Jeff Hawkins: The Thousand Brains Theory of Intelligence

Jeff Hawkins (02:56.640)

love. This is the Lex Friedman podcast, and here is my conversation with Jeff Hawkins.

Jeff Hawkins (03:04.080)

We previously talked over two years ago. Do you think there's still neurons in your brain

Jeff Hawkins (03:09.760)

that remember that conversation, that remember me and got excited? Like there's a Lex neuron

Jeff Hawkins (03:15.600)

in your brain that just like finally has a purpose? I do remember our conversation. I

Jeff Hawkins (03:19.920)

have some memories of it, and I formed additional memories of you in the meantime. I wouldn't

Jeff Hawkins (03:26.480)

say there's a neuron or neurons in my brain that know you. There are synapses in my brain

Jeff Hawkins (03:31.360)

that have formed that reflect my knowledge of you and the model I have of you in the

Jeff Hawkins (03:36.800)

world. Whether the exact same synapses were formed two years ago, it's hard to say because

Jeff Hawkins (03:41.520)

these things come and go all the time. One of the things to know about brains is that

Jeff Hawkins (03:46.480)

when you think of things, you often erase the memory and rewrite it again. Yes, but I have

Jeff Hawkins (03:50.400)

a memory of you, and that's instantiated in synapses. There's a simpler way to think about

Jeff Hawkins (03:55.360)

it. We have a model of the world in your head, and that model is continually being updated.

Jeff Hawkins (04:02.400)

I updated this morning. You offered me this water. You said it was from the refrigerator.

Jeff Hawkins (04:07.200)

I remember these things. The model includes where we live, the places we know, the words,

Jeff Hawkins (04:12.960)

the objects in the world. It's a monstrous model, and it's constantly being updated.

Jeff Hawkins (04:17.600)

People are just part of that model. Our animals, our other physical objects, our events we've

Jeff Hawkins (04:23.360)

done. In my mind, it's no special place for the memories of humans. Obviously, I know a lot about

Jeff Hawkins (04:33.440)

my wife and friends and so on, but it's not like a special place for humans or over here.

Jeff Hawkins (04:41.920)

We model everything, and we model other people's behaviors too. If I said there's a copy of your

Jeff Hawkins (04:46.640)

mind in my mind, it's just because I've learned how humans behave, and I've learned some things

Jeff Hawkins (04:53.280)

about you, and that's part of my world model. Well, I just also mean the collective intelligence

Jeff Hawkins (05:00.560)

of the human species. I wonder if there's something fundamental to the brain that enables that,

Lex Fridman (05:08.480)

so modeling other humans with their ideas. You're actually jumping into a lot of big

Jeff Hawkins (05:13.600)

topics. Collective intelligence is a separate topic that a lot of people like to talk about.

Jeff Hawkins (05:17.440)

We could talk about that. That's interesting. We're not just individuals. We live in society

Lex Fridman (05:24.640)

and so on. From our research point of view, again, let's just talk. We studied the neocortex.

Jeff Hawkins (05:30.960)

It's a sheet of neural tissue. It's about 75% of your brain. It runs on this very repetitive

Jeff Hawkins (05:37.040)

algorithm. It's a very repetitive circuit. You can apply that algorithm to lots of different

Jeff Hawkins (05:44.000)

problems, but underneath, it's the same thing. We're just building this model. From our point

Jeff Hawkins (05:48.640)

of view, we wouldn't look for these special circuits someplace buried in your brain that

Jeff Hawkins (05:52.720)

might be related to understanding other humans. It's more like, how do we build a model of

Jeff Hawkins (05:58.640)

anything? How do we understand anything in the world? Humans are just another part of

Lex Fridman (06:02.080)

the things we understand. There's nothing to the brain that knows the

Jeff Hawkins (06:08.720)

emergent phenomena of collective intelligence. Well, I certainly know about that. I've heard

Lex Fridman (06:13.120)

the terms, I've read. No, but that's as an idea.

Jeff Hawkins (06:16.800)

Well, I think we have language, which is built into our brains. That's a key part of collective

Jeff Hawkins (06:21.920)

intelligence. There are some prior assumptions about the world we're going to live in. When

Lex Fridman (06:27.680)

we're born, we're not just a blank slate. Did we evolve to take advantage of those situations?

Jeff Hawkins (06:35.520)

Yes. Again, we study only part of the brain, the neocortex. There's other parts of the

Jeff Hawkins (06:39.040)

brain that are very much involved in societal interactions and human emotions and how we

Jeff Hawkins (06:45.600)

interact and even societal issues about how we interact with other people, when we support

Jeff Hawkins (06:53.280)

them, when we're greedy and things like that. Certainly, the brain is a great place

Jeff Hawkins (07:00.160)

where to study intelligence. I wonder if it's the fundamental atom of intelligence.

Jeff Hawkins (07:06.720)

Well, I would say it's absolutely in a central component, even if you believe in collective

Jeff Hawkins (07:12.000)

intelligence as, hey, that's where it's all happening. That's what we need to study,

Jeff Hawkins (07:16.320)

which I don't believe that, by the way. I think it's really important, but I don't think that

Jeff Hawkins (07:19.120)

is the thing. Even if you do believe that, then you have to understand how the brain works in

Jeff Hawkins (07:26.080)

doing that. It's more like we are intelligent individuals and together, we are much more

Jeff Hawkins (07:32.880)

magnified, our intelligence. We can do things that we couldn't do individually, but even as

Jeff Hawkins (07:37.200)

individuals, we're pretty damn smart and we can model things and understand the world and interact

Jeff Hawkins (07:42.000)

with it. To me, if you're going to start someplace, you need to start with the brain. Then you could

Jeff Hawkins (07:48.160)

say, well, how do brains interact with each other? What is the nature of language? How do we share

Jeff Hawkins (07:53.760)

models that I've learned something about the world, how do I share it with you? Which is really

Lex Fridman (07:56.960)

what sort of communal intelligence is. I know something, you know something. We've had different

Jeff Hawkins (08:02.800)

experiences in the world. I've learned something about brains. Maybe I can impart that to you. You've

Jeff Hawkins (08:06.800)

learned something about physics and you can impart that to me. Even just the epistemological

Jeff Hawkins (08:15.200)

question of, well, what is knowledge and how do you represent it in the brain? That's where it's

Jeff Hawkins (08:20.880)

going to reside for in our writings. It's obvious that human collaboration, human interaction

Lex Fridman (08:27.280)

is how we build societies. But some of the things you talk about and work on,

Jeff Hawkins (08:34.560)

some of those elements of what makes up an intelligent entity is there with a single person.

Jeff Hawkins (08:40.560)

Absolutely. I mean, we can't deny that the brain is the core element here. At least I think it's

Jeff Hawkins (08:47.040)

obvious. The brain is the core element in all theories of intelligence. It's where knowledge

Jeff Hawkins (08:51.920)

is represented. It's where knowledge is created. We interact, we share, we build upon each other's

Jeff Hawkins (08:58.080)

work. But without a brain, you'd have nothing. There would be no intelligence without brains.

Lex Fridman (09:03.920)

And so that's where we start. I got into this field because I just was curious as to who I am.

Lex Fridman (09:11.520)

How do I think? What's going on in my head when I'm thinking? What does it mean to know something?

Jeff Hawkins (09:16.560)

I can ask what it means for me to know something independent of how I learned it from you or from

Jeff Hawkins (09:21.200)

someone else or from society. What does it mean for me to know that I have a model of you in my

Jeff Hawkins (09:25.600)

head? What does it mean to know I know what this microphone does and how it works physically,

Jeff Hawkins (09:28.880)

even when I can't see right now? How do I know that? What does it mean? How the neurons do that

Jeff Hawkins (09:34.480)

at the fundamental level of neurons and synapses and so on? Those are really fascinating questions.

Lex Fridman (09:40.240)

And I'm happy to be just happy to understand those if I could.

Lex Fridman (09:44.400)

So in your new book, you talk about our brain, our mind as being made up of many brains.

Lex Fridman (09:55.920)

So the book is called A Thousand Brain Theory of Intelligence. What is the key idea of this book?

Jeff Hawkins (10:02.720)

The book has three sections and it has sort of maybe three big ideas. So the first section is

Jeff Hawkins (10:09.360)

all about what we've learned about the neocortex and that's the thousand brains theory. Just to

Jeff Hawkins (10:13.760)

complete the picture, the second section is all about AI and the third section is about the future

Jeff Hawkins (10:16.960)

of humanity. So the thousand brains theory, the big idea there, if I had to summarize into one

Jeff Hawkins (10:27.440)

big idea, is that we think of the brain, the neocortex as learning this model of the world.

Lex Fridman (10:33.440)

But what we learned is actually there's tens of thousands of independent modeling systems going

Jeff Hawkins (10:38.560)

on. And so each, we call the column in the cortex is about 150,000 of them, is a complete modeling

Jeff Hawkins (10:44.560)

system. So it's a collective intelligence in your head in some sense. So the thousand brains theory

Lex Fridman (10:50.320)

says, well, where do I have knowledge about this coffee cup or where's the model of this cell phone?

Jeff Hawkins (10:55.760)

It's not in one place. It's in thousands of separate models that are complimentary and

Jeff Hawkins (10:59.360)

they communicate with each other through voting. So this idea that we feel like we're one person,

Jeff Hawkins (11:04.240)

that's our experience. We can explain that. But reality, there's lots of these, it's almost like

Jeff Hawkins (11:09.920)

little brains, but they're sophisticated modeling systems, about 150,000 of them in each human

Jeff Hawkins (11:16.320)

brain. And that's a total different way of thinking about how the neocortex is structured

Jeff Hawkins (11:21.280)

than we or anyone else thought of even just five years ago. So you mentioned you started

Jeff Hawkins (11:27.280)

this journey just looking in the mirror and trying to understand who you are.

Lex Fridman (11:31.840)

So if you have many brains, who are you then? So it's interesting. We have a singular perception,

Jeff Hawkins (11:38.080)

right? We think, oh, I'm just here. I'm looking at you. But it's composed of all these things,

Jeff Hawkins (11:42.560)

like there's sounds and there's vision and there's touch and all kinds of inputs. Yeah,

Jeff Hawkins (11:48.080)

we have the singular perception. And what the thousand brain theory says, we have these models

Jeff Hawkins (11:51.920)

that are visual models. We have a lot of models that are auditory models, models that talk to

Jeff Hawkins (11:55.040)

models and so on, but they vote. And so these things in the cortex, you can think about these

Jeff Hawkins (12:01.200)

columns as like little grains of rice, 150,000 stacked next to each other. And each one is its

Jeff Hawkins (12:07.920)

own little modeling system, but they have these long range connections that go between them.

Lex Fridman (12:12.640)

And we call those voting connections or voting neurons. And so the different columns try to

Jeff Hawkins (12:20.080)

reach a consensus. Like, what am I looking at? Okay. Each one has some ambiguity, but they come

Jeff Hawkins (12:24.640)

to a consensus. Oh, there's a water bottle I'm looking at. We are only consciously able to

Jeff Hawkins (12:30.640)

perceive the voting. We're not able to perceive anything that goes on under the hood. So the

Lex Fridman (12:35.680)

voting is what we're aware of. The results of the vote.

Jeff Hawkins (12:39.920)

Yeah. Well, you can imagine it this way. We were just talking about eye movements a moment ago. So

Jeff Hawkins (12:44.560)

as I'm looking at something, my eyes are moving about three times a second. And with each movement,

Jeff Hawkins (12:49.120)

a completely new input is coming into the brain. It's not repetitive. It's not shifting it around.

Jeff Hawkins (12:54.480)

I'm totally unaware of it. I can't perceive it. But yet if I looked at the neurons in your brain,

Jeff Hawkins (12:58.960)

they're going on and off, on and off, on and off, on and off. But the voting neurons are not.

Jeff Hawkins (13:03.040)

The voting neurons are saying, we all agree, even though I'm looking at different parts of this,

Jeff Hawkins (13:06.240)

this is a water bottle right now. And that's not changing. And it's in some position and

Jeff Hawkins (13:11.360)

pose relative to me. So I have this perception of the water bottle about two feet away from me

Jeff Hawkins (13:15.520)

at a certain pose to me. That is not changing. That's the only part I'm aware of. I can't be

Jeff Hawkins (13:20.480)

aware of the fact that the inputs from the eyes are moving and changing and all this other tapping.

Lex Fridman (13:25.040)

So these long range connections are the part we can be conscious of. The individual activity in

Jeff Hawkins (13:31.200)

each column doesn't go anywhere else. It doesn't get shared anywhere else. There's no way to extract

Jeff Hawkins (13:37.840)

it and talk about it or extract it and even remember it to say, oh, yes, I can recall that.

Lex Fridman (13:45.200)

But these long range connections are the things that are accessible to language and to our,

Jeff Hawkins (13:50.160)

like the hippocampus, our memories, our short term memory systems and so on. So we're not aware of

Jeff Hawkins (13:56.640)

95% or maybe it's even 98% of what's going on in your brain. We're only aware of this sort of

Jeff Hawkins (14:02.960)

stable, somewhat stable voting outcome of all these things that are going on underneath the hood.

Lex Fridman (14:09.920)

So what would you say is the basic element in the thousand brains theory of intelligence

Jeff Hawkins (14:15.520)

of intelligence? Like what's the atom of intelligence when you think about it? Is it

Jeff Hawkins (14:21.040)

the individual brains and then what is a brain? Well, let's, let's, can we just talk about what

Jeff Hawkins (14:25.920)

intelligence is first and then, and then we can talk about the elements are. So in my, in my book,

Jeff Hawkins (14:31.440)

intelligence is the ability to learn a model of the world, to build internal to your head,

Jeff Hawkins (14:38.560)

a model that represents the structure of everything, you know, to know what this is a

Jeff Hawkins (14:42.720)

table and that's a coffee cup and this is a gooseneck lamp and all this to know these things.

Lex Fridman (14:47.200)

I have to have a model of it in my head. I just don't look at them and go, what is that?

Jeff Hawkins (14:50.720)

I already have internal representations of these things in my head and I had to learn them. I wasn't

Jeff Hawkins (14:55.680)

born of any of that knowledge. You were, you know, we have some lights in the room here. I, you know,

Jeff Hawkins (15:00.320)

that's not part of my evolutionary heritage, right? It's not in my genes. So, um, we have this

Jeff Hawkins (15:05.360)

incredible model and the model includes not only what things look like and feel like, but where

Jeff Hawkins (15:09.040)

they are relative to each other and how they behave. I've never picked up this water bottle

Jeff Hawkins (15:12.800)

before, but I know that if I took my hand on that blue thing and I turn it, it'll probably make a

Jeff Hawkins (15:16.240)

funny little sound as the little plastic things detach and then it'll rotate and it'll rotate a

Jeff Hawkins (15:20.720)

certain way and it'll come off. How do I know that? Because I have this model in my head.

