Peter Norvig: Artificial Intelligence: A Modern Approach | Lex Fridman Podcast #42

1. 0:00 – 3:46

   How AI: A Modern Approach evolved: hardware progress and a shift toward values

   1. LFLex Fridman0:00

      The following is a conversation with Peter Norvig. He's the director of research at Google and the co-author with Stuart Russell of the book Artificial Intelligence: A Modern Approach that educated and inspired a whole generation of researchers, including myself, to get into the field of artificial intelligence. This is the Artificial Intelligence podcast. If you enjoy it, subscribe on YouTube, give it five stars on iTunes, support on Patreon, or simply connect with me on Twitter @lexfridman, spelled F-R-I-D-M-A-N. And now, here's my conversation with Peter Norvig. Most researchers in the AI community, including myself, own all three editions, red, green, and blue, of the, uh, Artificial Intelligence: A Modern Approach. It's a field-defining textbook, as many people are aware, that you wrote with Stuart Russell. How has the book changed and how have you changed-

   2. PNPeter Norvig0:55

      (laughs) Yeah.

   3. LFLex Fridman0:55

      ... uh, in relation to it from the first edition to the second to the third and now fourth edition as you work on it?

   4. PNPeter Norvig1:00

      Yeah. So it's, so it's been a lot of years, a lot of changes. One of the things changing from the first to m- m- maybe the second or third was just the rise of, uh, computing power, right? So I think in the, in the first edition we said, uh, "Here's predicate logic, but, uh, that only goes so far 'cause pretty soon you have millions of, uh, short little predicate expressions and they couldn't possibly fit in memory. Uh, so we're gonna use first-order logic that's, uh, more concise." And then we quickly r- realized, "Oh, predicate logic is pretty nice because there are really fast SAT solvers and other things, and look, there's only millions of expressions and that fits easily into memory, or maybe even billions fit into memory now." So, that was a change of, uh, the type of technology we needed just because the hardware expanded.

   5. LFLex Fridman1:56

      Even to the second edition?

   6. PNPeter Norvig1:58

      Yeah. Yeah.

   7. LFLex Fridman1:58

      So resource constraints were loosened significantly for the second edition?

   8. PNPeter Norvig2:01

      Yeah. Yeah. And then-

   9. LFLex Fridman2:02

      And that was early 2000s, second edition?

   10. PNPeter Norvig2:05

       Right. So '95-

   11. LFLex Fridman2:06

       Yeah.

   12. PNPeter Norvig2:06

       ... was the first and then, uh, 2000, 2001 or so. And then, uh, moving on from there, I think we're s- we're starting to see that again with the, uh, GPUs and then, uh, more specific type of, uh, machinery like the TPUs and w- we're seeing custom ASICs and so on, uh, for deep learning. So, we're seeing another advance in terms of the hardware. Then I think another thing that we especially noticed this time around is in all three of the first editions, we kind of said, "Well, we're gonna find AI as maximizing expected utility."

   13. LFLex Fridman2:42

       Mm-hmm.

   14. PNPeter Norvig2:43

       "And you tell me your utility function and now we've got 27 chapters worth of cool techniques for how to optimize that." I think in this edition, we're saying more, "You know what? Maybe that optimization part is the easy part-

   15. LFLex Fridman2:57

       Mm-hmm.

   16. PNPeter Norvig2:58

       ... and the hard part is deciding what is my utility function? What do I want? And if I'm a collection of agents or a society, uh, what do we want as a whole?"

   17. LFLex Fridman3:08

       So, you touched that topic in this edition. You get-

   18. PNPeter Norvig3:10

       Yeah.

   19. LFLex Fridman3:10

       ... a little bit more into utility.

   20. PNPeter Norvig3:12

       Yeah. Yeah.

   21. LFLex Fridman3:12

       That's really interesting. Uh, on a, uh, a technical level or almost pushing the philosophical?

   22. PNPeter Norvig3:17

       I guess it, it is philosophical, right? So we, we've always had a philosophy chapter, which, which I was, uh, glad to s- that we were supporting. And now, it's less kind of the, uh, you know, Chinese room-type argument-

   23. LFLex Fridman3:33

       Mm-hmm.

   24. PNPeter Norvig3:33

       ... and more of these, uh, ethical and societal-type issues. Uh, so we get into, uh, the issues of, uh, fairness and bias and, uh, and just the issue of, uh, aggregating utilities.

2. 3:46 – 4:49

   Encoding human values: inverse reinforcement learning and its limits

   1. LFLex Fridman3:46

      So, how do you encode human values into a utility function?

   2. PNPeter Norvig3:49

      (laughs)

   3. LFLex Fridman3:50

      Is, is there something that you can do purely through data in a learned way or is there some systematic... Obviously, there's no good answers yet. There's just, uh-

   4. PNPeter Norvig3:59

      Yeah.

   5. LFLex Fridman4:00

      ... beginnings to this, uh, to, to even opening the door to these questions.

   6. PNPeter Norvig4:02

      Right. So, there is no one answer. Yes, there are techniques, uh, to try to learn that. So we talk about inverse reinforcement learning.

   7. LFLex Fridman4:10

      Mm-hmm.

   8. PNPeter Norvig4:10

      Right? So, reinforcement learning, uh, you take some actions, you get some rewards, and you figure out what actions you should take. In inverse reinforcement learning, you observe somebody taking actions and you figure out, uh, "Well, th- this must be what they were trying to do. If they did this action, it must be because they want it." Of course, there is restrictions to that, right? So, lots of people take actions that are self-destructive-

   9. LFLex Fridman4:36

      Mm-hmm.

   10. PNPeter Norvig4:36

       ... uh, where they're, they're suboptimal in certain ways. So, you don't want to learn that.

   11. LFLex Fridman4:40

       Right.

   12. PNPeter Norvig4:40

       You want to, uh, somehow learn the, uh, the, the perfect actions, uh, rather than the ones they actually take. So, uh, so that's a challenge-

   13. LFLex Fridman4:48

       Mm-hmm.

3. 4:49 – 7:07

   Fairness trade-offs in real systems: recidivism prediction and impossibility results

   1. PNPeter Norvig4:49

      ... uh, for that field. Then another big part of it i- is just kind of, uh, theoretical of saying, uh, what can we accomplish? And so you look at, like, this, this work on the, uh, programs to, uh, predict recidivism and decide, uh, you know, who should get parole or who should get bail or whatever. Uh, and how are you gonna evaluate that? And one of the big issues is fairness across protected classes, protected classes being things like, uh, sex and race and so on. And, uh, so two things you want is you want to say, "Well, if I get a score of, say, uh, six out of 10, then I want that to mean the same whether... no matter what race I'm on."

   2. LFLex Fridman5:37

      Yes.

