that we don't know how to control it. And whoever says 20% or 5% is just talking. We cannot put a number of that because we don't have theoretical or technical instruments to predict.

Whether or not we can control it. We do not know what's gonna hap what's gonna develop. But what what I find a l alarming about those utterances is that I mean if you just imagine if the you know the physicist who gave us uh the bomb, you know, at the at the at you know uh the Manhattan Project, if when asked About their initial concern that it might ignite the atmosphere and destroy all of life on planet Earth.

If they had been the ones saying, Yeah, maybe it's twenty percent, maybe it's fifteen percent, and yet they were still moving forward with the work, that would have been alarming. But of course that's not what they were saying. They were th they did some calculation and they put the chances To to be, you know, infinitesimal, though not zero.

It just seems bizarre culturally that we have the people doing the work who are not expressing you know, fallaciously or not. I mean, I gr I'll grant you that all of this is made up and it's hard to come up with a with a rational estimate, but For the people doing the work, plowing, you know, trillions of dollars into the build out of AI to be giving numbers like twenty percent seems Culturally is strange. I don't know what I mean by twenty percent uh i look at me. I am d

Fairly sure. All I'm saying is there's no theoretical impediment for creating such a species, dominating species. Right. Okay. It that is true. And at the same time I'm working toward that. indirectly. Not toward that in order to create it, but to understand it well. The capabilities of intelligence in general. Because I want to understand ourselves because I'm curious. Do you have any thoughts on the other?

a system would have to be built so as to be perpetually aligned with our interests. I mean so we if you're taking intelligence seriously. Yes. Right. So we're talking about building an autonomous intelligence system that exceeds our own intelligence and in the limit improves itself, one one would imagine. Do you have any Notions about what a guarantee of an alignment could look like before we hit the play on that? No. I don't think we can imagine an effective alignment.

On a an effective architecture that will reassure us of alignment with with our survival. I think Stuart Russell, I it's been w it's a couple of years since I've spoken with him, but I recall his notion, again this is I'm sure this is a kind of a hand waving notion from the a computer science point of view, but to have as its utility function just to better and better approximate what we want.

to be perpetually uncertain that it's achieved our goals insofar as we can continue to articulate them in an in this open ended conversation that is the evolution of of human culture. Does that seem like a frame that's it's a nice frame, but I don't see any impediment for the new species to overcome or bypass those guard lines and uh and play. What what so people so people have an intuition that if we built it, there's no possibility of it forming its own goals. Like that we didn't anticipate.

What would you say to that? I mean you just very uh breezily articulated certainty that uh or something like certainty that an independently intelligent system can play, that c it can change its mind, it can discover new goals and and cognitive horizons just as we seem to be able to do. Why is there a difference of intuition on this front? I mean Your account seems obvious to me. I don't know why I have different intuition than uh Lacroon.

You want a a system uh that will explore explore its environment. That in in that required for any intelligent system. We wanted to play like a baby in a crib and find out why this toy makes noise and this doesn't, okay? So that's to play around. in order to get control over the environment, to understand the environment. Okay? So once you have the idea of playing, what will prevent from playing with us?

as instrument for his or her understanding, instrument for the environment and become part of its environment.

All right. So this is kind of a reckless pivot from the topic of AI, but it's uh I I think there's a bridge here. I mean I guess we could put this sort of in the frame of the cultural conditions that uh that allow us to reason effectively or fail to reason effectively. And this is on, you know, morally loaded topics like

you know, war and uh you know asymmetric violence, antisemitism, Islamism, again, Israel's status uh among nations. You know, unfortunately you are unusually well placed to have an opinion on these topics.

given your history and what happened to your son back in two thousand two, I don't wanna uh, you know, a awaken painful memories, but I I just feel like I I w I'd we need to I mean I'm happy to talk about this topic in any way you want, but I I just need to acknowledge that your son Danny was w one of the most prominent people killed by Al Qaeda, uh, when the war on terror, so called, uh became of uh you know salient to most people in America, certainly for the first time after

Nine eleven. Nine eleven. So you've spent, you know, now a quarter of a century witnessing uh, you know, as I have, but from a far um kind of deeper space.

the um kind of consistent misunderstanding around jihadism and Islamism that uh has happened especially on the left in our society. I mean that we I it to my eye, we have a um Kind of an anti colonial oppressor oppressed narrative that has captured the moral intuitions of the left such that it's very difficult to talk about some of the ideas within Islam that h that reliably beget the kind of violence we've seen and

Yeah. And you know, the the the groups like the Muslim Brotherhood has managed to play havoc with this moral confusion. They've found uh legions of useful idiots uh even on college campuses like your own. I mean, I don't know if you notice this, but the other day the UAE announced that it would no longer pay for its students to study in the UK at the UK universities for fear that they will be radicalized by the Muslim Brotherhood on UK campuses. So I mean that's how far the rot has spread.

We can take this from any side. We can we can talk about twenty plus years ago where how you came to this or your experience after you're I'm I wanna talk about your experience after October seventh. Just you know, please start wherever you want to start, but the my son's tragedy tragedy pushed me into uh public life and uh into my interest with uh social problem and cultural problem uh the way you are describing.

I uh we started the foundation after his death to with the same belief that it's a matter of communication, dialogue with between the East and West Jews and Muslims and um we I got uh pushed into that very heavily and I started uh together with the um Pakistani scholar, we start the Daniel Pearl dialogue between Muslims and Jews and we went from town to town and we had meetings and discussions.

audience discussions. I even took a trip which I describe in the book, a trip to Doha in 2005 as part of the a conference to bridge East-West relationship and to understand what prevents the Muslim world or the Arab world, the Muslim world, yeah, from modernizing and become enlightened as we are. And that was the first time that I found the barriers which I didn't believe existed. Uh and this with the barrier of Israel.

and progressive. And we came out, my conclusion is that they had a different idea in mind. And we are talking about moderate Muslim. scholars from all over the Muslim world gathering in Doha for this conference, the purpose of which was what how America can do to speed up the process of progress. and democratization of the Muslim world. Their idea was, if you want us to modernize,

We'll give you that favor. We are gonna do the you the favor of modernizing yourself on one condition. We want Israel head on a tray, on a silver plateau. Yeah. This is a condition. We cannot make any progress unless you chop off the head of Israel. Yeah. Well and you were you at this time you were s you were living in Los Angeles, right? You were not living in Israel in two thousand five. No, no, I was in Los Angeles.

Yeah, wait when did you cut when did you come If you'd like to continue listening to this conversation, you'll Subscribe at samharris.org. Once you do, you'll get access to all full-length episodes of the Making Sense Podcast. The Making Sense Podcast is ad-free and relies entirely on listener support, and you can subscribe now at samharris.org.