Lex Fridman (15:24.480)

So the essence of intelligence as our ability to learn a model and the more sophisticated our

Jeff Hawkins (15:29.360)

model is, the smarter we are. Uh, not that there is a single intelligence, because you can know

Jeff Hawkins (15:34.880)

about, you know, a lot about things that I don't know. And I know about things you don't know.

Lex Fridman (15:37.680)

And we can both be very smart, but we both learned a model of the world through interacting with it.

Lex Fridman (15:42.080)

So that is the essence of intelligence. Then we can ask ourselves, what are the mechanisms in the

Jeff Hawkins (15:46.320)

brain that allow us to do that? And what are the mechanisms of learning, not just the neural

Jeff Hawkins (15:50.560)

mechanisms, what are the general process by how we learn a model? So that was a big insight for us.

Jeff Hawkins (15:54.800)

It's like, what are the, what is the actual things that, how do you learn this stuff? It turns out

Jeff Hawkins (15:59.840)

you have to learn it through movement. Um, you can't learn it just by that's how we learn. We

Jeff Hawkins (16:04.000)

learn through movement. We learn. Um, so you build up this model by observing things and

Jeff Hawkins (16:07.840)

touching them and moving them and walking around the world and so on. So either you move or the

Jeff Hawkins (16:11.680)

thing moves somehow. Yeah. You obviously can learn things just by reading a book, something like that.

Lex Fridman (16:16.960)

But think about if I were to say, oh, here's a new house. I want you to learn, you know,

Lex Fridman (16:21.120)

what do you do? You have to walk, you have to walk from room to the room. You have to open the doors,

Jeff Hawkins (16:25.440)

look around, see what's on the left, what's on the right. As you do this, you're building a model in

Jeff Hawkins (16:29.680)

your head. It's just, that's what you're doing. You can't just sit there and say, I'm going to grok

Jeff Hawkins (16:34.000)

the house. No. You know, or you can do it. You don't even want to just sit down and read some

Jeff Hawkins (16:37.360)

description of it, right? Yeah. You literally physically interact. The same with like a smartphone.

Jeff Hawkins (16:41.600)

If I'm going to learn a new app, I touch it and I move things around. I see what happens when I,

Jeff Hawkins (16:45.760)

when I do things with it. So that's the basic way we learn in the world. And by the way,

Jeff Hawkins (16:49.600)

when you say model, you mean something that can be used for prediction in the future.

Jeff Hawkins (16:54.720)

It's used for prediction and for behavior and planning. Right. And does a pretty good job

Jeff Hawkins (17:02.000)

doing so. Yeah. Here's the way to think about the model. A lot of people get hung up on this. So

Jeff Hawkins (17:08.320)

you can imagine an architect making a model of a house, right? So there's a physical model that's

Jeff Hawkins (17:13.360)

small. And why do they do that? Well, we do that because you can imagine what it would look like

Jeff Hawkins (17:17.520)

from different angles. Okay. Look from here, look from there. And you can also say, well,

Jeff Hawkins (17:21.200)

how, how far to get from the garage to the, to the swimming pool or something like that. Right. You

Jeff Hawkins (17:25.760)

can imagine looking at this and you can say, what would be the view from this location? So we build

Jeff Hawkins (17:29.120)

these physical models to let you imagine the future and imagine behaviors. Now we can take

Jeff Hawkins (17:34.720)

that same model and put it in a computer. So we now, today they'll build models of houses in a

Jeff Hawkins (17:39.840)

computer and they, and they do that using a set of, we'll come back to this term in a moment,

Jeff Hawkins (17:45.840)

reference frames, but basically you assign a reference frame for the palace and you assign

Jeff Hawkins (17:49.680)

different things for the house in different locations. And then the computer can generate

Jeff Hawkins (17:53.280)

an image and say, okay, this is what it looks like in this direction. The brain is doing something

Jeff Hawkins (17:56.960)

remarkably similar to this surprising. It's using reference frames. It's building these,

Jeff Hawkins (18:02.160)

it's similar to a model on a computer, which has the same benefits of building a physical model.

Jeff Hawkins (18:06.160)

It allows me to say, what would this thing look like if it was in this orientation? What would

Jeff Hawkins (18:10.480)

likely happen if I push this button? I've never pushed this button before, or how would I accomplish

Jeff Hawkins (18:15.120)

something? I want to, I want to convey a new idea I've learned. How would I do that? I can imagine

Jeff Hawkins (18:21.520)

in my head, well, I could talk about it. I could write a book. I could do some podcasts. I could,

Jeff Hawkins (18:28.400)

you know, maybe tell my neighbor, you know, and I can imagine the outcomes of all these things

Jeff Hawkins (18:32.720)

before I do any of them. That's what the model lets you do. It lets us plan the future and

Jeff Hawkins (18:36.880)

imagine the consequences of our actions. Prediction, you asked about prediction. Prediction

Jeff Hawkins (18:42.720)

is not the goal of the model. Prediction is an inherent property of it, and it's how the model

Jeff Hawkins (18:48.800)

corrects itself. So prediction is fundamental to intelligence. It's fundamental to building a model,

Lex Fridman (18:55.600)

and the model's intelligent. And let me go back and be very precise about this. Prediction,

Jeff Hawkins (19:00.000)

you can think of prediction two ways. One is like, hey, what would happen if I did this? That's a

Jeff Hawkins (19:03.520)

type of prediction. That's a key part of intelligence. But using prediction is like, oh,

Jeff Hawkins (19:07.920)

what's this water bottle going to feel like when I pick it up, you know? And that doesn't seem very

Jeff Hawkins (19:13.120)

intelligent. But one way to think about prediction is it's a way for us to learn where our model is

Jeff Hawkins (19:20.400)

wrong. So if I picked up this water bottle and it felt hot, I'd be very surprised. Or if I picked

Jeff Hawkins (19:26.080)

it up and it was very light, I'd be surprised. Or if I turned this top and I had to turn it the other

Jeff Hawkins (19:32.720)

way, I'd be surprised. And so all those might have a prediction like, okay, I'm going to do it. I'll

Jeff Hawkins (19:38.480)

drink some water. I'm okay. Okay, I do this. There it is. I feel opening, right? What if I had to turn

Jeff Hawkins (19:42.720)

it the other way? Or what if it's split in two? Then I say, oh my gosh, I misunderstood this. I

Jeff Hawkins (19:47.360)

didn't have the right model of this thing. My attention would be drawn to it. I'd be looking at

Jeff Hawkins (19:50.400)

it going, well, how the hell did that happen? Why did it open up that way? And I would update my

Jeff Hawkins (19:55.360)

model by doing it. Just by looking at it and playing around with that update and say, this is

Jeff Hawkins (19:58.400)

a new type of water bottle. So you're talking about sort of complicated things like a water bottle,

Jeff Hawkins (1:00:06.000)

perhaps. Okay. Yeah. Some people decide they don't want to live. Some people decide the desire to

Jeff Hawkins (1:00:13.120)

live is built in DNA, right? But I think what I'm trying to get to is in order to accomplish goals,

Jeff Hawkins (1:00:18.880)

it's useful to have the urgency of mortality. It's what the Stoics talked about,

Jeff Hawkins (1:00:23.200)

is meditating in your mortality. It might be a very useful thing to do to die and have the urgency

Jeff Hawkins (1:00:31.600)

of death and to realize that to conceive yourself as an entity that operates in this world that

Jeff Hawkins (1:00:38.400)

eventually will no longer be a part of this world and actually conceive of yourself as a conscious

Jeff Hawkins (1:00:43.280)

entity might be very useful for you to be a system that makes sense of the world. Otherwise,

Jeff Hawkins (1:00:49.760)

you might get lazy. Well, okay. We're going to build these machines, right? So we're talking

Jeff Hawkins (1:00:55.360)

about building AIs. But we're building the equivalent of the cortical columns.

Jeff Hawkins (1:01:03.360)

The neocortex. The neocortex. And the question is, where do they arrive at? Because we're not

Jeff Hawkins (1:01:11.120)

hard coding everything in. Well, in terms of if you build the neocortex equivalent,

Jeff Hawkins (1:01:17.360)

it will not have any of these desires or emotional states. Now, you can argue that

Jeff Hawkins (1:01:22.640)

that neocortex won't be useful unless I give it some agency, unless I give it some desire,

Lex Fridman (1:01:28.240)

unless I give it some motivation. Otherwise, you'll be just lazy and do nothing, right?

Jeff Hawkins (1:01:31.600)

You could argue that. But on its own, it's not going to do those things. It's just not going

Jeff Hawkins (1:01:37.040)

to sit there and say, I understand the world. Therefore, I care to live. No, it's not going

Jeff Hawkins (1:01:41.120)

to do that. It's just going to say, I understand the world. Why is that obvious to you? Do you think

Jeff Hawkins (1:01:46.240)

it's possible? Okay, let me ask it this way. Do you think it's possible it will at least assign to

Jeff Hawkins (1:01:52.960)

itself agency and perceive itself in this world as being a conscious entity as a useful way to

Jeff Hawkins (1:02:04.240)

operate in the world and to make sense of the world? I think an intelligent machine can be

Jeff Hawkins (1:02:08.640)

conscious, but that does not, again, imply any of these desires and goals that you're worried about.

Lex Fridman (1:02:18.160)

We can talk about what it means for a machine to be conscious.

Jeff Hawkins (1:02:20.560)

By the way, not worry about, but get excited about. It's not necessary that we should worry

Jeff Hawkins (1:02:24.640)

about it. I think there's a legitimate problem or not problem, a question asked,

Jeff Hawkins (1:02:29.200)

if you build this modeling system, what's it going to model? What's its desire? What's its

Jeff Hawkins (1:02:35.600)

goal? What are we applying it to? That's an interesting question. One thing, and it depends

Jeff Hawkins (1:02:42.720)

on the application, it's not something that inherent to the modeling system. It's something

Jeff Hawkins (1:02:46.800)

we apply to the modeling system in a particular way. If I wanted to make a really smart car,

Jeff Hawkins (1:02:52.320)

it would have to know about driving and cars and what's important in driving and cars.

Jeff Hawkins (1:02:58.320)

It's not going to figure that on its own. It's not going to sit there and say, I've understood

Jeff Hawkins (1:03:01.760)

the world and I've decided, no, no, no, no, we're going to have to tell it. We're going to have to

Jeff Hawkins (1:03:06.000)

say, so I imagine I make this car really smart. It learns about your driving habits. It learns

Jeff Hawkins (1:03:10.880)

about the world. Is it one day going to wake up and say, you know what? I'm tired of driving

Lex Fridman (1:03:17.760)

and doing what you want. I think I have better ideas about how to spend my time.

Jeff Hawkins (1:03:22.080)

Okay. No, it's not going to do that. Well, part of me is playing a little bit of devil's advocate,

Lex Fridman (1:03:26.160)

but part of me is also trying to think through this because I've studied cars quite a bit and

Jeff Hawkins (1:03:32.560)

I studied pedestrians and cyclists quite a bit. And there's part of me that thinks

Jeff Hawkins (1:03:38.560)

that there needs to be more intelligence than we realize in order to drive successfully.

Jeff Hawkins (1:03:46.160)

That game theory of human interaction seems to require some deep understanding of human nature

Jeff Hawkins (1:03:54.720)

that, okay. When a pedestrian crosses the street, there's some sense. They look at a car usually,

Lex Fridman (1:04:04.880)

and then they look away. There's some sense in which they say, I believe that you're not going

Jeff Hawkins (1:04:10.960)

to murder me. You don't have the guts to murder me. This is the little dance of pedestrian car

Jeff Hawkins (1:04:16.320)

interaction is saying, I'm going to look away and I'm going to put my life in your hands because

Jeff Hawkins (1:04:22.960)

I think you're human. You're not going to kill me. And then the car in order to successfully

Jeff Hawkins (1:04:28.240)

operate in like Manhattan streets has to say, no, no, no, no. I am going to kill you like a little

Jeff Hawkins (1:04:34.400)

bit. There's a little bit of this weird inkling of mutual murder. And that's a dance and somehow

Jeff Hawkins (1:04:40.480)

successfully operate through that. Do you think you were born of that? Did you learn that social

Jeff Hawkins (1:04:44.160)

interaction? I think it might have a lot of the same elements that you're talking about,

Jeff Hawkins (1:04:50.800)

which is we're leveraging things we were born with and applying them in the context that.

Jeff Hawkins (1:04:57.600)

All right. I would have said that that kind of interaction is learned because people in different

Jeff Hawkins (1:05:03.440)

cultures to have different interactions like that. If you cross the street in different cities and

Jeff Hawkins (1:05:06.880)

different parts of the world, they have different ways of interacting. I would say that's learned.

Lex Fridman (1:05:10.400)

And I would say an intelligent system can learn that too, but that does not lead. And the intelligent

Jeff Hawkins (1:05:15.360)

system can understand humans. It could understand that just like I can study an animal and learn

Jeff Hawkins (1:05:24.320)

something about that animal. I could study apes and learn something about their culture and so on.

Jeff Hawkins (1:05:28.640)

I don't have to be an ape to know that. I may not be completely, but I can understand something.

Lex Fridman (1:05:34.160)

So intelligent machine can model that. That's just part of the world. It's just part of the

Jeff Hawkins (1:05:37.360)

interactions. The question we're trying to get at, will the intelligent machine have its own personal

Jeff Hawkins (1:05:42.640)

agency that's beyond what we assign to it or its own personal goals or will it evolve and create

Jeff Hawkins (1:05:49.440)

these things? My confidence comes from understanding the mechanisms I'm talking about creating.

Jeff Hawkins (1:05:55.920)

This is not hand wavy stuff. It's down in the details. I'm going to build it. And I know what

Jeff Hawkins (1:06:00.880)

it's going to look like. And I know what it's going to behave. I know what the kind of things

Jeff Hawkins (1:06:03.760)

it could do and the kind of things it can't do. Just like when I build a computer, I know it's

Jeff Hawkins (1:06:08.000)

not going to, on its own, decide to put another register inside of it. It can't do that. No way.