   3. PNPeter Norvig5:37

      Right? So I want to have a 60% chance of, uh, of reoccurring, uh, regardless. Uh, and the makers of the, uh, o- one of the makers of a commercial program to do that says, "That's what we're trying to optimize and look, we achieved that. We've, uh, we've reached that kind of, uh, of balance." And then on the other side, you also want to say, uh, "Well, if, if it makes mistakes, I want that to affect both sides of the protected class equally."... and it turns out they don't do that, right? So they're- they're twice as likely to make a mistake that would harm a- a Black person over a white person, so that seems unfair. So you'd like to say, "Well, I want to achieve both those goals," and then it turns out, you do the analysis, and it's theoretically impossible to achieve both those goals. So you have to trade them off, one against the other. So that analysis is really helpful to know, uh, what you can aim for and how much you can get, that you can't have everything. But the analysis certainly can't tell you where should we make that trade-off point.

   4. LFLex Fridman6:38

      But nevertheless, then we can, uh, as humans, deliberate where that trade-off should be. Okay.

   5. PNPeter Norvig6:43

      Yeah. So at least we now, we're- we're arguing in an informed way.

   6. LFLex Fridman6:45

      Right.

   7. PNPeter Norvig6:45

      Uh, we're not asking for something impossible. We're saying, uh, "Here's where we are and- and here's what we aim for, and, uh, this, uh, strategy is better than that strategy."

   8. LFLex Fridman6:55

      So that's, I would argue, is- is a really powerful and really important first step, uh, but it's a doable one, sort of removing, uh, undesirable degrees of bias in, uh, in systems-

   9. PNPeter Norvig7:07

      Mm-hmm.

4. 7:07 – 9:09

   Attention economy and “dopamine optimization” vs long-term human benefit

   1. LFLex Fridman7:07

      ... in terms of protected classes. And then there's something, I listened to your, uh, commencement speech or there's some fuzzier things, like you mentioned Angry Birds.

   2. PNPeter Norvig7:16

      Yeah. (laughs)

   3. LFLex Fridman7:17

      Do you want- do you want to create systems that feed the dopamine enjoyment, uh, that feed- that optimize for you returning to the system, enjoying the moment of playing the game, of getting likes or whatever-

   4. PNPeter Norvig7:30

      Mm-hmm.

   5. LFLex Fridman7:30

      ... this kind of thing, or some kind of long-term improvement?

   6. PNPeter Norvig7:34

      Right.

   7. LFLex Fridman7:35

      It's, uh, v- are you even thinking about that? Uh, that's ex- well, th- that's really going into the phil- philosophical area.

   8. PNPeter Norvig7:43

      Yeah. Uh, I think that's a, a really important issue too, certainly thinking about that. I- I don't think about that...

   9. LFLex Fridman7:48

      Uh-huh.

   10. PNPeter Norvig7:49

       ... as a- as an AI issue as much.

   11. LFLex Fridman7:50

       Mm-hmm.

   12. PNPeter Norvig7:52

       But as you say, you know, the- the point is we've built this society and this infrastructure where we say, "We have a marketplace for attention and, uh, we've decided as a society that we like things that are free and so we want all the apps on our phone to be free." Uh, and that means they're all competing for your attention and then eventually they- they make some money some way through, uh, ads or in-game sales or whatever. But they can only win by, uh, defeating all the other apps by- in stealing your attention. And we build a marketplace where it seems like they're, uh, working against you rather than working with you. And I'd like to find a way where we can change the playing field so you feel more like, "Well, these things are on my side. Yes, they're ha- letting me have some fun in the short term, but they're also helping me in the long term rather than competing against me."

   13. LFLex Fridman8:55

       And those aren't necessarily conflicting objectives.

   14. PNPeter Norvig8:57

       Right.

   15. LFLex Fridman8:57

       They're just, uh, the incentives, the direct current incentives as we try to figure out this whole new world, uh, seem to be, um, uh, the- the easier part of that, which is feeding the dopamine, uh, the rush.

   16. PNPeter Norvig9:09

       Right.

5. 9:09 – 11:40

   Writing AIMA in the 1990s: why it happened and what wave it captured

   1. LFLex Fridman9:09

      But, uh, so maybe taking a quick step back, uh, at the beginning of the Artificial Intelligence: The Modern Approach book or writing, so here you are in the '90s when you first sa- uh, sat down with Stuart to write the book, uh, to cover an entire field, which is one of the only books that has successfully done that for AI and actually in- in a lot of other computer science fields. You know, it's a diff- it's a h- it's a huge undertaking. So, m- th- it must've been quite daunting. What was that process like? Did you envision that you would be trying to cover the entire field? Was there a systematic approach to it that was more step-by-step? How was-

   2. PNPeter Norvig9:50

      Yeah.

   3. LFLex Fridman9:50

      ... how did it feel?

   4. PNPeter Norvig9:52

      So I guess it came about, you know, I'd go to lunch with the other AI faculty at Berkeley and we'd say, uh, you know, "The field is changing. Seems like the current books are a little bit behind. Nobody's come out with a new book recently. We should do that." And everybody said, "Yeah, yeah. That's a great thing to do." Uh, and we never did anything.

   5. LFLex Fridman10:10

      (laughs) Right.

   6. PNPeter Norvig10:11

      And then I ended up, uh, heading off to, uh, industry. I went to, uh, Sun Labs, so I thought, "Well, that's the end of my possible academic publishing career."

   7. LFLex Fridman10:20

      Mm-hmm.

   8. PNPeter Norvig10:21

      But I met Stuart again at a conference, like a year later and said, "You know that book we were always talking about? You guys must be half done with it by now, right?"

   9. LFLex Fridman10:30

      (laughs)

   10. PNPeter Norvig10:32

       And he said, "Well, we keep talking. We never do anything."

   11. LFLex Fridman10:34

       Right.

   12. PNPeter Norvig10:34

       So I said, "Well, you know, we should do it." And I think the reason is that we all felt it was a time where the field was changing, and th- that was in two ways. So, you know, the good old-fashioned AI was based, uh, primarily on Boolean logic, and you had a few tricks to deal with uncertainty. And it was based pr- primarily on knowledge engineering-

   13. LFLex Fridman11:00

       Mm-hmm.

   14. PNPeter Norvig11:00

       ... that the way you got something done is you went out and you interviewed an expert and you wrote down by hand everything they knew. And we saw in- in, uh, '95 that the field was changing in- in two ways. One, we were moving more towards probability ru- rather than Boolean logic, and we were moving more towards machine learning rather than knowledge engineering.

   15. LFLex Fridman11:20

       Mm-hmm.

   16. PNPeter Norvig11:21

       Uh, and the other books, uh, hadn't caught that wave. They were still in the, uh, more in the- in the old school, although certainly they had part of that, uh, on the way. But we said, "If we start now completely taking that point of view, we can have a- a different kind of book." And we were able to put that together.

6. 11:40 – 13:38

   How they built the textbook remotely—and what they missed about the future

   1. LFLex Fridman11:40

      And, uh, what was literally the process if you remember? What... Did you start writing a chapter? Did you outline s-

   2. PNPeter Norvig11:48

      Yeah. I guess, I guess we did an- an outline and then we sort of assigned chapters to each person. At the time, uh, I had moved to Boston and Stuart- Stuart was in Berkeley, so basically, uh, we did it, uh, uh, over the internet and- and, uh, you know, that's n- that wasn't the same as doing it today. (laughs) It meant, uh, you know, uh, dial-up lines and Telnetting in-

   3. LFLex Fridman12:13

      Right.

   4. PNPeter Norvig12:13

      ... and, and (laughs) -

   5. LFLex Fridman12:13

      (laughs)

   6. PNPeter Norvig12:14

      ... you know, you, uh, you Telnet it into, uh, one shell and you type cat filename and you-

   7. LFLex Fridman12:21

      Right.

   8. PNPeter Norvig12:21

      ... hoped it was captured at the other end and...

   9. LFLex Fridman12:23

      And certainly you're not sending, uh, images and figures back and forth.

   10. PNPeter Norvig12:27

       Right, right, that didn't work.

   11. LFLex Fridman12:29

       (laughs) But, you know, did you anticipate where the field would go, uh, from that day, uh, from, uh, from the '90s? Did you see the growth into learning-based methods, into data-driven methods, that followed in the future decades?

   12. PNPeter Norvig12:47

       W- we certainly thought that, uh, learning was important. I guess we, w- we missed it as, uh, being as important as it, as it is today. We mi- we missed this idea of big data. We missed it, uh, uh... the idea of deep learning hadn't been invented yet. We could have, uh, taken the book from a complete, uh, machine learning point of view right from the start. We chose to do it more from a point of view of, we're gonna first develop the different types of representations and we're gonna talk about different types of environments, of, uh, is it fully observable or partially observable and is it, uh, deterministic or stochastic and so on. And we, uh, made it more complex along those axes rather than, uh, focusing on the machine learning axis first.

7. 13:38 – 15:41

   Deep learning’s place in the bigger AI toolbox (and Ian Goodfellow’s chapter)

   1. LFLex Fridman13:38

      Do you think... You know, there's some sense in which the deep learning craze, uh, is extremely successful for a particular set of problems and, you know, eventually it's going to, in the general case, hit challenges. And so in terms of the difference between, uh, perception systems and robots that have to act in the world, do you think, uh, we're gonna return to AI modern approach type breadth in edition five and six-

   2. PNPeter Norvig14:08

      Yeah.

   3. LFLex Fridman14:09

      ... in, uh, in future decades? Do you think, uh, deep learning will take its place as a chapter in this bigger, as, in this bigger, uh, view of AI?

   4. PNPeter Norvig14:18

      Yeah, I think we don't know yet how it's all gonna play out. So, uh, i- in the new edition, uh, we have a chapter on deep learning. Uh-

   5. LFLex Fridman14:26

      Right.

   6. PNPeter Norvig14:26

      We got Ian Goodfellow to be the, uh-

   7. LFLex Fridman14:28

      (laughs)

   8. PNPeter Norvig14:28

      ... guest author for that chapter.

   9. LFLex Fridman14:30

      Great.

   10. PNPeter Norvig14:30

       So he, he said he could condense his whole, uh-

   11. LFLex Fridman14:33

       (laughs)

   12. PNPeter Norvig14:33

       ... deep learning book into one chapter. I, I think he did a great job. We were also encouraged that he's, you know, we gave him the old, uh, neural net chapter and said, uh-

   13. LFLex Fridman14:45

       (laughs) "Have fun with it."

   14. PNPeter Norvig14:46

       ... "Modernize that." And he said, "You know, half of that was okay."

   15. LFLex Fridman14:49

       Yeah.

   16. PNPeter Norvig14:50

       That, uh, certainly there's lots of new things that have been developed, but some of the core was still the same. So I think we'll gain a better understanding of what you can do there. I think we'll need to incorporate all the things we can do with the other technologies.

   17. LFLex Fridman15:09

       Mm-hmm.

   18. PNPeter Norvig15:09

       Right? So deep learning started out doing convolutional networks and, uh, very close to perception, uh, and has since moved to be, uh, to be able to do more with, uh, actions and some degree of, of longer term planning. Uh, but we need to do a better job with, uh, representation and reasoning and, uh, one-shot learning and so on, and w- well, I think we don't know yet how that's gonna play out.

8. 15:41 – 18:31

   Symbolic AI’s lessons: representation, messy concepts, and when reasoning applies

   1. LFLex Fridman15:41

      So do you think looking at the, some success but certainly, uh, eventual demise, the partial demise of experts as symbolic, uh, systems in the '80s, do you think there is kernels of wisdom in the work that was done there with logic and reasoning and so on that will rise again in your view?

   2. PNPeter Norvig16:05

      So certainly I think the idea of representation and reasoning is crucial, that, uh, you know, sometimes you just don't have enough data about the world to learn de novo, uh, so you've got to have a, a, some idea of representation, whether that was programmed in or told or whatever, and then be able to take, uh, steps of reasoning. I, I think the problem, uh, with, uh, you know, the good old-fashioned AI was, uh, one, we tried to base everything on these, uh, symbols that were atomic. And that's great if you're, like, trying to define the properties of a triangle.

   3. LFLex Fridman16:47

      Right.

   4. PNPeter Norvig16:47

      Right? Because they have necessary and sufficient conditions. Uh, but things in the real world don't. The real world is, is messy and doesn't have sharp edges, and atomic symbols do. So that was a, a poor match. And then the other aspect was that the, uh, reasoning was universal and applied anywhere, which in some sense is good, but it also means there's no guidance as to where to apply.

   5. LFLex Fridman17:14

      Mm-hmm.

   6. PNPeter Norvig17:15

      And so you, you know, you started getting these paradoxes like, uh, uh, "Well, if I have a mountain and I remove one grain of sand, uh, then it's still a mountain," and, "But if I do that repeatedly, at some point it's not," right? And, uh, with logic, you know, there's nothing to stop you from applying things, uh, repeatedly. Uh, but maybe with, uh, something like, uh, deep learning, and I don't really know what the right name for it is, uh, we could separate out those ideas. So, one, we could say, uh, you know, a mountain isn't just an atomic notion. It, it's some sort of, something like a word embedding that, uh, uh, has a, uh, a more complex representation.