Jeff Hawkins (1:06:13.440)

No matter what your software does, it can't add a register to the computer.

Lex Fridman (1:06:17.440)

So in this way, when we build AI systems, we have to make choices about how we embed them.

Lex Fridman (1:06:26.560)

So I talk about this in the book. I said intelligent system is not just the neocortex

Jeff Hawkins (1:06:30.880)

equivalent. You have to have that. But it has to have some kind of embodiment, physical or virtual.

Jeff Hawkins (1:06:36.800)

It has to have some sort of goals. It has to have some sort of ideas about dangers,

Jeff Hawkins (1:06:41.040)

about things it shouldn't do. We build in safeguards into systems. We have them in our

Jeff Hawkins (1:06:47.360)

bodies. We put them into cars. My car follows my directions until the day it sees I'm about to hit

Jeff Hawkins (1:06:53.440)

something and it ignores my directions and puts the brakes on. So we can build those things in.

Lex Fridman (1:06:58.240)

So that's a very interesting problem, how to build those in. I think my differing opinion about the

Jeff Hawkins (1:07:06.480)

risks of AI for most people is that people assume that somehow those things will disappear

Jeff Hawkins (1:07:11.440)

automatically and evolve. And intelligence itself begets that stuff or requires it.

Lex Fridman (1:07:17.600)

But it's not. Intelligence of the neocortex equipment doesn't require this. The neocortex

Jeff Hawkins (1:07:21.120)

equipment just says, I'm a learning system. Tell me what you want me to learn and ask me questions

Lex Fridman (1:07:26.880)

and I'll tell you the answers. And that, again, it's again like a map. A map has no intent about

Jeff Hawkins (1:07:33.920)

things, but you can use it to solve problems. Okay. So the building, engineering the neocortex

Lex Fridman (1:07:41.920)

in itself is just creating an intelligent prediction system.

Jeff Hawkins (1:07:45.840)

Modeling system. Sorry, modeling system. You can use it to then make predictions.

Lex Fridman (1:07:52.480)

But you can also put it inside a thing that's actually acting in this world.

Jeff Hawkins (1:07:56.800)

You have to put it inside something. Again, think of the map analogy, right? A map on its own doesn't

Jeff Hawkins (1:08:02.160)

do anything. It's just inert. It can learn, but it's just inert. So we have to embed it somehow

Jeff Hawkins (1:08:07.920)

in something to do something. So what's your intuition here? You had a conversation with

Jeff Hawkins (1:08:13.360)

Sam Harris recently that was sort of, you've had a bit of a disagreement and you're sticking on

Jeff Hawkins (1:08:20.320)

this point. Elon Musk, Stuart Russell kind of have us worry existential threats of AI.

Jeff Hawkins (1:08:29.520)

What's your intuition? Why, if we engineer increasingly intelligent neocortex type of system

Jeff Hawkins (1:08:36.720)

in the computer, why that shouldn't be a thing that we...

Jeff Hawkins (1:08:40.240)

It was interesting to use the word intuition and Sam Harris used the word intuition too.

Lex Fridman (1:08:44.240)

And we didn't use that intuition, that word. I immediately stopped and said,

Jeff Hawkins (1:08:47.840)

oh, that's the crux of the problem. He's using intuition. I'm not speaking about my intuition.

Jeff Hawkins (1:08:52.960)

I'm speaking about something I understand, something I'm going to build, something I am

Jeff Hawkins (1:08:56.080)

building, something I understand completely, or at least well enough to know what... I'm guessing,

Jeff Hawkins (1:09:01.840)

I know what this thing's going to do. And I think most people who are worried, they have trouble

Jeff Hawkins (1:09:08.160)

separating out... They don't have the knowledge or the understanding about what is intelligence,

Jeff Hawkins (1:09:13.280)

how's it manifest in the brain, how's it separate from these other functions in the brain.

Lex Fridman (1:09:17.280)

And so they imagine it's going to be human like or animal like. It's going to have the same sort of

Jeff Hawkins (1:09:21.680)

drives and emotions we have, but there's no reason for that. That's just because there's an unknown.

Jeff Hawkins (1:09:27.680)

If the unknown is like, oh my God, I don't know what this is going to do. We have to be careful.

Jeff Hawkins (1:09:31.520)

It could be like us, but really smarter. I'm saying, no, it won't be like us. It'll be really

Jeff Hawkins (1:09:35.680)

smarter, but it won't be like us at all. But I'm coming from that, not because I'm just guessing,

Jeff Hawkins (1:09:42.080)

I'm not using intuition. I'm basing it on like, okay, I understand this thing works. This is what

Jeff Hawkins (1:09:46.640)

it does. It makes money to you. Okay. But to push back, so I also disagree with the intuitions that

Jeff Hawkins (1:09:54.400)

Sam has, but I also disagree with what you just said, which, you know, what's a good analogy. So

Jeff Hawkins (1:10:02.080)

if you look at the Twitter algorithm in the early days, just recommender systems, you can understand

Lex Fridman (1:10:08.720)

how recommender systems work. What you can't understand in the early days is when you apply

Jeff Hawkins (1:10:14.640)

that recommender system at scale to thousands and millions of people, how that can change societies.

Jeff Hawkins (1:10:20.400)

Yeah. So the question is, yes, you're just saying this is how an engineer in your cortex works,

Lex Fridman (1:10:27.840)

but the, like when you have a very useful, uh, TikTok type of service that goes viral when your

Lex Fridman (1:10:35.040)

neural cortex goes viral and then millions of people start using it, can that destroy the world?

Jeff Hawkins (1:10:40.160)

No. Uh, well, first of all, this is back. One thing I want to say is that, um, AI is a dangerous

Jeff Hawkins (1:10:44.880)

technology. I don't, I'm not denying that. All technology is dangerous. Well, and AI,

Jeff Hawkins (1:10:48.880)

maybe particularly so. Okay. So, um, am I worried about it? Yeah, I'm totally worried about it.

Lex Fridman (1:10:54.400)

The thing where the narrow component we're talking about now is the existential risk of AI, right?

Jeff Hawkins (1:11:00.320)

Yeah. So I want to make that distinction because I think AI can be applied poorly. It can be applied

Jeff Hawkins (1:11:05.360)

in ways that, you know, people are going to understand the consequences of it. Um, these are

Jeff Hawkins (1:11:11.200)

all potentially very bad things, but they're not the AI system creating this existential risk on

Jeff Hawkins (1:11:18.400)

its own. And that's the only place that I disagree with other people. Right. So I, I think the

Jeff Hawkins (1:11:23.440)

existential risk thing is, um, humans are really damn good at surviving. So to kill off the human

Jeff Hawkins (1:11:29.360)

race, it'd be very, very difficult. Yes, but you can even, I'll go further. I don't think AI systems

Jeff Hawkins (1:11:36.000)

are ever going to try to, I don't think AI systems are ever going to like say, I'm going to ignore

Jeff Hawkins (1:11:40.720)

you. I'm going to do what I think is best. Um, I don't think that's going to happen, at least not

Jeff Hawkins (1:11:46.480)

in the way I'm talking about it. So you, the Twitter recommendation algorithm is an interesting

Jeff Hawkins (1:11:52.720)

example. Let's, let's use computers as an analogy again, right? I build a computer. It's a universal

Jeff Hawkins (1:11:59.600)

computing machine. I can't predict what people are going to use it for. They can build all kinds of

Jeff Hawkins (1:12:03.440)

things. They can, they can even create computer viruses. It's, you know, all kinds of stuff. So

Jeff Hawkins (1:12:09.040)

there's some unknown about its utility and about where it's going to go. But on the other hand,

Jeff Hawkins (1:12:13.360)

I pointed out that once I build a computer, it's not going to fundamentally change how it computes.

Jeff Hawkins (1:12:18.960)

It's like, I use the example of a register, which is a part, internal part of a computer. Um, you

Jeff Hawkins (1:12:23.520)

know, I say it can't just sit there because computers don't evolve. They don't replicate,

Jeff Hawkins (1:12:27.600)

they don't evolve. They don't, you know, the physical manifestation of the computer itself

Jeff Hawkins (1:12:31.120)

is not going to, there's certain things that can't do right. So we can break into things like things

Jeff Hawkins (1:12:36.320)

that are possible to happen. We can't predict and things that are just impossible to happen.

Jeff Hawkins (1:12:40.400)

Unless we go out of our way to make them happen, they're not going to happen unless somebody makes

Jeff Hawkins (1:12:44.320)

them happen. Yeah. So there's, there's a bunch of things to say. One is the physical aspect,

Jeff Hawkins (1:12:49.120)

which you're absolutely right. We have to build a thing for it to operate in the physical world

Lex Fridman (1:12:54.640)

and you can just stop building them. Uh, you know, the moment they're not doing the thing you want

Jeff Hawkins (1:13:01.280)

them to do or just change the design or change the design. The question is, I mean, there's,

Jeff Hawkins (1:13:05.760)

uh, it's possible in the physical world. This is probably longer term is you automate the building.

Jeff Hawkins (1:13:10.640)

It makes, it makes a lot of sense to automate the building. There's a lot of factories that

Jeff Hawkins (1:13:14.000)

are doing more and more and more automation to go from raw resources to the final product.

Jeff Hawkins (1:13:19.360)

It's possible to imagine that obviously much more efficient to keep, to create a factory that's

Jeff Hawkins (1:13:25.040)

creating robots that do something, uh, you know, that do something extremely useful for society.

Jeff Hawkins (1:13:30.880)

It could be a personal assistance. It could be, uh, it could, it could be your toaster, but a

Jeff Hawkins (1:13:35.840)

toaster as much as deeper knowledge of your culinary preferences. Yeah. And that could,

Jeff Hawkins (1:13:41.680)

uh, I think now you've hit on the right thing. The real thing we need to be worried about is

Jeff Hawkins (1:13:46.000)

self replication. Right. That is the thing that we're in the physical world or even the virtual

Jeff Hawkins (1:13:51.440)

world self replication because self replication is dangerous. It's probably more likely to be

Jeff Hawkins (1:13:56.560)

killed by a virus, you know, or a human hand veneered virus. Anybody can create a, you know,

Jeff Hawkins (1:14:01.760)

there's the technology is getting so almost anybody, but not anybody, but a lot of people

Jeff Hawkins (1:14:05.680)

could create a human engineered virus that could wipe out humanity. That is really dangerous. No

Jeff Hawkins (1:14:11.360)

intelligence required, just self replication. So, um, so we need to be careful about that.

Lex Fridman (1:14:18.480)

So when I think about, you know, AI, I'm not thinking about robots, building robots. Don't

Jeff Hawkins (1:14:24.240)

do that. Don't build a, you know, just, well, that's because you're interesting creating

Jeff Hawkins (1:14:28.320)

intelligence. It seems like self replication is a good way to make a lot of money. Well,

Jeff Hawkins (1:14:35.360)

fine. But so is, you know, maybe editing viruses is a good way too. I don't know. The point is,

Jeff Hawkins (1:14:41.120)

if as a society, when we want to look at existential risks, the existential risks we face

Jeff Hawkins (1:14:46.880)

that we can control almost all evolve around self replication. Yes. The question is, I don't see a

Jeff Hawkins (1:14:54.880)

good, uh, way to make a lot of money by engineering viruses and deploying them on the world. There

Jeff Hawkins (1:15:00.240)

could be, there could be applications that are useful, but let's separate out, let's separate out.

Jeff Hawkins (1:15:04.880)

I mean, you don't need to, you only need some, you know, terrorists who wants to do it. Cause

Jeff Hawkins (1:15:08.000)

it doesn't take a lot of money to make viruses. Um, let's just separate out what's risky and what's

Jeff Hawkins (1:15:13.520)

not risky. I'm arguing that the intelligence side of this equation is not risky. It's not risky at

Jeff Hawkins (1:15:18.560)

all. It's the self replication side of the equation that's risky. And I'm arguing that

Jeff Hawkins (1:15:23.520)

it's not risky. And I'm not dismissing that. I'm scared as hell. It's like the paperclip

Jeff Hawkins (1:15:28.880)

maximizer thing. Yeah. Those are often like talked about in the same conversation.

Jeff Hawkins (1:15:35.200)

Um, I think you're right. Like creating ultra intelligent, super intelligent systems

Jeff Hawkins (1:15:42.000)

is not necessarily coupled with a self replicating arbitrarily self replicating systems. Yeah. And

Jeff Hawkins (1:15:47.600)

you don't get evolution unless you're self replicating. Yeah. And so I think that's the gist

Jeff Hawkins (1:15:52.560)

of this argument that people have trouble separating those two out. They just think,

Jeff Hawkins (1:15:56.720)

Oh yeah, intelligence looks like us. And look how, look at the damage we've done to this planet,

Jeff Hawkins (1:16:00.960)

like how we've, you know, destroyed all these other species. Yeah. Well we replicate,

Jeff Hawkins (1:16:04.640)

which the 8 billion of us are 7 billion of us now. So, um, I think the idea is that the,

Jeff Hawkins (1:16:10.400)

the more intelligent we're able to build systems, the more tempting it becomes from a capitalist

Jeff Hawkins (1:16:17.120)

perspective of creating products, the more tempting it becomes to create self, uh, reproducing

Lex Fridman (1:16:21.920)

systems. All right. So let's say that's true. So does that mean we don't build intelligent systems?

Jeff Hawkins (1:16:26.720)

No, that means we regulate, we, we understand the risks. Uh, we regulate them. Uh, you know,

Jeff Hawkins (1:16:33.760)

look, there's a lot of things we could do as society, which have some sort of financial

Jeff Hawkins (1:16:37.200)

benefit to someone, which could do a lot of harm. And we have to learn how to regulate those things.

Jeff Hawkins (1:16:42.560)

We have to learn how to deal with those things. I will argue this. I would say the opposite. Like I

Jeff Hawkins (1:16:46.400)

would say having intelligent machines at our disposal will actually help us in the end more,

Jeff Hawkins (1:16:52.000)

because it'll help us understand these risks better. It'll help us mitigate these risks

Jeff Hawkins (1:16:55.040)

better. It might be ways of saying, oh, well, how do we solve climate change problems? You know,

Lex Fridman (1:16:59.040)

how do we do this? Or how do we do that? Um, that just like computers are dangerous in the hands of

Jeff Hawkins (1:17:05.600)

the wrong people, but they've been so great for so many other things. We live with those dangers.

Lex Fridman (1:17:09.840)

And I think we have to do the same with intelligent machines. We just, but we have to be

Jeff Hawkins (1:17:13.520)

constantly vigilant about this idea of a bad actors doing bad things with them and be,

Jeff Hawkins (1:17:19.360)

um, don't ever, ever create a self replicating system. Um, uh, and, and by the way, I don't even

Jeff Hawkins (1:17:25.440)

know if you could create a self replicating system that uses a factory. That's really dangerous.

Jeff Hawkins (1:17:30.320)

You know, nature's way of self replicating is so amazing. Um, you know, it doesn't require

Jeff Hawkins (1:17:36.000)

anything. It just, you know, the thing and resources and it goes right. Um, if I said to

Jeff Hawkins (1:17:41.680)

you, you know what we have to build, uh, our goal is to build a factory that can make that builds

Jeff Hawkins (1:17:46.880)

new factories and it has to end to end supply chain. It has to bind the resources, get the

Jeff Hawkins (1:17:54.000)

energy. I mean, that's really hard. It's, you know, no one's doing that in the next, you know,

Jeff Hawkins (1:18:00.000)

a hundred years. I've been extremely impressed by the efforts of Elon Musk and Tesla to try to do

Jeff Hawkins (1:18:06.400)

exactly that. Not, not from raw resource. Well, he actually, I think states the goal is to go from

Jeff Hawkins (1:18:12.720)

raw resource to the, uh, the final car in one factory. Yeah. That's the main goal. Of course,

Jeff Hawkins (1:18:19.440)

it's not currently possible, but they're taking huge leaps. Well, he's not the only one to do

Jeff Hawkins (1:18:23.600)

that. This has been a goal for many industries for a long, long time. Um, it's difficult to do.

Jeff Hawkins (1:18:28.720)

Well, a lot of people, what they do is instead they have like a million suppliers and then they

Jeff Hawkins (1:18:34.480)

like there's everybody's, they all co locate them and they, and they tie the systems together.

Jeff Hawkins (1:18:40.480)

It's a fundamental, I think that's, that also is not getting at the issue I was just talking about,

Jeff Hawkins (1:18:45.840)

um, which is self replication. It's, um, I mean, self replication means there's no

Jeff Hawkins (1:18:53.840)

entity involved other than the entity that's replicating. Um, right. And so if there are

Jeff Hawkins (1:18:58.800)

humans in this, in the loop, that's not really self replicating, right? It's unless somehow we're

Jeff Hawkins (1:19:04.400)

duped into doing it. But it's also, I don't necessarily

Lex Fridman (1:19:11.920)

agree with you because you've kind of mentioned that AI will not say no to us.

Jeff Hawkins (1:19:16.480)

I just think they will. Yeah. Yeah. So like, uh, I think it's a useful feature to build in. I'm

Jeff Hawkins (1:19:23.520)

just trying to like, uh, put myself in the mind of engineers to sometimes say no, you know, if you,

Jeff Hawkins (1:19:32.480)

I gave the example earlier, right? I gave the example of my car, right? My car turns the wheel

Jeff Hawkins (1:19:38.000)

and, and applies the accelerator and the brake as I say, until it decides there's something dangerous.

Jeff Hawkins (1:19:43.760)

Yes. And then it doesn't do that. Now that was something it didn't decide to do. It's something

Jeff Hawkins (1:19:50.240)

we programmed into the car. And so good. It was a good idea, right? The question again, isn't like

Jeff Hawkins (1:19:57.600)

if we create an intelligent system, will it ever ignore our commands? Of course it will. And

Jeff Hawkins (1:20:02.640)

sometimes is it going to do it because it came up, came up with its own goals that serve its purposes

Lex Fridman (1:20:08.560)

and it doesn't care about our purposes? No, I don't think that's going to happen.

Jeff Hawkins (1:20:12.480)

Okay. So let me ask you about these, uh, super intelligent cortical systems that we engineer

Lex Fridman (1:20:16.960)

and us humans, do you think, uh, with these entities operating out there in the world,

Lex Fridman (1:20:24.320)

what is the future most promising future look like? Is it us merging with them or is it us?

Jeff Hawkins (1:20:33.040)

Like, how do we keep us humans around when you have increasingly intelligent beings? Is it, uh,

Lex Fridman (1:20:38.880)

one of the dreams is to upload our minds in the digital space. So can we just

Jeff Hawkins (1:20:42.960)

give our minds to these, uh, systems so they can operate on them? Is there some kind of more

Jeff Hawkins (1:20:48.400)

interesting merger or is there more, more communication? I talked about all these

Jeff Hawkins (1:20:52.240)

scenarios and let me just walk through them. Sure. Um, the uploading the mind one. Yes. Extremely,

Jeff Hawkins (1:21:00.560)

really difficult to do. Like, like, we have no idea how to do this even remotely right now. Um,

Lex Fridman (1:21:06.480)

so it would be a very long way away, but I make the argument you wouldn't like the result.

Jeff Hawkins (1:21:11.280)

Um, and you wouldn't be pleased with the result. It's really not what you think it's going to be.

Jeff Hawkins (1:21:16.080)

Um, imagine I could upload your brain into a, into a computer right now. And now the computer

Jeff Hawkins (1:21:20.000)

sitting there going, Hey, I'm over here. Great. Get rid of that old bio person. I don't need them.

Jeff Hawkins (1:21:24.160)

You're still sitting here. Yeah. What are you going to do? No, no, that's not me. I'm here.

Jeff Hawkins (1:21:28.560)

Right. Are you going to feel satisfied then? Then you, but people imagine, look, I'm on my deathbed

Lex Fridman (1:21:33.600)

and I'm about to, you know, expire and I pushed the button and now I'm uploaded. But think about

Jeff Hawkins (1:21:38.240)

it a little differently. And, and so I don't think it's going to be a thing because people,

Jeff Hawkins (1:21:42.640)

by the time we're able to do this, if ever, because you have to replicate the entire body,

Jeff Hawkins (1:21:47.760)

not just the brain. It's, it's really, it's, I walked through the issues. It's really substantial.

Jeff Hawkins (1:21:52.240)

Um, do you have a sense of what makes us us? Is there, is there a shortcut to what can only save

Jeff Hawkins (1:21:59.520)

a certain part that makes us truly ours? No, but I think that machine would feel like it's you too.

Jeff Hawkins (1:22:04.720)

Right. Right. You have two people, just like I have a child, I have a child, right? I have two

Jeff Hawkins (1:22:08.400)

daughters. They're independent people. I created them. Well, partly. Yeah. And, um, uh, I don't,

Jeff Hawkins (1:22:16.160)

just because they're somewhat like me, I don't feel on them and they don't feel like I'm me. So

Jeff Hawkins (1:22:20.400)

if you split apart, you have two people. So we can tell them, come back to what, what makes,

Lex Fridman (1:22:24.080)

what consciousness do you want? We can talk about that, but we don't have like remote consciousness.

Jeff Hawkins (1:22:28.400)

I'm not sitting there going, Oh, I'm conscious of that. You know, I mean, that system of,

Lex Fridman (1:22:32.000)

so let's say, let's, let's stay on our topic. One was uploading a brand. Yep. It ain't gonna happen

Jeff Hawkins (1:22:38.480)

in a hundred years, maybe a thousand, but I don't think people are going to want to do it. The

Jeff Hawkins (1:22:44.080)

merging your mind with, uh, you know, the neural link thing, right? Like again, really, really

Jeff Hawkins (1:22:50.240)

difficult. It's, it's one thing to make progress, to control a prosthetic arm. It's another to have

Jeff Hawkins (1:22:54.720)

like a billion or several billion, you know, things and understanding what those signals

Jeff Hawkins (1:22:58.960)

mean. Like it's the one thing that like, okay, I can learn to think some patterns to make something

Jeff Hawkins (1:23:03.680)

happen. It's quite another thing to have a system, a computer, which actually knows exactly what

Jeff Hawkins (1:23:08.800)

cells it's talking to and how it's talking to them and interacting in a way like that. Very,

Jeff Hawkins (1:23:12.960)

very difficult. We're not getting anywhere closer to that. Um, interesting. Can I, can I, uh, can

Jeff Hawkins (1:23:18.160)

I ask a question here? What, so for me, what makes that merger very difficult practically in the next

Jeff Hawkins (1:23:24.880)

10, 20, 50 years is like literally the biology side of it, which is like, it's just hard to do

Jeff Hawkins (1:23:32.000)

that kind of surgery in a safe way. But your intuition is even the machine learning part of it,

Jeff Hawkins (1:23:38.640)

where the machine has to learn what the heck it's talking to. That's even hard. I think it's even

Jeff Hawkins (1:23:43.280)

harder. And it's not, it's, it's easy to do when you're talking about hundreds of signals. It's,

Jeff Hawkins (1:23:49.200)

it's a totally different thing to say, talking about billions of years. It's, it's a totally

Jeff Hawkins (1:23:53.840)

different thing to say, talking about billions of signals. So you don't think it's the raw,

Jeff Hawkins (1:23:57.440)

the it's a machine learning problem. You don't think it could be learned? Well, I'm just saying,

Jeff Hawkins (1:24:01.360)

no, I think you'd have to have detailed knowledge. You'd have to know exactly what the types of

Jeff Hawkins (1:24:05.440)

neurons you're connecting to. I mean, in the brain, there's these, there are all different

Jeff Hawkins (1:24:09.440)

types of things. It's not like a neural network. It's a very complex organism system up here. We

Jeff Hawkins (1:24:13.520)

talked about the grid cells or the place cells, you know, you have to know what kind of cells

Jeff Hawkins (1:24:16.640)

you're talking to and what they're doing and how their timing works and all, all this stuff,

Jeff Hawkins (1:24:20.640)

which you can't today. There's no way of doing that. Right. But I think it's, I think it's a,

Jeff Hawkins (1:24:24.960)

I think the problem you're right. That the biological aspect of like who wants to have

Jeff Hawkins (1:24:28.400)

a surgery and have this stuff inserted in your brain. That's a problem. But this is when we

Jeff Hawkins (1:24:32.640)

solve that problem. I think the, the information coding aspect is much worse. I think that's much

Jeff Hawkins (1:24:38.080)

worse. It's not like what they're doing today. Today. It's simple machine learning stuff

Jeff Hawkins (1:24:42.240)

because you're doing simple things. But if you want to merge your brain, like I'm thinking on

Jeff Hawkins (1:24:46.720)

the internet, I'm merged my brain with the machine and we're both doing, that's a totally different

Jeff Hawkins (1:24:51.440)

issue. That's interesting. I tend to think if the, okay. If you have a super clean signal

Jeff Hawkins (1:24:57.760)

from a bunch of neurons at the start, you don't know what those neurons are. I think that's much

Jeff Hawkins (1:25:04.400)

easier than the getting of the clean signal. I think if you think about today's machine learning,

Jeff Hawkins (1:25:10.880)

that's what you would conclude. Right. I'm thinking about what's going on in the brain

Lex Fridman (1:25:14.960)

and I don't reach that conclusion. So we'll have to see. Sure. But I don't think even, even then,

Jeff Hawkins (1:25:20.080)

I think this kind of a sad future. Like, you know, do I, do I have to like plug my brain

Jeff Hawkins (1:25:26.240)

into a computer? I'm still a biological organism. I assume I'm still going to die.

Lex Fridman (1:25:30.000)

So what have I achieved? Right. You know, what have I achieved? Oh, I disagree that we don't

Jeff Hawkins (1:25:36.640)

know what those are, but it seems like there could be a lot of different applications. It's

Jeff Hawkins (1:25:40.320)

like virtual reality is to expand your brain's capability to, to like, to read Wikipedia.

Lex Fridman (1:25:47.280)

Yeah. But, but fine. But, but you're still a biological organism.

Jeff Hawkins (1:25:50.080)

Yes. Yes. You know, you're still, you're still mortal. All right. So,

Lex Fridman (1:25:53.280)

so what are you accomplishing? You're making your life in this short period of time better. Right.

Jeff Hawkins (1:25:58.000)

Just like having the internet made our life better. Yeah. Yeah. Okay. So I think that's of,

Jeff Hawkins (1:26:03.760)

of, if I think about all the possible gains we can have here, that's a marginal one.

Jeff Hawkins (1:26:08.080)

It's an individual, Hey, I'm better, you know, I'm smarter. But you know, fine. I'm not against it.

Jeff Hawkins (1:26:15.280)

I just don't think it's earth changing. I, but, but it w so this is the true of the internet.

Jeff Hawkins (1:26:20.240)

When each of us individuals are smarter, we get a chance to then share our smartness.

Jeff Hawkins (1:26:24.800)

We get smarter and smarter together as like, as a collective, this is kind of like this

Jeff Hawkins (1:26:28.560)

ant colony. Why don't I just create an intelligent machine that doesn't have any of this biological

Jeff Hawkins (1:26:32.480)

nonsense that has all the same. It's everything except don't burden it with my brain. Yeah.

Jeff Hawkins (1:26:39.360)

Right. It has a brain. It is smart. It's like my child, but it's much, much smarter than me.

Lex Fridman (1:26:43.680)

So I have a choice between doing some implant, doing some hybrid, weird, you know, biological

Jeff Hawkins (1:26:48.320)

thing that bleeding and all these problems and limited by my brain or creating a system,

Jeff Hawkins (1:26:53.760)

which is super smart that I can talk to. Um, that helps me understand the world that can

Jeff Hawkins (1:26:58.240)

read the internet, you know, read Wikipedia and talk to me. I guess my, the open questions there

Jeff Hawkins (1:27:03.600)

are what does the men manifestation of super intelligence look like? So like, what are we

Jeff Hawkins (1:27:10.000)

going to, you, you talked about why do I want to merge with AI? Like what, what's the actual

Jeff Hawkins (1:27:14.880)

marginal benefit here? If I, if we have a super intelligent system, how will it make our life

Jeff Hawkins (1:27:23.680)

better? So let's, let's, that's a great question, but let's break it down to little pieces. All

Jeff Hawkins (1:27:28.240)

right. On the one hand, it can make our life better in lots of simple ways. You mentioned

Jeff Hawkins (1:27:32.400)

like a care robot or something that helps me do things. It cooks. I don't know what it does. Right.