   7. LFLex Fridman18:01

      Mm-hmm. Yeah.

   8. PNPeter Norvig18:02

      And secondly, we could somehow learn, yeah, there's this rule that you can remove one grain of sand, uh, and you can do that a bunch of times but you can't do it, uh, a near infinite amount of times. But on the other hand, when you're doing induction on the integers, sure, then it's fine to do it an infinite number of times. And if we could l- uh, somehow we have to learn when these strategies are applicable-... rather than having the strategies be completely neutral and avai- uh, available everywhere.

9. 18:31 – 23:12

   Beyond explainability: trust, verification, and adversarial robustness

   1. LFLex Fridman18:31

      Anytime you use neural networks, anytime you learn from data or form representation from data in an automated way, it's not very explainable as to, uh, or it's not introspective to us humans in terms of, uh, how this neural network sees the world. Where... Why does it succeed so brilliantly on so many, in so many cases and fail so miserably in surprising-

   2. PNPeter Norvig18:55

      Yeah.

   3. LFLex Fridman18:55

      ... ways and small. So, what do you think is this... is, um, the future there? Can simply more data, better data, more organized data solve that problem, or is there elements of symbolic systems that need to be brought in which are a little bit more explainable?

   4. PNPeter Norvig19:12

      Yeah. So, I prefer to talk about trust and, uh, validation and verification rather than just about explainability. And then I think, uh, explanations are one tool that you use towards those goals. And I think it is an important issue that, uh, we don't want to use these systems unless we trust them and we want to understand where they work and where they don't work, and- and an explanation can be part of that, right? So, I apply for loan and I get, uh, denied, uh, I want some explanation of why, and uh, you have, uh, in Europe we have the GDPR that says, uh, you're required to be able to get that. But on the other hand, an explanation alone is not enough, right? So, you know, we were used to dealing with people and with, uh, organizations and corporations and so on, and they can give you an explanation and you have no guarantee that that explanation relates to reality.

   5. LFLex Fridman20:10

      Right.

   6. PNPeter Norvig20:10

      Right? So, the bank can tell me, "Well, you didn't get the loan 'cause you didn't have enough collateral," and that may be true or it may be true that they just didn't like my, uh, religion or- or something else. Uh, I can't tell from the explanation, and that's, uh, that's true whether the decision was made by a computer or by a person. So, I want more. I do want to have the explanations and I want to be able to, uh, have a conversation to go back and forth-

   7. LFLex Fridman20:38

      Mm-hmm.

   8. PNPeter Norvig20:38

      ... and said, "Well, you gave this explanation, but what about this?"

   9. LFLex Fridman20:41

      Mm-hmm.

   10. PNPeter Norvig20:41

       "And what would have happened if this had happened? And, uh, what would- what would I need to change that?" So, I think a conversation is- is a better way to think about it than just, uh, an explanation as a single output. Uh, and I think we need testing of various kinds, right? So, in order to know, was the decision really based on my collateral or was it based on my, uh, religion or skin color or whatever? I can't tell if I'm only looking at my case, but if I look across all the cases, then I can detect a pattern.

   11. LFLex Fridman21:15

       Right.

   12. PNPeter Norvig21:15

       Right? So, you want to have that kind of capability. Uh, you want to have these adversarial testing, right? So, we thought we were doing pretty good at, uh, object recognition in- in images. We said, "Look, we're- we're at sort of pretty close to human level of performance on ImageNet," and so on. Uh, and then you start seeing these adversarial images and you say, "Wait a minute, that part is nothing like (laughs) human performance." Uh-

   13. LFLex Fridman21:39

       Yeah, you can mess with it really easily.

   14. PNPeter Norvig21:40

       You can mess with it really easily, right?

   15. LFLex Fridman21:42

       Yeah.

   16. PNPeter Norvig21:42

       And, uh, yeah, you can do that to humans too, right? So...

   17. LFLex Fridman21:46

       In a different way, perhaps.

   18. PNPeter Norvig21:47

       Right. Humans don't know what color the dress was.

   19. LFLex Fridman21:49

       Right.

   20. PNPeter Norvig21:50

       And so they're vulnerable to certain attacks that are different than the attacks on the- on the machines, but the, you know, the attacks on the machines are so striking, uh, they really change the way you think about what we've done, right?

   21. LFLex Fridman22:03

       Mm-hmm.

   22. PNPeter Norvig22:03

       And the- and the way I think about it is, I think part of the problem is we're seduced by, uh, our low-dimensional metaphors, right?

   23. LFLex Fridman22:13

       (laughs) Yeah. I like-

   24. PNPeter Norvig22:14

       So, you know, you look-

   25. LFLex Fridman22:15

       I like that phrase. (laughs)

   26. PNPeter Norvig22:15

       You look in a- in a textbook and you say, "Okay, now we've mapped out the space and, you know, uh, cat is here and dog is here, and maybe there's a tiny little spot in the middle-

   27. LFLex Fridman22:27

       Yeah.

   28. PNPeter Norvig22:27

       ... where you can't tell the difference, but mostly we've got it all covered." And if you believe that metaphor, uh, then you say, "Well, we're nearly there," and, uh, you know, there's only going to be a couple adversarial images.

   29. LFLex Fridman22:38

       Yeah.

   30. PNPeter Norvig22:39

       Uh, but I think that's the wrong metaphor, and what you should really say is, "It's not a 2D flat space that we've got mostly covered, it's a million dimension space and, uh, cat is this string that goes out in this crazy path, and if you step a little bit off the path in any direction, you're in Nowhere's, m- m- land and you don't know what's going to happen." And so I think that's where we are, and- and now we've got to deal with that. So, uh, it wasn't so much an explanation, but it was an- an understanding of what the models are and what they're doing, and now we can start exploring how do you fix that.

10. 23:12 – 25:45

    Humans vs AI: why we demand higher standards and what trust looks like socially

    1. LFLex Fridman23:12

       Yeah, validating the robustness of the system, so on. But take it back to the... this, uh, this word trust. Uh, do you think we're a little too hard on our robots in terms of-

    2. PNPeter Norvig23:23

       (laughs) .

    3. LFLex Fridman23:23

       ... uh, the standards we apply? So, you know, of, uh, there's a dance, there's a- there's a- there's a dance in non-verbal and verbal communication between humans. You know, if we apply the same kind of standard in terms of humans, you know, we trust each other pretty quickly. Uh, you know, you and I haven't met before and there's some degree of trust (laughs) .

    4. PNPeter Norvig23:48

       Yeah.

    5. LFLex Fridman23:48

       Right? That, uh, nothing's gonna go crazy wrong. And yet to AI, when we look at AI systems or... we seem to approach, uh, through skepticism always, always.