Jeff Hawkins (1:27:36.960)

Little things like that. We have super better, smarter cars. We can have, you know, better agents

Jeff Hawkins (1:27:42.640)

aids helping us in our work environment and things like that. To me, that's like the easy stuff, the

Jeff Hawkins (1:27:47.360)

simple stuff in the beginning. Um, um, and so in the same way that computers made our lives better

Jeff Hawkins (1:27:53.200)

in ways, many, many ways, I will have those kinds of things. To me, the really exciting thing about AI

Jeff Hawkins (1:28:00.560)

is the sort of it's transcendent, transcendent quality in terms of humanity. We're still

Jeff Hawkins (1:28:05.760)

biological organisms. We're still stuck here on earth. It's going to be hard for us to live

Jeff Hawkins (1:28:09.760)

anywhere else. Uh, I don't think you and I are going to want to live on Mars anytime soon. Um,

Jeff Hawkins (1:28:14.960)

um, and, um, and we're flawed, you know, we may end up destroying ourselves. It's totally possible.

Jeff Hawkins (1:28:23.440)

Uh, we, if not completely, we could destroy our civilizations. You know, it's this face the fact

Jeff Hawkins (1:28:28.320)

we have issues here, but we can create intelligent machines that can help us in various ways. For

Jeff Hawkins (1:28:33.680)

example, one example I gave, and that sounds a little sci fi, but I believe this. If we really

Jeff Hawkins (1:28:38.160)

wanted to live on Mars, we'd have to have intelligent systems that go there and build

Jeff Hawkins (1:28:42.560)

the habitat for us, not humans. Humans are never going to do this. It's just too hard. Um, but could

Jeff Hawkins (1:28:48.240)

we have a thousand or 10,000, you know, engineer workers up there doing this stuff, building things,

Jeff Hawkins (1:28:53.120)

terraforming Mars? Sure. Maybe we can move Mars. But then if we want to, if we want to go around

Jeff Hawkins (1:28:57.840)

the universe, should I send my children around the universe or should I send some intelligent machine,

Jeff Hawkins (1:29:02.400)

which is like a child that represents me and understands our needs here on earth that could

Jeff Hawkins (1:29:07.520)

travel through space. Um, so it's sort of, it, in some sense, intelligence allows us to transcend

Jeff Hawkins (1:29:13.280)

our, the limitations of our biology, uh, with, and, and don't think of it as a negative thing.

Jeff Hawkins (1:29:19.920)

It's in some sense, my children transcend my, the, my biology too, cause they, they live beyond me.

Jeff Hawkins (1:29:26.000)

Yeah. Um, and we impart, they represent me and they also have their own knowledge and I can

Jeff Hawkins (1:29:30.480)

impart knowledge to them. So intelligent machines will be like that too, but not limited like us.

Jeff Hawkins (1:29:34.400)

I mean, but the question is, um, there's so many ways that transcendence can happen

Lex Fridman (1:29:40.320)

and the merger with AI and humans is one of those ways. So you said intelligent,

Jeff Hawkins (1:29:46.960)

basically beings or systems propagating throughout the universe, representing us humans.

Jeff Hawkins (1:29:53.280)

They represent us humans in the sense they represent our knowledge and our history,

Jeff Hawkins (1:29:56.560)

not us individually. Right. Right. But I mean, the question is, is it just a database

Lex Fridman (1:30:04.960)

with, uh, with the really damn good, uh, model of the world?

Lex Fridman (1:30:09.600)

It's conscious, it's conscious just like us. Okay. But just different?

Jeff Hawkins (1:30:12.800)

They're different. Uh, just like my children are different. They're like me, but they're

Lex Fridman (1:30:16.560)

different. Um, these are more different. I guess maybe I've already, I kind of,

Lex Fridman (1:30:22.560)

I take a very broad view of our life here on earth. I say, you know, why are we living here?

Jeff Hawkins (1:30:28.320)

Are we just living because we live? Is it, are we surviving because we can survive? Are we fighting

Jeff Hawkins (1:30:32.960)

just because we want to just keep going? What's the point of it? Right. So to me, the point,

Jeff Hawkins (1:30:38.880)

if I asked myself, what's the point of life is what's transcends that ephemeral sort of biological

Jeff Hawkins (1:30:46.000)

experience is to me, this is my answer is the acquisition of knowledge to understand more about

Jeff Hawkins (1:30:53.520)

the universe, uh, and to explore. And that's partly to learn more. Right. Um, I don't view it as

Jeff Hawkins (1:31:01.920)

a terrible thing. If the ultimate outcome of humanity is we create systems that are intelligent

Jeff Hawkins (1:31:09.040)

that are offspring, but they're not like us at all. And we stay, we stay here and live on earth

Jeff Hawkins (1:31:13.680)

as long as we can, which won't be forever, but as long as we can and, but that would be a great

Jeff Hawkins (1:31:20.960)

thing to do. It's not a, it's not like a negative thing. Well, would, uh, you be okay then if, uh,

Jeff Hawkins (1:31:29.760)

the human species vanishes, but our knowledge is preserved and keeps being expanded by intelligence

Jeff Hawkins (1:31:37.440)

systems. I want our knowledge to be preserved and expanded. Yeah. Am I okay with humans dying? No,

Jeff Hawkins (1:31:44.960)

I don't want that to happen. But if it, if it does happen, what if we were sitting here and this is

Jeff Hawkins (1:31:50.400)

all the real, the last two people on earth and we're saying, Lex, we blew it. It's all over.

Jeff Hawkins (1:31:53.920)

Right. Wouldn't I feel better if I knew that our knowledge was preserved and that we had agents

Jeff Hawkins (1:32:00.080)

that knew about that, that were trans, you know, there were that left earth. I wouldn't want that.

Jeff Hawkins (1:32:04.800)

Mm. It's better than not having that, you know, I make the analogy of like, you know,

Jeff Hawkins (1:32:08.240)

the dinosaurs, the poor dinosaurs, they live for, you know, tens of millions of years.

Jeff Hawkins (1:32:11.520)

They raised their kids. They, you know, they, they fought to survive. They were hungry. They,

Jeff Hawkins (1:32:15.840)

they did everything we do. And then they're all gone. Yeah. Like, you know, and, and if we didn't

Jeff Hawkins (1:32:20.960)

discover their bones, nobody would ever know that they ever existed. Right. Do we want to be like

Jeff Hawkins (1:32:27.600)

that? I don't want to be like that. There's a sad aspect to it. And it's kind of, it's jarring to

Jeff Hawkins (1:32:32.720)

think about that. It's possible that a human like intelligence civilization has previously existed

Jeff Hawkins (1:32:39.600)

on earth. The reason I say this is like, it is jarring to think that we would not, if they went

Jeff Hawkins (1:32:46.640)

extinct, we wouldn't be able to find evidence of them after a sufficient amount of time. Of course,

Jeff Hawkins (1:32:53.040)

there's like, like basically humans, like if we destroy ourselves now, the human civilization

Jeff Hawkins (1:32:58.800)

destroyed ourselves. Now, after a sufficient amount of time, we would not be, we'd find evidence of

Jeff Hawkins (1:33:03.280)

the dinosaurs would not find evidence of humans. Yeah. That's kind of an odd thing to think about.

Jeff Hawkins (1:33:08.640)

Although I'm not sure if we have enough knowledge about species going back for billions of years,

Lex Fridman (1:33:14.880)

but we could, we could, we might be able to eliminate that possibility, but it's an interesting

Jeff Hawkins (1:33:18.960)

question. Of course, this is a similar question to, you know, there were lots of intelligent

Jeff Hawkins (1:33:23.200)

species throughout our galaxy that have all disappeared. That's super sad that they're,

Jeff Hawkins (1:33:30.320)

exactly that there may have been much more intelligent alien civilizations in our galaxy

Jeff Hawkins (1:33:36.000)

that are no longer there. Yeah. You actually talked about this, that humans might destroy

Jeff Hawkins (1:33:42.480)

ourselves and how we might preserve our knowledge and advertise that knowledge to other. Advertise

Jeff Hawkins (1:33:53.920)

is a funny word to use. From a PR perspective. There's no financial gain in this.

Jeff Hawkins (1:34:00.720)

You know, like make it like from a tourism perspective, make it interesting. Can you

Jeff Hawkins (1:34:04.480)

describe how you think about this problem? Well, there's a couple things. I broke it down

Jeff Hawkins (1:34:07.600)

into two parts, actually three parts. One is, you know, there's a lot of things we know that,

Lex Fridman (1:34:14.960)

what if, what if we were, what if we ended, what if our civilization collapsed? Yeah. I'm not

Jeff Hawkins (1:34:19.280)

talking tomorrow. Yeah. We could be a thousand years from now, like, so, you know, we don't

Jeff Hawkins (1:34:22.400)

really know, but, but historically it would be likely at some point. Time flies when you're

Jeff Hawkins (1:34:26.720)

having fun. Yeah. That's a good way to put it. You know, could we, and then intelligent life

Jeff Hawkins (1:34:33.200)

evolved again on this planet. Wouldn't they want to know a lot about us and what we knew? Wouldn't

Jeff Hawkins (1:34:37.680)

they wouldn't be able to ask us questions? So one very simple thing I said, how would we archive

Lex Fridman (1:34:42.000)

what we know? That was a very simple idea. I said, you know what, that wouldn't be that hard to put

Jeff Hawkins (1:34:46.080)

a few satellites, you know, going around the sun and we'd upload Wikipedia every day and that kind

Jeff Hawkins (1:34:51.200)

of thing. So, you know, if we end up killing ourselves, well, it's up there and the next intelligent

Jeff Hawkins (1:34:55.600)

species will find it and learn something. They would like that. They would appreciate that.

Jeff Hawkins (1:34:58.720)

Um, uh, so that's one thing. The next thing I said, well, what if, you know, how outside,

Jeff Hawkins (1:35:05.360)

outside of our solar system, we have the SETI program. We're looking for these intelligent

Jeff Hawkins (1:35:09.680)

signals from everybody. And if you do a little bit of math, which I did in the book, uh, and

Jeff Hawkins (1:35:14.320)

you say, well, what if intelligent species only live for 10,000 years before, you know,

Jeff Hawkins (1:35:18.800)

technologically intelligent species, like ones are really able to do the stuff we're just starting

Jeff Hawkins (1:35:22.560)

to be able to do. Um, well, the chances are we wouldn't be able to see any of them because they

Jeff Hawkins (1:35:26.800)

would have all been disappeared by now. Um, they would, they've lived for 10,000 years and now

Jeff Hawkins (1:35:31.040)

they're gone. And so we're not going to find these signals being sent from these people because, um,

Lex Fridman (1:35:36.080)

but I said, what kind of signal could you create that would last a million years or a billion years

Jeff Hawkins (1:35:41.120)

that someone would say, dammit, someone smart lived there that we know that that would be a

Jeff Hawkins (1:35:46.080)

life changing event for us to figure that out. Well, what we're looking for today in the study

Jeff Hawkins (1:35:49.760)

program, isn't that we're looking for very coded signals in some sense. Um, and so I asked myself,

Lex Fridman (1:35:54.560)

what would be a different type of signal one could create? Um, I've always thought about

Jeff Hawkins (1:35:58.160)

this throughout my life. And in the book, I gave one, one possible suggestion, which was, um, uh,

Jeff Hawkins (1:36:04.480)

we now detect planets going around other, other suns, uh, other stars, uh, excuse me. And we do

Jeff Hawkins (1:36:11.040)

that by seeing this, the, the slight dimming of the light as the planets move in front of them.

Jeff Hawkins (1:36:14.800)

That's how, uh, we detect, uh, planets elsewhere in our galaxy. Um, what if we created something

Jeff Hawkins (1:36:21.040)

like that, that just rotated around our, our, our, around the sun and it blocked out a little

Jeff Hawkins (1:36:26.480)

bit of light in a particular pattern that someone said, Hey, that's not a planet. That is a sign

Jeff Hawkins (1:36:31.760)

that someone was once there. You can say, what if it's beating up pie, you know, three point,

Jeff Hawkins (1:36:36.000)

whatever. Um, so I did it from a distance. Broadly broadcast takes no continue activation on our

Jeff Hawkins (1:36:44.960)

part. This is the key, right? No one has to be senior running a computer and supplying it with

Jeff Hawkins (1:36:48.320)

power. It just goes on. So we go, it's continuous. And, and I argued that part of the study program

Jeff Hawkins (1:36:55.200)

should be looking for signals like that. And to look for signals like that, you ought to figure

Jeff Hawkins (1:36:58.880)

out what the, how would we create a signal? Like what would we create that would be like that,

Jeff Hawkins (1:37:03.440)

that would persist for millions of years that would be broadcast broadly. You could see from

Jeff Hawkins (1:37:07.680)

a distance that was unequivocal, came from an intelligent species. And so I gave that one

Jeff Hawkins (1:37:13.760)

example. Um, cause they don't know what I know of actually. And then, and then finally, right.

Jeff Hawkins (1:37:19.760)

If, if our, ultimately our solar system will die at some point in time, you know, how do we go

Jeff Hawkins (1:37:26.640)

beyond that? And I think it's possible if it all possible, we'll have to create intelligent machines

Jeff Hawkins (1:37:31.600)

that travel throughout the, throughout the solar system or the galaxy. And I don't think that's

Jeff Hawkins (1:37:36.880)

going to be humans. I don't think it's going to be biological organisms. So these are just things to

Jeff Hawkins (1:37:41.040)

think about, you know, like, what's the old, you know, I don't want to be like the dinosaur. I

Jeff Hawkins (1:37:44.560)

don't want to just live in, okay, that was it. We're done. You know, well, there is a kind of

Jeff Hawkins (1:37:48.400)

presumption that we're going to live forever, which, uh, I think it is a bit sad to imagine

Jeff Hawkins (1:37:55.280)

that the message we send as, as you talk about is that we were once here instead of we are here.

Jeff Hawkins (1:38:03.680)

Well, it could be, we are still here. Uh, but it's more of a, it's more of an insurance policy

Jeff Hawkins (1:38:09.520)

in case we're not here, you know? Well, I don't know, but there is something I think about,

Lex Fridman (1:38:16.080)

we as humans don't often think about this, but it's like, like whenever I, um,

Jeff Hawkins (1:38:23.680)

record a video, I've done this a couple of times in my life. I've recorded a video for my future

Jeff Hawkins (1:38:28.160)

self, just for personal, just for fun. And it's always just fascinating to think about

Jeff Hawkins (1:38:34.400)

that preserving yourself for future civilizations. For me, it was preserving myself for a future me,

Lex Fridman (1:38:41.600)

but that's a little, that's a little fun example of archival.