    6. PNPeter Norvig23:58

       Yeah.

    7. LFLex Fridman23:58

       And it's like they have to prove through a lot (laughs) of hard work that they're even worthy of, uh, even inkling of our trust. What do- what do you- what do you think about that? How- how do we break that barrier, close that gap?

    8. PNPeter Norvig24:11

       I think that's right. I think that's a big issue. Uh, just listening, uh, my friend, uh, Mark Moffett is a naturalist and he says, uh, "The most amazing thing about humans is that you can walk into a- a coffee shop or a, uh, a busy street in a city..."... and there's lots of people around you that you've never met before, and you don't kill each other.

    9. LFLex Fridman24:33

       (laughs) Yeah.

    10. PNPeter Norvig24:34

        He says, "Chimpanzees cannot do that."

    11. LFLex Fridman24:36

        Yeah, right. (laughs)

    12. PNPeter Norvig24:37

        Right? If a chimpanzee is in a situation where, "Here's some, uh, that aren't from my tribe..." Bad things happen.

    13. LFLex Fridman24:46

        Especially in a coffee shop, there's delicious food around, you never know.

    14. PNPeter Norvig24:49

        Yeah, yeah. But, but we humans have figured that out.

    15. LFLex Fridman24:52

        Yeah.

    16. PNPeter Norvig24:52

        Right? Uh, and you know-

    17. LFLex Fridman24:54

        For the most part.

    18. PNPeter Norvig24:55

        ... for the most part. We still go to war, we still do terrible things, uh, but for the most part, we've learned to trust each other and, and live together. Uh, so that's gonna be important for our, uh, our AI systems as well, and I th- also, I think, uh, you know, a lot of the emphasis is on AI, uh, but in many cases, uh, AI is part of the technology, but isn't really the main thing. So a lot of, of what we've seen is more due to communications technology-

    19. LFLex Fridman25:24

        Mm-hmm.

    20. PNPeter Norvig25:25

        ... than AI te- AI technology. Yeah, you wanna make these good decisions, but the reason, uh, we're able to have any kind of system at all is we've got the communication so that we're collecting the data and so that we can reach lots of people around the world. I think that's a, a bigger change that we're dealing with.

11. 25:45 – 28:56

    MOOCs at massive scale: motivation beats information, community beats content alone

    1. LFLex Fridman25:45

       Speaking of reaching a lot of people around the world, on the side of education, you've, uh ... one of the many things in terms of education you've done, you taught the Intro to Artificial Intelligence course that signed up 100,000- 160,000 students. It was one of the first successful example of an massive, uh, of a MOOC, massive open online course. What did you learn from that experience? Uh, what do you think is the future of MOOCs, of education online?

    2. PNPeter Norvig26:12

       Yeah. It was a great fun doing it, particularly, uh, being right at the start just because it was exciting and new, but it also meant that we had less competition.

    3. LFLex Fridman26:24

       (laughs) Yeah.

    4. PNPeter Norvig26:24

       Right? So, uh, one of the things you hear about, uh, "Well, the problem with MOOCs is, uh, the completion rates are, are so low, so they must be a failure." And, and I got to admit, I'm a prime contributor, right?

    5. LFLex Fridman26:37

       Mm-hmm.

    6. PNPeter Norvig26:37

       I've probably, uh, started 50 different courses that I haven't finished, but I got exactly what I wanted out of them because I had never intended to finish them. I just wanted to, uh, dabble in a little bit, either to see the topic matter or just to see the pedagogy of, "How are they doing this class?" So I guess the main thing I learned is, when I came in, I thought, uh, the challenge was information, saying, "If I'm just to take the stuff I want you to know and I'm very clear and explain it well, then my job is done and, uh, good things are gonna happen." Uh, and then in, in doing the course, I learned, uh, well, yeah, you gotta have the information, but really, the motivation is the most important thing, that, uh, if students don't stick with it, then it doesn't matter how good the content is.

    7. LFLex Fridman27:29

       Mm-hmm.

    8. PNPeter Norvig27:29

       Uh, and I think being one of the first classes, we were helped by a sort of exterior motivation. So, we tried to do a good job at making it enticing and setting up, uh, uh, ways for, uh, you know, the community to work with each other, to make it more motivating, but really, a lot of it was, "Hey, this is a, a new thing, and I'm really excited to be part of a new thing." And so the students brought their own motivation. And so I think this is great because there's lots of people around the world who have never had this before-

    9. LFLex Fridman28:00

       Mm-hmm.

    10. PNPeter Norvig28:00

        ... you know, uh, who'd never have the opportunity to, uh, go to Stanford and take a class, or go to MIT, or go to one of the other schools. Uh, but now we can bring that to them, and if they bring their own motivation, uh, they can be successful in a way they couldn't before. But that's really just the top tier of people that are ready to do that. The rest of the people, uh, j- just don't see or, you know, don't have the motivation and don't see how, if they push through and were able to do it, what advantage that would get them. Uh, so I think we got a long way to go before we're able to do that, and I think it'll be m- some of it is based on technology, but more of it's based on the idea of community, that you gotta actually get people together. Some of that getting together can be done online. I think some of it really has to be done in person to be able to, in order to build that type of, uh, community and trust.

12. 28:56 – 32:42

    Online vs in-person education: commitment, social pressure, and what universities will do

    1. LFLex Fridman28:56

       You know, there's an intentional mechanism that we've developed, uh, a short attention span, especially younger people, um, because sort of shorter and shorter videos online, uh, there's a, whatever the, th- the way the brain is dev- is developing now with people that have grown up with the internet, they have a, quite a short attention span. So, and I, I would say I had the same when I was growing up too, probably for different reasons. So, I probably wouldn't have, uh, learned as much as I have if I wasn't forced to sit in a physical classroom-

    2. PNPeter Norvig29:30

       Mm-hmm.

    3. LFLex Fridman29:31

       ... sort of bored, sometimes falling asleep-

    4. PNPeter Norvig29:34

       Sure. (laughs)

    5. LFLex Fridman29:34

       ... but sort of forcing myself through that process in sometimes extremely difficult computer science courses. What, what's the difference, in your view, between in-person education experience, which you, uh, first of all, yourself had and you yourself taught-

    6. PNPeter Norvig29:50

       Yeah.

    7. LFLex Fridman29:50

       ... and online education?

    8. PNPeter Norvig29:52

       Right.

    9. LFLex Fridman29:52

       And how do we close that gap, if it's even possible?

    10. PNPeter Norvig29:54

        Yeah. So, I think there's two issues. One is whether it's in person or online, so sort of the physical location, and then the other is, uh, kind of the affiliation, right? So, you stuck with it in part because you were in the classroom and you saw everybody else was suffering-

    11. LFLex Fridman30:14

        Right. (laughs)

    12. PNPeter Norvig30:15

        ... the same, the same way you were.

    13. LFLex Fridman30:16

        Yeah.

    14. PNPeter Norvig30:16

        Uh, but also because you were enrolled, you had paid tuition.

    15. LFLex Fridman30:21

        Yeah.

    16. PNPeter Norvig30:22

        Sort of everybody was expecting you to stick with it. Uh-

    17. LFLex Fridman30:25

        Mm-hmm. Society, parents-

    18. PNPeter Norvig30:26

        Yeah.

    19. LFLex Fridman30:27

        ... c- uh, peers.

    20. PNPeter Norvig30:28

        Right.

    21. LFLex Fridman30:29

        Yeah.

    22. PNPeter Norvig30:29

        And so those are two separate things. I mean, you could certainly imagine I pay a huge amount of tuition (laughs) and everybody signed up and says, "Yes, you're doing this," uh...... but then I'm in my room and my classmates are in, are in different rooms, right? We c- we could have things set up that way. Uh, so it's not just the online versus offline. I think what's more important is the commitment, uh, that you've made. And certainly, it is important to have that kind of informal, uh, you know, I meet people outside of class, we talk together because we're all in it together. Uh, I think that's, uh, really important both in keeping your motivation and also that's where some of the most important learning goes on. So, you wanna have that. Uh, maybe, you know, e- especially now, we start getting into higher bandwidths and augmented reality and virtual reality, you might be able to get that without being in the same physical place.

    23. LFLex Fridman31:26

        Do you think it's possible we'll see a course at Stanford, for example, that, for students, enrolled students, is only online in the near future? Where literally, sort of it's part of the curriculum and there is no...

    24. PNPeter Norvig31:41

        Yeah. So, you're starting to see that. Uh, I know, uh, Georgia Tech has a master's, uh, that's done that way.

    25. LFLex Fridman31:46

        Oftentimes, it's sort of they're creeping in, in terms of a master's program or sort of a further education-

    26. PNPeter Norvig31:54

        Yeah.

    27. LFLex Fridman31:54

        ... considering the constraints of students and so on. But I mean literally, is it possible that we, uh, you know, Stanford, MIT, Berkeley, all these places go online only in, uh, in the next few decades?

    28. PNPeter Norvig32:07

        Y- yeah, probably not 'cause, you know, they've got a big, uh, commitment to a physical campus.

    29. LFLex Fridman32:13

        Sure.

    30. PNPeter Norvig32:13

        Right?
13. 32:42 – 37:16

    Learning to program: problem-solving, modeling, and being comfortable with uncertainty

    1. LFLex Fridman32:42

       Right. So our field, programming, you've also done a lot of, you've done a lot of programming yourself. In, uh, 2001, you wrote a great article about programming called Teach Yourself Programming in 10 Years. Sort of responds to-

    2. PNPeter Norvig32:58

       Yeah.

    3. LFLex Fridman32:58

       ... all the books that say Teach Yourself Programming in 21 Days. So, if you were giving advice to someone getting into programming today, this is, uh, a few years since you've written that article, what's the best way to undertake that journey?

    4. PNPeter Norvig33:10

       I think there's lots of different ways and I think, uh, programming means more things now. And I guess, you know, when I wrote that article, I was thinking more about becoming a professional software engineer. And I thought that's a, you know, a c- sort of a career-long, uh, field of study. Uh, but I think there's lots of things now that people can do where programming is a part of solving what they wanna solve, uh, without it achieving that professional-level status.

    5. LFLex Fridman33:44

       Yeah.

    6. PNPeter Norvig33:44

       Right? So, I'm not gonna be going and writing a million lines of code, but, you know, I'm a biologist or a physicist or something, or a, even a historian, and I've got some data, and I wanna ask a question of that data. And I think for that, uh, you don't need 10 years, right? So, eith- there are many shortcuts to, uh, being able to a- answer those kinds of questions. And, and you know, you see today a lot of, uh, emphasis on, uh, learning to code-

    7. LFLex Fridman34:13

       Mm-hmm.

    8. PNPeter Norvig34:14

       ... teaching kids how to code. Uh, I think that's great. Uh, but I wish they would change the message a little bit, right? So, I think code isn't the main thing. I don't really care if you know the syntax of JavaScript or if you can, uh, connect these blocks together in this visual language. Uh, but what I do care about is that you can analyze a problem, uh, you can, uh, think of a solution, you can, uh, carry out, uh, you know, make a model, run that model, test the model, see the results, uh, uh, verify that they're reasonable, uh, ask questions and answer them. Right? So, it's more, uh, modeling and problem-solving-

    9. LFLex Fridman34:58

       Mm-hmm.

    10. PNPeter Norvig34:58

        ... and you use coding in order to do that, uh, but it's not just learning coding for its own sake.

    11. LFLex Fridman35:04

        That's really interesting. So, it's actually almost, in many cases, it's learning to work with data, to extract-

    12. PNPeter Norvig35:10

        Yeah.

    13. LFLex Fridman35:10

        ... something useful-

    14. PNPeter Norvig35:11

        Yeah.

    15. LFLex Fridman35:11

        ... out of data. So, when you say problem-solving, you really mean taking some kind of, maybe collecting some kind of dataset, cleaning it up, and saying something interesting about it-

    16. PNPeter Norvig35:20

        Yeah.

    17. LFLex Fridman35:20

        ... which is useful in all kinds of domains.

    18. PNPeter Norvig35:23

        And, uh, you know, and I see myself, uh, being stuck sometimes in kind of the, the old ways.

    19. LFLex Fridman35:29

        Mm-hmm.

    20. PNPeter Norvig35:30

        Right? So, you know, I'll be working on a project, uh, maybe with a, a younger employee and we say, "Oh, well, here's this new package that could help solve this, uh, problem." And I'll go and I'll start reading the manuals and, you know, I'll be-

    21. LFLex Fridman35:45

        (laughs)

    22. PNPeter Norvig35:45

        ... two hours into reading the manuals and then, uh, my colleague comes back and says, "I'm done."

    23. LFLex Fridman35:51

        Yup.

    24. PNPeter Norvig35:51

        You know? "I downloaded the package. I installed it. I tried calling some things. The first one didn't work. The second one worked. Now I'm done." And, and I say, "But I have 100 questions about how does this work and how does that work?" And they say, "Who cares," right? "I don't need to understand the whole thing. I unders- I answered my question. It's a big complicated package. I don't understand the rest of it, but I got the right answer." And I'm just, it's hard for me to get into that mindset. I want to understand the whole thing and, you know, if they wrote a manual, I should probably read it. And, but that's not necessarily the right way. And I, I think I have to get used to dealing with more, being more comfortable with uncertainty and not knowing everything.