Jeff Hawkins (1:38:46.160)

Well, these podcasts are, are, are preserving you and I in a way. Yeah. For future,

Jeff Hawkins (1:38:51.280)

hopefully well after we're gone. But you don't often, we're sitting here talking about this.

Jeff Hawkins (1:38:56.640)

You are not thinking about the fact that you and I are going to die and there'll be like 10 years

Jeff Hawkins (1:39:02.800)

after somebody watching this and we're still alive. You know, in some sense I do. I'm here

Jeff Hawkins (1:39:09.440)

cause I want to talk about ideas and these ideas transcend me and they transcend this time and, and

Jeff Hawkins (1:39:16.720)

on our planet. Um, we're talking here about ideas that could be around a thousand years from now.

Jeff Hawkins (1:39:23.520)

Or a million years from now. I, when I wrote my book, I had an audience in mind and one of the

Jeff Hawkins (1:39:29.360)

clearest audiences was aliens. No. Were people reading this a hundred years from now? Yes.

Lex Fridman (1:39:35.200)

I said to myself, how do I make this book relevant to someone reading this a hundred years from now?

Lex Fridman (1:39:39.360)

What would they want to know that we were thinking back then? What would make it like,

Jeff Hawkins (1:39:44.160)

that was an interesting, it's still an interesting book. I'm not sure I can achieve that, but that was

Lex Fridman (1:39:49.360)

how I thought about it because these ideas, like especially in the third part of the book, the ones

Jeff Hawkins (1:39:53.440)

we were just talking about, you know, these crazy, sounds like crazy ideas about, you know,

Jeff Hawkins (1:39:56.960)

storing our knowledge and, and, you know, merging our brains with computers and, and sending, you

Jeff Hawkins (1:40:01.680)

know, our machines out into space. It's not going to happen in my lifetime. Um, and they may not

Lex Fridman (1:40:07.360)

have been happening in the next hundred years. They may not happen for a thousand years. Who knows?

Jeff Hawkins (1:40:10.640)

Uh, but we have the unique opportunity right now. We, you, me, and other people in the world,

Jeff Hawkins (1:40:17.440)

right now, we, you, me, and other people like this, um, to sort of at least propose the agenda,

Jeff Hawkins (1:40:24.640)

um, that might impact the future like that. That's a fascinating way to think, uh, both like

Jeff Hawkins (1:40:29.840)

writing or creating, try to make, try to create ideas, try to create things that, uh, hold up

Jeff Hawkins (1:40:38.400)

in time. Yeah. You know, when understanding how the brain works, we're going to figure that out

Jeff Hawkins (1:40:42.240)

once. That's it. It's going to be figured out once. And after that, that's the answer. And

Jeff Hawkins (1:40:46.720)

people will, people will study that thousands of years now. We still, we still, you know,

Jeff Hawkins (1:40:51.600)

venerate Newton and, and Einstein and, um, and, you know, because, because ideas are exciting,

Jeff Hawkins (1:40:59.040)

even well into the future. Well, the interesting thing is like big ideas, even if they're wrong,

Jeff Hawkins (1:41:05.520)

are still useful. Like, yeah, especially if they're not completely wrong, right? Right.

Jeff Hawkins (1:41:12.800)

Newton's laws are not wrong. They're just Einstein's they're better. Um, so yeah, I mean,

Lex Fridman (1:41:19.840)

but we're talking with Newton and Einstein, we're talking about physics. I wonder if we'll ever

Jeff Hawkins (1:41:23.440)

achieve that kind of clarity, but understanding, um, like complex systems and the, this particular

Jeff Hawkins (1:41:30.880)

manifestation of complex systems, which is the human brain. I'm totally optimistic. We can do

Jeff Hawkins (1:41:36.160)

that. I mean, we're making progress at it. I don't see any reasons why we can't completely. I mean,

Jeff Hawkins (1:41:41.440)

completely understand in the sense, um, you know, we don't really completely understand what all

Jeff Hawkins (1:41:46.080)

the molecules in this water bottle are doing, but, you know, we have laws that sort of capture it

Jeff Hawkins (1:41:50.080)

pretty good. Um, and, uh, so we'll have that kind of understanding. I mean, it's not like you're

Jeff Hawkins (1:41:54.960)

gonna have to know what every neuron in your brain is doing. Um, but enough to, um, first of all,

Jeff Hawkins (1:42:00.880)

to build it. And second of all, to do, you know, do what physics does, which is like have, uh,

Jeff Hawkins (1:42:06.400)

concrete experiments where we can validate this is happening right now. Like it's not,

Jeff Hawkins (1:42:12.400)

this is not some future thing. Um, you know, I'm very optimistic about it because I know about our,

Jeff Hawkins (1:42:17.760)

our work and what we're doing. We'll have to prove it to people. Um, but, um,

Jeff Hawkins (1:42:24.480)

I, I consider myself a rational person and, um, you know, until fairly recently,

Jeff Hawkins (1:42:30.640)

I wouldn't have said that, but right now I'm, where I'm sitting right now, I'm saying, you know,

Jeff Hawkins (1:42:33.840)

we, we could, this is going to happen. There's no big obstacles to it. Um, we finally have a

Jeff Hawkins (1:42:39.200)

framework for understanding what's going on in the cortex and, um, and that's liberating. It's,

Jeff Hawkins (1:42:44.960)

it's like, Oh, it's happening. So I can't see why we wouldn't be able to understand it. I just can't.

Jeff Hawkins (1:42:50.880)

Okay. So, I mean, on that topic, let me ask you to play devil's advocate.

Jeff Hawkins (1:42:54.560)

Is it possible for you to imagine, look, look a hundred years from now and looking at your book,

Jeff Hawkins (1:43:02.320)

uh, in which ways might your ideas be wrong? Oh, I worry about this all the time. Um,

Lex Fridman (1:43:11.840)

yeah, it's still useful. Yeah. Yeah.

Jeff Hawkins (1:43:15.200)

Yeah. I think there's, you know, um, well I can, I can best relate it to like things I'm worried

Jeff Hawkins (1:43:24.800)

about right now. So we talked about this voting idea, right? It's happening. There's no question.

Jeff Hawkins (1:43:29.920)

It's happening, but it could be far more, um, um, there's, there's enough things I don't know about

Jeff Hawkins (1:43:36.480)

it that it might be working into ways differently than I'm thinking about the kind of what's voting,

Jeff Hawkins (1:43:41.520)

who's voting, you know, where are representations? I talked about, like, you have a thousand models

Jeff Hawkins (1:43:45.680)

of a coffee cup like that. That could turn out to be wrong. Um, because it may be, maybe there are a

Jeff Hawkins (1:43:52.320)

thousand models that are sub models, but not really a single model of the coffee cup. Um,

Jeff Hawkins (1:43:57.120)

I mean, there's things, these are all sort of on the edges, things that I present as like,

Jeff Hawkins (1:44:02.000)

Oh, it's so simple and clean. Well, it's not that it's always going to be more complex.

Jeff Hawkins (1:44:05.440)

And, um, and there's parts of the theory, which I don't understand the complexity well. So I think,

Jeff Hawkins (1:44:14.640)

I think the idea that this brain is a distributed modeling system is not controversial at all. Right.

Lex Fridman (1:44:19.440)

It's not, that's well understood by many people. The question then is,

Jeff Hawkins (1:44:22.720)

are each quarter of a column an independent modeling system? Um, I could be wrong about that.

Jeff Hawkins (1:44:29.040)

Um, I don't think so, but I worry about it. My intuition, not even thinking why you could

Jeff Hawkins (1:44:35.600)

be wrong is the same intuition I have about any sort of physicist, uh, like string theory

Jeff Hawkins (1:44:42.480)

that we as humans desire for a clean explanation. And, uh, a hundred years from now, uh,

Jeff Hawkins (1:44:50.160)

intelligent systems might look back at us and laugh at how we try to get rid of the whole mess

Jeff Hawkins (1:44:56.560)

by having simple explanation when the reality is it's way messier. And in fact, it's impossible

Jeff Hawkins (1:45:03.680)

to understand. You can only build it. It's like this idea of complex systems and cellular automata

Jeff Hawkins (1:45:08.960)

is you can only launch the thing. You cannot understand it. Yeah. I think that, you know,

Jeff Hawkins (1:45:13.840)

the history of science suggests that's not likely to occur. Um, the history of science suggests that

Jeff Hawkins (1:45:20.240)

as a theorist and we're theorists, you look for simple explanations, right? Fully knowing

Jeff Hawkins (1:45:25.920)

that whatever simple explanation you're going to come up with is not going to be completely correct.

Jeff Hawkins (1:45:30.640)

I mean, it can't be, I mean, it's just, it's just more complexity, but that's the role of theorists

Jeff Hawkins (1:45:35.840)

play. They, they sort of, they give you a framework on which you now can talk about a problem and

Jeff Hawkins (1:45:41.600)

figure out, okay, now we can start digging more details. The best frameworks stick around while

Jeff Hawkins (1:45:46.480)

the details change. You know, again, you know, the classic example is Newton and Einstein, right? You

Jeff Hawkins (1:45:53.440)

know, um, Newton's theories are still used. They're still valuable. They're still practical. They're

Jeff Hawkins (1:46:00.000)

not like wrong. It's just, they've been refined. Yeah. But that's in physics. It's not obvious,

Jeff Hawkins (1:46:05.120)

by the way, it's not obvious for physics either that the universe should be such that's amenable

Jeff Hawkins (1:46:10.400)

to these simple. But it's so far, it appears to be as far as we can tell. Um, yeah. I mean,

Lex Fridman (1:46:17.920)

but as far as we could tell, and, but it's also an open question whether the brain is amenable to

Jeff Hawkins (1:46:23.040)

such clean theories. That's the, uh, not the brain, but intelligence. Well, I, I, I don't know. I would

Jeff Hawkins (1:46:28.960)

take intelligence out of it. Just say, you know, um, well, okay. Um, the evidence we have suggests

Jeff Hawkins (1:46:37.120)

that the human brain is, is a, at the one time extremely messy and complex, but there's some

Jeff Hawkins (1:46:42.960)

parts that are very regular and structured. That's why we started the neocortex. It's extremely

Jeff Hawkins (1:46:48.240)

regular in its structure. Yeah. And unbelievably so. And then I mentioned earlier, the other thing is

Jeff Hawkins (1:46:53.440)

it's, it's universal abilities. It is so flexible to learn so many things. We don't, we haven't

Jeff Hawkins (1:47:00.560)

figured out what it can't learn yet. We don't know, but we haven't figured it out yet, but it

Jeff Hawkins (1:47:03.440)

can learn things that it never was evolved to learn. So those give us hope. Um, that's why I

Jeff Hawkins (1:47:09.040)

went into this field because I said, you know, this regular structure, it's doing this amazing

Jeff Hawkins (1:47:14.880)

number of things. There's gotta be some underlying principles that are, that are common and other,

Jeff Hawkins (1:47:19.680)

other scientists have come up with the same conclusions. Um, and so it's promising and,

Jeff Hawkins (1:47:25.600)

um, and that's, and whether the theories play out exactly this way or not, that is the role that

Jeff Hawkins (1:47:32.400)

theorists play. And so far it's worked out well, even though, you know, maybe, you know, we don't

Jeff Hawkins (1:47:38.080)

understand all the laws of physics, but so far it's been pretty damn useful. The ones we have

Jeff Hawkins (1:47:42.000)

are our theories are pretty useful. You mentioned that, uh, we should not necessarily be,

Jeff Hawkins (1:47:49.840)

at least to the degree that we are worried about the existential risks of artificial intelligence

Jeff Hawkins (1:47:55.200)

relative to, uh, human risks from human nature being existential risk.

Lex Fridman (1:48:02.720)

What aspect of human nature worries you the most in terms of the survival of the human species?

Jeff Hawkins (1:48:07.600)

I mean, I'm disappointed in humanity, humans. I mean, all of us, I'm one. So I'm disappointed

Jeff Hawkins (1:48:15.440)

myself too. Um, it's kind of a sad state. There's two things that disappoint me. One is

Lex Fridman (1:48:24.880)

how it's difficult for us to separate our rational component of ourselves from our evolutionary

Jeff Hawkins (1:48:30.640)

heritage, which is, you know, not always pretty, you know, um, uh, rape is a, is an evolutionary

Jeff Hawkins (1:48:38.800)

good strategy for reproduction. Murder can be at times too, you know, making other people miserable

Jeff Hawkins (1:48:45.760)

at times is a good strategy for reproduction. It's just, and it's just, and, and so now that

Jeff Hawkins (1:48:50.640)

we know that, and yet we have this sort of, you know, we, you and I can have this very rational

Jeff Hawkins (1:48:54.640)

discussion talking about, you know, intelligence and brains and life and so on. So many, it seems

Jeff Hawkins (1:48:59.680)

like it's so hard. It's just a big, big transition to get humans, all humans to, to make the

Jeff Hawkins (1:49:05.520)

transition from be like, let's pay no attention to all that ugly stuff over here. Let's just focus

Jeff Hawkins (1:49:11.360)

on the interesting. What's unique about humanity is our knowledge and our intellect. But the fact

Jeff Hawkins (1:49:16.720)

that we're striving is in itself amazing, right? The fact that we're able to overcome that part.

Lex Fridman (1:49:22.480)

And it seems like we are more and more becoming successful at overcoming that part. That is the

Jeff Hawkins (1:49:28.720)

optimistic view. And I agree with you, but I worry about it. I'm not saying I'm worrying about it. I

Jeff Hawkins (1:49:33.760)

think that was your question. I still worry about it. Yes. You know, we could be in tomorrow because

Jeff Hawkins (1:49:38.320)

some terrorists could get nuclear bombs and, you know, blow us all up. Who knows? Right. The other

Jeff Hawkins (1:49:43.200)

thing I think I'm disappointed is, and it's just, I understand it. It's, I guess you can't really

Jeff Hawkins (1:49:47.760)

be disappointed. It's just a fact is that we're so prone to false beliefs that we, you know, we have

Jeff Hawkins (1:49:53.120)

a model in our head, the things we can interact with directly, physical objects, people, that

Jeff Hawkins (1:50:00.080)

model is pretty good. And we can test it all the time, right? I touch something, I look at it,

Jeff Hawkins (1:50:04.800)

talk to you, see if my model is correct. But so much of what we know is stuff I can't directly

Jeff Hawkins (1:50:09.760)

interact with. I only know because someone told me about it. And so we're prone, inherently prone

Jeff Hawkins (1:50:16.560)

to having false beliefs because if I'm told something, how am I going to know it's right

Jeff Hawkins (1:50:20.560)

or wrong? Right. And so then we have the scientific process, which says we are inherently flawed.

Lex Fridman (1:50:26.800)

So the only way we can get closer to the truth is by looking for contrary evidence.

Jeff Hawkins (1:50:34.800)

Yeah. Like this conspiracy theory, this theory that scientists keep telling me about that the

Jeff Hawkins (1:50:41.600)

earth is round. As far as I can tell, when I look out, it looks pretty flat.

Jeff Hawkins (1:50:46.960)

Yeah. So, yeah, there is a tension, but it's also, I tend to believe that we haven't figured

Jeff Hawkins (1:50:55.440)

out most of this thing, right? Most of nature around us is a mystery. And so it...

Lex Fridman (1:51:02.240)

But that doesn't, does that worry you? I mean, it's like, oh, that's like a pleasure,

Jeff Hawkins (1:51:06.080)

more to figure out, right? Yeah. That's exciting. But I'm saying like

Jeff Hawkins (1:51:09.760)

there's going to be a lot of quote unquote, wrong ideas. I mean, I've been thinking a lot about

Jeff Hawkins (1:51:16.320)

engineering systems like social networks and so on. And I've been worried about censorship

Lex Fridman (1:51:21.760)

and thinking through all that kind of stuff, because there's a lot of wrong ideas. There's a

Jeff Hawkins (1:51:25.520)

lot of dangerous ideas, but then I also read a history, read history and see when you censor

Jeff Hawkins (1:51:33.360)

ideas that are wrong. Now this could be a small scale censorship, like a young grad student who

Jeff Hawkins (1:51:39.760)

comes up, who like raises their hand and says some crazy idea. A form of censorship could be,

Jeff Hawkins (1:51:46.320)

I shouldn't use the word censorship, but like de incentivize them from no, no, no, no,

Jeff Hawkins (1:51:52.000)

this is the way it's been done. Yeah. Yeah. You're a foolish kid. Don't

Lex Fridman (1:51:54.800)

think that's it. Yeah. You're foolish. So in some sense,

Jeff Hawkins (1:51:59.760)

those wrong ideas, most of the time end up being wrong, but sometimes end up being

Jeff Hawkins (1:52:05.520)

I agree with you. So I don't like the word censorship. Um, at the very end of the book, I,

Jeff Hawkins (1:52:11.280)

I ended up with a sort of a, um, a plea or a recommended force of action. Um, the best way I

Jeff Hawkins (1:52:20.000)

could, I know how to deal with this issue that you bring up is if everybody understood as part of

Jeff Hawkins (1:52:26.240)

your upbringing in life, something about how your brain works, that it builds a model of the world,

Jeff Hawkins (1:52:31.120)

uh, how it works, you know, how basically it builds that model of the world and that the model

Jeff Hawkins (1:52:34.960)

is not the real world. It's just a model and it's never going to reflect the entire world. And it

Jeff Hawkins (1:52:39.760)

can be wrong and it's easy to be wrong. And here's all the ways you can get a wrong model in your

Jeff Hawkins (1:52:44.320)

head. Right? It's not prescribed what's right or wrong. Just understand that process. If we all

Jeff Hawkins (1:52:50.960)

understood the processes and I got together and you say, I disagree with you, Jeff. And I said,

Jeff Hawkins (1:52:54.720)

Lex, I disagree with you that at least we understand that we're both trying to model

Jeff Hawkins (1:52:59.680)

something. We both have different information, which leads to our different models. And therefore

Jeff Hawkins (1:53:03.760)

I shouldn't hold it against you and you shouldn't hold it against me. And we can at least agree that,

Jeff Hawkins (1:53:07.760)

well, what can we look for in that's common ground to test our, our beliefs, as opposed to so much,

Jeff Hawkins (1:53:13.600)

uh, as we raise our kids on dogma, which is this is a fact, this is a fact, and these people are

Jeff Hawkins (1:53:20.080)

bad. And, and, and, you know, where every, if everyone knew just to, to be skeptical of every

Jeff Hawkins (1:53:31.120)

belief and why, and how their brains do that, I think we might have a better world.

Lex Fridman (1:53:36.560)

Do you think the human mind is able to comprehend reality? So you talk about this creating models

Lex Fridman (1:53:45.600)

how close do you think we get to, uh, to reality? There's so the wildest ideas is like Donald

Lex Fridman (1:53:51.440)

Hoffman saying, we're very far away from reality. Do you think we're getting close to reality?

Jeff Hawkins (1:53:56.560)

Well, it depends on what you define reality. Uh, we are getting, we have a model of the world

Jeff Hawkins (1:54:02.000)

that's very useful, right? For, for basic goals. Well, for our survival and our pleasure right

Jeff Hawkins (1:54:10.000)

now. Right. Um, so that's useful. Um, I mean, it's really useful. Oh, we can build planes. We can build computers. We can do these things. Right.

Jeff Hawkins (1:54:17.200)

Uh, I don't think, I don't know the answer to that question. Um, I think that's part of the

Jeff Hawkins (1:54:24.080)

question we're trying to figure out, right? Like, you know, obviously if you end up with a theory of

Jeff Hawkins (1:54:27.920)

everything that really is a theory of everything and all of a sudden everything comes into play

Lex Fridman (1:54:32.960)

and there's no room for something else, then you might feel like we have a good model of the world.

Jeff Hawkins (1:54:37.120)

Yeah. But if we have a theory of everything and somehow, first of all, you'll never be able to

Jeff Hawkins (1:54:41.440)

really conclusively say it's a theory of everything, but say somehow we are very damn sure it's a theory

Jeff Hawkins (1:54:46.480)

of everything. We understand what happened at the big bang and how just the entirety of the

Jeff Hawkins (1:54:51.680)

physical process. I'm still not sure that gives us an understanding of, uh, the next

Jeff Hawkins (1:54:58.240)

many layers of the hierarchy of abstractions that form. Well, also what if string theory

Jeff Hawkins (1:55:03.600)

turns out to be true? And then you say, well, we have no reality, no modeling what's going on in

Jeff Hawkins (1:55:09.360)

those other dimensions that are wrapped into it on each other. Right. Or, or the multiverse,

Jeff Hawkins (1:55:14.880)

you know, I honestly don't know how for us, for human interaction, for ideas of intelligence,

Lex Fridman (1:55:21.600)

how it helps us to understand that we're made up of vibrating strings that are

Jeff Hawkins (1:55:26.800)

like 10 to the whatever times smaller than us. I don't, you know, you could probably build better

Jeff Hawkins (1:55:33.040)

weapons, a better rockets, but you're not going to be able to understand intelligence. I guess,

Jeff Hawkins (1:55:37.200)

I guess maybe better computers. No, you won't be. I think it's just more purely knowledge.

Jeff Hawkins (1:55:41.680)

You might lead to a better understanding of the, of the beginning of the universe,

Jeff Hawkins (1:55:46.240)

right? It might lead to a better understanding of, uh, I don't know. I guess I think the acquisition

Jeff Hawkins (1:55:52.720)

of knowledge has always been one where you, you pursue it for its own pleasure. Um, and you don't

Jeff Hawkins (1:56:01.200)

always know what is going to make a difference. Yeah. Uh, you're pleasantly surprised by the,

Jeff Hawkins (1:56:06.400)

the weird things you find. Do you think, uh, for the, for the neocortex in general, do you,

Lex Fridman (1:56:11.760)

do you think there's a lot of innovation to be done on the machine side? You know,

Jeff Hawkins (1:56:16.960)

you use the computer as a metaphor quite a bit. Is there different types of computer that would

Jeff Hawkins (1:56:21.600)

help us build intelligence manifestations of intelligent machines? Yeah. Or is it, oh no,

Jeff Hawkins (1:56:26.880)

it's going to be totally crazy. Uh, we have no idea how this is going to look out yet.

Jeff Hawkins (1:56:32.720)

You can already see this. Um, today we've, of course, we model these things on traditional

Jeff Hawkins (1:56:37.760)

computers and now, now GPUs are really popular with, with, uh, you know, neural networks and so

Jeff Hawkins (1:56:43.840)

on. Um, but there are companies coming up with fundamentally new physical substrates, um, that

Jeff Hawkins (1:56:50.640)

are just really cool. I don't know if they're going to work or not. Um, but I think there'll

Jeff Hawkins (1:56:55.840)

be decades of innovation here. Yeah. Totally. Do you think the final thing will be messy,

Jeff Hawkins (1:57:01.360)

like our biology is messy? Or do you think, uh, it's, it's the, it's the old bird versus

Jeff Hawkins (1:57:07.360)

airplane question, or do you think we could just, um, build airplanes that, that fly way better

Jeff Hawkins (1:57:16.320)

than birds in the same way we could build, uh, uh, electrical neocortex? Yeah. You know,

Jeff Hawkins (1:57:23.280)

can I, can I, can I riff on the bird thing a bit? Because I think that's interesting.

Jeff Hawkins (1:57:27.040)

People really misunderstand this. The Wright brothers, um, the problem they were trying to

Jeff Hawkins (1:57:33.120)

solve was controlled flight, how to turn an airplane, not how to propel an airplane.

Jeff Hawkins (1:57:38.320)

They weren't worried about that. Interesting. Yeah. They already had, at that time,

Jeff Hawkins (1:57:41.600)

there was already wing shapes, which they had from studying birds. There was already gliders

Jeff Hawkins (1:57:45.520)

that carry people. The problem was if you put a rudder on the back of a glider and you turn it,

Jeff Hawkins (1:57:49.440)

the plane falls out of the sky. So the problem was how do you control flight? And they studied

Jeff Hawkins (1:57:55.680)

birds and they actually had birds in captivity. They watched birds in wind tunnels. They observed

Jeff Hawkins (1:58:00.240)

them in the wild and they discovered the secret was the birds twist their wings when they turn.

Lex Fridman (1:58:05.200)

And so that's what they did on the Wright brothers flyer. They had these sticks that

Jeff Hawkins (1:58:07.840)

you would twist the wing. And that was the, that was their innovation, not the propeller.

Lex Fridman (1:58:12.320)

And today airplanes still twist their wings. We don't twist the entire wing. We just twist

Jeff Hawkins (1:58:16.720)

the tail end of it, the flaps, which is the same thing. So today's airplanes fly on the

Jeff Hawkins (1:58:22.000)

same principles as birds would observe. So everyone get that analogy wrong, but let's

Jeff Hawkins (1:58:26.960)

step back from that. Once you understand the principles of flight, you can choose

Lex Fridman (1:58:32.640)

how to implement them. No one's going to use bones and feathers and muscles, but they do have wings

Lex Fridman (1:58:39.120)

and we don't flap them. We have propellers. So when we have the principles of computation that

Jeff Hawkins (1:58:45.040)

goes on to modeling the world in a brain, we understand those principles very clearly.

Jeff Hawkins (1:58:50.160)

We have choices on how to implement them. And some of them will be biological like and some won't.

Lex Fridman (1:58:54.400)

And, but I do think there's going to be a huge amount of innovation here.

Jeff Hawkins (1:58:59.600)

Just think about the innovation when in the computer, they had to invent the transistor,

Jeff Hawkins (1:59:03.920)

they invented the Silicon chip. They had to invent, you know, then this software. I mean,

Jeff Hawkins (1:59:09.200)

it's millions of things they had to do, memory systems. We're going to do, it's going to be

Jeff Hawkins (1:59:13.360)

similar. Well, it's interesting that the deep learning, the effectiveness of deep learning for

Jeff Hawkins (1:59:19.760)

specific tasks is driving a lot of innovation in the hardware, which may have effects for actually

Jeff Hawkins (1:59:27.120)

allowing us to discover intelligence systems that operate very differently or at least much

Jeff Hawkins (1:59:31.760)

bigger than deep learning. Yeah. Interesting. So ultimately it's good to have an application

Jeff Hawkins (1:59:37.040)

that's making our life better now because the capitalist process, if you can make money.

Jeff Hawkins (1:59:42.960)

Yeah. Yeah. That works. I mean, the other way, I mean, Neil deGrasse Tyson writes about this

Jeff Hawkins (1:59:48.080)

is the other way we fund science, of course, is through military. So like, yeah. Conquests.

Lex Fridman (1:59:53.360)

So here's an interesting thing we're doing on this regard. So we've decided, we used to have

Jeff Hawkins (1:59:57.920)

a series of these biological principles and we can see how to build these intelligent machines,

Lex Fridman (20:05.040)

but this also applies for just basic vision, just like seeing things. It's almost like a

Jeff Hawkins (20:10.640)

precondition of just perceiving the world is predicting it. So just everything that you see

Jeff Hawkins (20:18.000)

is first passed through your prediction. Everything you see and feel. In fact,

Jeff Hawkins (20:23.600)

this was the insight I had back in the early 80s. And I know that people have reached the same idea

Jeff Hawkins (20:31.680)

is that every sensory input you get, not just vision, but touch and hearing, you have an

Jeff Hawkins (20:37.600)

expectation about it and a prediction. Sometimes you can predict very accurately. Sometimes you

Jeff Hawkins (20:43.440)

can't. I can't predict what next word is going to come out of your mouth. But as you start talking,

Jeff Hawkins (20:47.520)

I'll get better and better predictions. And if you talk about some topics, I'd be very surprised.

Lex Fridman (20:51.440)

So I have this sort of background prediction that's going on all the time for all of my senses.

Jeff Hawkins (20:58.160)

Again, the way I think about that is this is how we learn. It's more about how we learn.

Jeff Hawkins (21:04.960)

It's a test of our understanding. Our predictions are a test. Is this really a water bottle? If it

Jeff Hawkins (21:10.000)

is, I shouldn't see a little finger sticking out the side. And if I saw a little finger sticking

Jeff Hawkins (21:14.960)

out, I was like, oh, what the hell's going on? That's not normal. I mean, that's fascinating

Jeff Hawkins (21:20.480)

that... Let me linger on this for a second. It really honestly feels that prediction is

Jeff Hawkins (21:27.760)

fundamental to everything, to the way our mind operates, to intelligence. So it's just a different

Jeff Hawkins (21:35.760)

way to see intelligence, which is like everything starts a prediction. And prediction requires a

Jeff Hawkins (21:41.040)

model. You can't predict something unless you have a model of it. Right. But the action is

Jeff Hawkins (21:46.880)

prediction. So the thing the model does is prediction. But it also... Yeah. But you can

Jeff Hawkins (21:53.280)

then extend it to things like, oh, what would happen if I took this today? I went and did this.