    25. LFLex Fridman36:32

        Yeah. So, I struggle with the same. It's sort of the, the spectrum between Donald, Don Knuth-

    26. PNPeter Norvig36:37

        Yeah.

    27. LFLex Fridman36:37

        ... who's kind of the very, you know, b- before he can say anything m-... about a problem, he really has to get down to the machine code as, uh, assembly.

    28. PNPeter Norvig36:45

        Yeah.

    29. LFLex Fridman36:46

        Versus exactly what you said, of, uh... of several students in my group that, uh, you know, 20 years old and they can solve almost any problem within a few hours that would take me probably weeks, because I would try to, as you said, read the manual. So, do you think the nature of mastery is... You're, you're mentioning biology, sort of outside disciplines, applying programming, but computer scientists... So over time, there's higher and higher levels

14. 37:16 – 48:31

    Modern software engineering: abstraction, hiring, code review, and Lisp’s legacy to Python

    1. LFLex Fridman37:16

       of abstraction available now. So with, with, uh, this week-

    2. PNPeter Norvig37:20

       Hm. Yeah.

    3. LFLex Fridman37:20

       ... there's the, the TensorFlows Summit, right?

    4. PNPeter Norvig37:23

       Yeah.

    5. LFLex Fridman37:23

       (laughs) So if you're, if you're not particularly into deep learning, but you're still a computer scientist, uh, you can accomplish an incredible amount with, uh, TensorFlow without really knowing any fundamental internals of machine learning. Do you think the nature of mastery is, is changing, uh, even for computer scientists, like what it means to be an expert programmer?

    6. PNPeter Norvig37:45

       Yeah, I think that's true. You know, we never really should have focused on programmer, right? Because that's still... it's, it's a skill and what we really want to focus on is the result. So we, we built this, uh, ecosystem where the way you can get stuff done is by programming it yourself.

    7. LFLex Fridman38:03

       Right.

    8. PNPeter Norvig38:04

       At least when I started, it... you know, library functions meant you had square root and that was about it.

    9. LFLex Fridman38:10

       (laughs)

    10. PNPeter Norvig38:10

        (laughs) Right?

    11. LFLex Fridman38:10

        Yeah.

    12. PNPeter Norvig38:10

        Everything else you built from scratch.

    13. LFLex Fridman38:12

        Yeah.

    14. PNPeter Norvig38:13

        And then we built up an ecosystem where a lot of times, well, you can download a lot of stuff that does-

    15. LFLex Fridman38:18

        Yeah.

    16. PNPeter Norvig38:18

        ... a big part of what you need. And so now, it's more a question of, uh, assembly rather than, uh, uh, manufacturing, and, uh, that's a different way of looking at problems.

    17. LFLex Fridman38:32

        From another perspective, in terms of mastery and looking at programmers or people that reason about problems in a computational way, so Google, uh, you know, the... from the hiring perspective, from the perspective of hiring or building a team of programmers, uh, how do you determine if someone's a good programmer? Or if somebody... again-

    18. PNPeter Norvig38:52

        Yeah.

    19. LFLex Fridman38:52

        ... because I don't want to deviate from... I want to move away from the word programmer, but somebody who could solve problems of large-scale data and so on. What's, what's, uh... how do you build a team like that through the interviewing process for them?

    20. PNPeter Norvig39:03

        Yeah, and I, and I think, uh, as a company grows, uh, you get more, uh, expansive in the types of people you're looking for, right? So, uh, I think, you know, in the early days, we'd interview people and the question we were trying to ask is, uh, how close are they to Jeff Dean?

    21. LFLex Fridman39:23

        (laughs)

    22. PNPeter Norvig39:25

        (laughs) And most people-

    23. LFLex Fridman39:26

        Sure.

    24. PNPeter Norvig39:26

        ... were pretty far away.

    25. LFLex Fridman39:26

        Yeah.

    26. PNPeter Norvig39:26

        But we'd take the ones that were, you know, not that far away.

    27. LFLex Fridman39:28

        Yeah.

    28. PNPeter Norvig39:29

        And so we got kind of a homogeneous group of people who were really great programmers.

    29. LFLex Fridman39:33

        Yeah.

    30. PNPeter Norvig39:34

        Uh, then as the company grows, you say, "Well, we don't want everybody to be the same, to have the same skillset." And so now, we're, uh, hiring, uh, biologists in our health areas and we're hiring physicists and we're hiring, uh, mechanical engineers and we're hiring, uh, you know, uh, social scientists and ethnographers and people with different backgrounds, uh, who bring different skills.
15. 48:31 – 53:23

    Early Google search quality: metrics, adversarial SEO, and reshaping the web

    1. LFLex Fridman48:31

       Okay, awesome. So, you were the director of search quality at Google from-

    2. PNPeter Norvig48:36

       Yeah.

    3. LFLex Fridman48:36

       ... 2001, '02, to 2005. In the early days, uh, when there was just a few employees and when the-

    4. PNPeter Norvig48:43

       Yeah.

    5. LFLex Fridman48:43

       ... when the company was growing like crazy, right? So-Th- I mean, Google revolutionized the way we discover, and share, and aggregate knowledge, so just, this is, uh, this is one of the fundamental aspects of civilization, right, is information being shared, and there's different mechanisms throughout history, but Google has just 10X improved that, right? And you're a part of that, right? People discovering that information. So, what, what were some of the challenges on a philosophical or the technical level in those early days?

    6. PNPeter Norvig49:18

       It definitely was an exciting time, and as you say, we were doubling in size every year. And the challenges were, we wanted to get the right answers, (laughs) right? And, uh, we had to figure out what that meant. We had to implement that, and we had to make it all, uh, efficient and, uh... We had to keep on testing and seeing if we were delivering good answers. Uh-

    7. LFLex Fridman49:44

       And now when you say good answers, it means whatever people are typing in, in terms of keywords, in terms of that kind of thing, that the- that the results they get are ordered by the desirability for them of those results. Like, they're like... The first thing they click on will likely be the thing that they were actually looking for.

    8. PNPeter Norvig50:01

       Right. One of the metrics we had was focused on the first thing. Uh, some of it wa- fo- was focused on the whole page. Some of it was focused on, you know, the top three or so. So, we looked at a lot of different metrics for- for how well we were doing, and we broke it down into subclasses of, you know, maybe here's a type of, uh- of, uh, query that we're not doing well on. Then we try to fix that. Uh, early on, we started to realize that we were in an adversarial position, right? So, we started thinking, uh, "Well, we're kind of like the card catalog in the library."

    9. LFLex Fridman50:35

       Mm-hmm.

    10. PNPeter Norvig50:36

        Right? So, the books are here, and we're off to the side and- and we're just, uh, reflecting what's there. And then we realized every time we make a change, the webmasters make a change.

    11. LFLex Fridman50:47

        (laughs)

    12. PNPeter Norvig50:48

        And it's, uh, game theoretic. And so we had to think not only of, is this the right move for us to make now? But also, if we make this move, what's the countermove gonna be? Is that gonna get us into a work- worse place? In which case we won't make that move, we'll make a different move.

    13. LFLex Fridman51:05

        And did you find... I mean, I assume with the popularity and the growth of the internet, that people were creating new content. So, you're almost helping guide the creation of new content.

    14. PNPeter Norvig51:14

        Yeah. So, that's certainly true, right? So, we- we kn- we definitely changed, uh, the structure of the network, right? So, if you think back, you know, in the- in the very early days, uh, uh, Larry and Sergey had the PageRank paper, and Jon Kleinberg had this, uh, hubs and authorities model, which says the web is made out of these, uh, hubs, which will be my page of cool links about dogs or whatever-

    15. LFLex Fridman51:44

        Mm-hmm.

    16. PNPeter Norvig51:44

        ... and people would just list links. Uh, and then there'd be authorities, which were the ones, uh, that- page about dogs that most people link to. That doesn't happen anymore. People don't bother to say, "My page of cool links-"

    17. LFLex Fridman51:57

        Mm-hmm.

    18. PNPeter Norvig51:57

        ... uh, 'cause we took over that function, right? So- so, uh, we changed th- the way that worked.

    19. LFLex Fridman52:03

        Did you imagine back then that the internet would be as massively vibrant as it is today? I mean, it was already growing quickly, but it's just another... I- I don't know if you've ever-

    20. PNPeter Norvig52:14

        You know-

    21. LFLex Fridman52:14

        ... if you- today, if you sit back and (laughs) just look at the internet with wonder, the amount of content that's just constantly being created, constantly being shared-

    22. PNPeter Norvig52:23

        Yeah.

    23. LFLex Fridman52:23

        ... and deployed.

    24. PNPeter Norvig52:24

        Yeah. It- it's, uh, it's always been surprising to me.

    25. LFLex Fridman52:27

        (laughs)

    26. PNPeter Norvig52:27

        I- I guess I'm not very good at- at, uh-

    27. LFLex Fridman52:29

        Predicting the future?

    28. PNPeter Norvig52:29

        ... predicting the future.

    29. LFLex Fridman52:30

        Okay. (laughs)

    30. PNPeter Norvig52:31

        Uh, and I remember, you know, being a graduate student in- in 1980 or so, and, uh, you know, we had the ARPANET, and then there was this, uh, proposal to, uh, commercialize it-
16. 53:23 – 1:02:57

    Human-level intelligence, assistants, love, tests, risks—and what to work on next

    1. LFLex Fridman53:23

       So, in terms of predicting the future, what do you think it takes to build a system that approaches human-level intelligence? Y- you've talked about, of course, that w- you know, we shouldn't be so obsessed about creating human-level intelligence, just cr- create systems that are very useful for humans. But what do you think it takes-

    2. PNPeter Norvig53:41

       Yeah.

    3. LFLex Fridman53:42

       ... to, uh, to- uh, to, uh, yeah, approach that level?

    4. PNPeter Norvig53:45

       Right. So, certainly, I- I don't think human-level intelligence is one thing, right? So, I think-

    5. LFLex Fridman53:50

       Right.

    6. PNPeter Norvig53:50

       ... there's lots of different tasks, lots of different capabilities. I also don't think, uh, that should be the goal, right? So, I, you know, I wouldn't want to create a, uh, calculator that could do multiplication at human level, right?

    7. LFLex Fridman54:03

       Right.

    8. PNPeter Norvig54:04

       That would- that would be a step backwards. And so for many things, we should be aiming far beyond human level. Uh, for other things, uh, maybe human level is a good level to aim at. Uh, and for others we'd say, "Well, let's not bother doing this 'cause we al- we already have humans who can take on those tasks." So, as you say, I like to focus on, uh, what- what's a useful tool?

    9. LFLex Fridman54:26

       Right.

    10. PNPeter Norvig54:26

        And- and in some cases, being at human level is an important part of crossing that threshold to- to make the tool useful. So, we see in- in things like these, uh, uh, personal assistants now that you get either on your phone or on a- a speaker that sits on the table, uh, you wanna be able to have a conversation with those.

    11. LFLex Fridman54:46

        Mm-hmm.

    12. PNPeter Norvig54:47

        And- and I think as an industry, we haven't quite figured out what the right model is for what these things can do.

    13. LFLex Fridman54:54

        Right.

    14. PNPeter Norvig54:54

        Uh, and we're aiming towards, well, you just have a conversation with them the way you can with a person.

    15. LFLex Fridman54:59

        Right.

    16. PNPeter Norvig54:59

        Uh, but we haven't delivered on that model yet, right? So, you can ask it, "What's the weather?" You can ask it-... play some nice songs, uh, and, uh, you know, five or six other things and then you run out of stuff that it can do.

    17. LFLex Fridman55:14

        In terms of, uh, deep meaningful connection, so you've mentioned the movie Her as one of your favorite AI movies. Do you think it's possible for a human being to fall in love with an AI system, AI assistant as you mentioned? So taking this big leap from, uh, "What's the weather?" to, you know-

    18. PNPeter Norvig55:29

        Yeah.

    19. LFLex Fridman55:29

        ... having a, a deep connection.

    20. PNPeter Norvig55:31

        Yeah. I, I think, uh, as people, that's what we love to do.

    21. LFLex Fridman55:35

        Yeah.

    22. PNPeter Norvig55:36

        And, uh, I was at a, uh, a showing of Her where we had a panel discussion and, and somebody asked me, uh, "What other movie do you think Her is similar to?" And my answer was, uh, "Life of Brian." Which, which is not a science fiction movie.

    23. LFLex Fridman55:52

        Mm-hmm.

    24. PNPeter Norvig55:53

        Uh, but both movies are about wanting to believe in something that's not necessarily real.

    25. LFLex Fridman55:59

        (laughs) Yeah. By the way, for people who don't know, it's Monty Python. Yeah.

    26. PNPeter Norvig56:03

        Yeah.

    27. LFLex Fridman56:03

        (laughs) That's been brilliantly put.

    28. PNPeter Norvig56:05

        Right?

    29. LFLex Fridman56:05

        But, so-

    30. PNPeter Norvig56:05

        So, uh, I mean, I think that's just the way we are. We, we want to trust, we want to believe, we want to fall in love and, uh, it doesn't necessarily take that much, right? So, uh, you know, my kids, uh, fell in love with their teddy bear.

Episode duration: 1:03:12