Lex Fridman (21:59.600)

What would be likely? Or how... You can extend prediction to like, oh, I want to get a promotion

Jeff Hawkins (22:04.320)

at work. What action should I take? And you can say, if I did this, I predict what might happen.

Jeff Hawkins (22:09.280)

If I spoke to someone, I predict what might happen. So it's not just low level predictions.

Jeff Hawkins (22:13.360)

Yeah. It's all predictions. It's all predictions. It's like this black box so you can ask basically

Jeff Hawkins (22:17.440)

any question, low level or high level. So we started off with that observation. It's

Jeff Hawkins (22:21.920)

this nonstop prediction. And I write about this in the book. And then we asked, how do neurons

Lex Fridman (22:27.120)

actually make predictions physically? Like what does the neuron do when it makes a prediction?

Jeff Hawkins (22:32.400)

Or the neural tissue does when it makes a prediction. And then we asked, what are the

Jeff Hawkins (22:35.760)

mechanisms by how we build a model that allows you to make predictions? So we started with prediction

Jeff Hawkins (22:40.400)

as sort of the fundamental research agenda, if in some sense. And say, well, we understand how

Jeff Hawkins (22:47.520)

the brain makes predictions. We'll understand how it builds these models and how it learns.

Lex Fridman (22:51.360)

And that's the core of intelligence. So it was the key that got us in the door

Lex Fridman (22:55.680)

to say, that is our research agenda. Understand predictions.

Lex Fridman (22:59.360)

So in this whole process, where does intelligence originate, would you say?

Lex Fridman (23:05.200)

So if we look at things that are much less intelligence to humans and you start to build

Jeff Hawkins (23:12.560)

up a human through the process of evolution, where's this magic thing that has a prediction

Lex Fridman (23:19.920)

model or a model that's able to predict that starts to look a lot more like intelligence?

Jeff Hawkins (23:24.720)

Is there a place where Richard Dawkins wrote an introduction to your book, an excellent

Jeff Hawkins (23:30.960)

introduction? I mean, it's, it puts a lot of things into context and it's funny just looking

Jeff Hawkins (23:36.320)

at parallels for your book and Darwin's Origin of Species. So Darwin wrote about the origin

Lex Fridman (23:42.640)

of species. So what is the origin of intelligence?

Jeff Hawkins (23:47.760)

Well, we have a theory about it and it's just that, it's a theory. The theory goes as follows.

Jeff Hawkins (23:53.200)

As soon as living things started to move, they're not just floating in sea, they're not just a

Jeff Hawkins (23:58.720)

plant, you know, grounded someplace. As soon as they started to move, there was an advantage to

Jeff Hawkins (24:03.920)

moving intelligently, to moving in certain ways. And there's some very simple things you can do,

Jeff Hawkins (24:08.960)

you know, bacteria or single cell organisms can move towards the source of gradient of

Jeff Hawkins (24:14.480)

food or something like that. But an animal that might know where it is and know where it's been

Lex Fridman (24:19.280)

and how to get back to that place, or an animal that might say, oh, there was a source of food

Jeff Hawkins (24:23.520)

someplace, how do I get to it? Or there was a danger, how do I get to it? There was a mate, how

Jeff Hawkins (24:29.040)

do I get to them? There was a big evolutionary advantage to that. So early on, there was a

Jeff Hawkins (24:34.480)

pressure to start understanding your environment, like where am I and where have I been? And what

Jeff Hawkins (24:40.640)

happened in those different places? So we still have this neural mechanism in our brains. In the

Jeff Hawkins (24:49.600)

mammals, it's in the hippocampus and entorhinal cortex, these are older parts of the brain.

Lex Fridman (24:55.520)

And these are very well studied. We build a map of the of our environment. So these neurons in

Jeff Hawkins (25:02.240)

these parts of the brain know where I am in this room, and where the door was and things like that.

Lex Fridman (25:07.360)

So a lot of other mammals have this?

Jeff Hawkins (25:09.360)

All mammals have this, right? And almost any animal that knows where it is, and get around

Jeff Hawkins (25:15.600)

must have some mapping system, must have some way of saying, I've learned a map of my environment,

Jeff Hawkins (25:21.360)

I have hummingbirds in my backyard. And they go to the same places all the time. They must know

Jeff Hawkins (25:26.640)

where they are. They just know where they are when they're not just randomly flying around. They

Jeff Hawkins (25:30.000)

know. They know particular flowers they come back to. So we all have this. And it turns out it's

Jeff Hawkins (25:36.160)

very tricky to get neurons to do this, to build a map of an environment. And so we now know,

Jeff Hawkins (25:42.320)

there's these famous studies that are still very active about place cells and grid cells and these

Jeff Hawkins (25:47.440)

other types of cells in the older parts of the brain, and how they build these maps of the world.

Jeff Hawkins (25:51.920)

It's really clever. It's obviously been under a lot of evolutionary pressure over a long period

Jeff Hawkins (25:55.920)

of time to get good at this. So animals now know where they are. What we think has happened,

Lex Fridman (26:01.920)

and there's a lot of evidence to suggest this, is that that mechanism we learned to map,

Jeff Hawkins (26:06.080)

like a space, was repackaged. The same type of neurons was repackaged into a more compact form.

Lex Fridman (26:17.840)

And that became the cortical column. And it was in some sense, genericized, if that's a word. It

Jeff Hawkins (26:23.760)

was turned into a very specific thing about learning maps of environments to learning maps

Jeff Hawkins (26:28.800)

of anything, learning a model of anything, not just your space, but coffee cups and so on. And

Jeff Hawkins (26:34.960)

it got sort of repackaged into a more compact version, a more universal version,

Lex Fridman (26:41.280)

and then replicated. So the reason we're so flexible is we have a very generic version of

Jeff Hawkins (26:46.800)

this mapping algorithm, and we have 150,000 copies of it. Sounds a lot like the progress

Jeff Hawkins (26:52.800)

of deep learning. How so? So take neural networks that seem to work well for a specific task,

Jeff Hawkins (27:00.480)

compress them, and multiply it by a lot. And then you just stack them on top of it. It's like the

Jeff Hawkins (27:07.680)

story of transformers in natural language processing. Yeah. But in deep learning networks,

Jeff Hawkins (27:12.640)

they end up, you're replicating an element, but you still need the entire network to do anything.

Jeff Hawkins (27:18.160)

Right. Here, what's going on, each individual element is a complete learning system. This is

Lex Fridman (27:24.240)

why I can take a human brain, cut it in half, and it still works. It's the same thing.

Jeff Hawkins (27:29.680)

It's pretty amazing. It's fundamentally distributed. It's fundamentally distributed,

Jeff Hawkins (27:34.000)

complete modeling systems. But that's our story we like to tell. I would guess it's likely largely

Jeff Hawkins (27:42.560)

right. But there's a lot of evidence supporting that story, this evolutionary story. The thing

Jeff Hawkins (27:50.080)

which brought me to this idea is that the human brain got big very quickly. So that led to the

Jeff Hawkins (27:58.720)

proposal a long time ago that, well, there's this common element just instead of creating

Jeff Hawkins (28:02.640)

new things, it just replicated something. We also are extremely flexible. We can learn things that

Jeff Hawkins (28:07.680)

we had no history about. And that tells it that the learning algorithm is very generic. It's very

Jeff Hawkins (28:15.360)

kind of universal because it doesn't assume any prior knowledge about what it's learning.

Lex Fridman (28:20.960)

And so you combine those things together and you say, okay, well, how did that come about? Where

Jeff Hawkins (28:26.160)

did that universal algorithm come from? It had to come from something that wasn't universal. It

Jeff Hawkins (28:29.760)

came from something that was more specific. So anyway, this led to our hypothesis that

Jeff Hawkins (28:34.000)

you would find grid cells and place cell equivalents in the neocortex. And when we

Jeff Hawkins (28:38.960)

first published our first papers on this theory, we didn't know of evidence for that. It turns out

Jeff Hawkins (28:43.760)

there was some, but we didn't know about it. So then we became aware of evidence for grid

Jeff Hawkins (28:48.960)

cells in parts of the neocortex. And then now there's been new evidence coming out. There's some

Jeff Hawkins (28:53.200)

interesting papers that came out just January of this year. So one of our predictions was if this

Jeff Hawkins (28:59.360)

evolutionary hypothesis is correct, we would see grid cell place cell equivalents, cells that work

Jeff Hawkins (29:04.000)

like them through every column in the neocortex. And that's starting to be seen. What does it mean

Jeff Hawkins (29:08.640)

that, why is it important that they're present? Because it tells us, well, we're asking about the

Jeff Hawkins (29:13.920)

evolutionary origin of intelligence, right? So our theory is that these columns in the cortex

Jeff Hawkins (29:19.120)

are working on the same principles, they're modeling systems. And it's hard to imagine how

Jeff Hawkins (29:25.120)

neurons do this. And so we said, hey, it's really hard to imagine how neurons could learn these

Jeff Hawkins (29:30.240)

models of things. We can talk about the details of that if you want. But there's this other part

Jeff Hawkins (29:36.480)

of the brain, we know that learns models of environments. So could that mechanism to learn

Jeff Hawkins (29:41.840)

to model this room be used to learn to model the water bottle? Is it the same mechanism? So we said

Jeff Hawkins (29:47.280)

it's much more likely the brain's using the same mechanism, which case it would have these equivalent

Jeff Hawkins (29:52.400)

cell types. So it's basically the whole theory is built on the idea that these columns have

Jeff Hawkins (29:57.920)

reference frames and they're learning these models and these grid cells create these reference frames.

Lex Fridman (2:00:01.920)

but we've decided to apply some of these principles to today's machine learning techniques.

Lex Fridman (2:00:07.280)

So one of the, we didn't talk about this principle. One is a sparsity in the brain,

Jeff Hawkins (2:00:11.600)

um, most of the neurons are active at any point in time. It's sparse and the connectivity is sparse

Lex Fridman (2:00:15.440)

and that's different than deep learning networks. Um, so we've already shown that we can speed up

Jeff Hawkins (2:00:20.800)

existing deep learning networks, uh, anywhere from 10 to a factor of a hundred. I mean,

Jeff Hawkins (2:00:26.400)

literally a hundred, um, and make a more robust at the same time. So this is commercially very,

Jeff Hawkins (2:00:31.760)

very valuable. Um, and so, you know, if we can prove this actually in the largest systems that

Jeff Hawkins (2:00:38.960)

are commercially applied today, there's a big commercial desire to do this. Well,

Jeff Hawkins (2:00:44.240)

sparsity is something that doesn't run really well on existing hardware. It doesn't really run

Jeff Hawkins (2:00:50.640)

really well, um, on, um, GPUs, um, and on CPUs. And so that would be a way of sort of bringing more,

Jeff Hawkins (2:00:59.520)

more brain principles into the existing system on a, on a commercially valuable basis.

Jeff Hawkins (2:01:03.920)

Another thing we can think we can do is we're going to use these dendrites,

Jeff Hawkins (2:01:06.960)

um, models that we, uh, I talked earlier about the prediction occurring inside a neuron

Jeff Hawkins (2:01:13.200)

that that basic property can be applied to existing neural networks and allow them to

Jeff Hawkins (2:01:18.000)

learn continuously, which is something they don't do today. And so the dendritic spikes that you

Jeff Hawkins (2:01:22.960)

were talking about. Yeah. Well, we wouldn't model the spikes, but the idea that you have

Jeff Hawkins (2:01:26.960)

that neuron today's neural networks have this company called the point neurons is a very simple

Jeff Hawkins (2:01:30.640)

model of a neuron. And, uh, by adding dendrites to them at just one more level of complexity,

Jeff Hawkins (2:01:36.000)

uh, that's in biological systems, you can solve problems in continuous learning, um,

Lex Fridman (2:01:41.520)

and rapid learning. So we're trying to take, we're trying to bring the existing field,

Lex Fridman (2:01:47.760)

and we'll see if we can do it. We're trying to bring the existing field of machine learning,

Jeff Hawkins (2:01:51.360)

um, commercially along with us, you brought up this idea of keeping, you know,

Jeff Hawkins (2:01:55.040)

paying for it commercially along with us as we move towards the ultimate goal of a true AI system.

Jeff Hawkins (2:02:00.320)

Even small innovations on your own networks are really, really exciting.

Lex Fridman (2:02:04.000)

Yeah.

Jeff Hawkins (2:02:04.480)

Is it seems like such a trivial model of the brain and applying different insights

Jeff Hawkins (2:02:11.920)

that just even, like you said, continuous, uh, learning or, uh, making it more asynchronous

Jeff Hawkins (2:02:19.360)

or maybe making more dynamic or like, uh, incentivizing, making it robust and making it

Jeff Hawkins (2:02:28.720)

somehow much better incentivizing sparsity, uh, somehow. Yeah. Well, if you can make things a

Jeff Hawkins (2:02:35.840)

hundred times faster, then there's plenty of incentive. That's true. People, people are

Jeff Hawkins (2:02:40.480)

spending millions of dollars, you know, just training some of these networks. Now these, uh,

Jeff Hawkins (2:02:44.400)

these transforming networks, let me ask you the big question for young people listening to this

Jeff Hawkins (2:02:51.520)

today in high school and college, what advice would you give them in terms of, uh, which career

Jeff Hawkins (2:02:57.280)

path to take and, um, maybe just about life in general? Well, in my case, um, I didn't start

Lex Fridman (2:03:06.720)

life with any kind of goals. I was, when I was going to college, it's like, Oh, what do I study?

Jeff Hawkins (2:03:11.040)

Well, maybe I'll do this electrical engineering stuff, you know? Um, it wasn't like, you know,

Jeff Hawkins (2:03:15.840)

today you see some of these young kids are so motivated, like I'm changing the world. I was

Jeff Hawkins (2:03:18.720)

like, you know, whatever. And, um, but then I did fall in love with something besides my wife,

Lex Fridman (2:03:25.920)

but I fell in love with this, like, Oh my God, it would be so cool to understand how the brain works.

Lex Fridman (2:03:30.800)

And then I, I said to myself, that's the most important thing I could work on. I can't imagine

Jeff Hawkins (2:03:34.800)

anything more important because if we understand how the brains work, you build tells the machines

Lex Fridman (2:03:38.240)

and they could figure out all the other big questions of the world. Right. So, and then I

Jeff Hawkins (2:03:42.240)

said, but I want to understand how I work. So I fell in love with this idea and I became passionate

Jeff Hawkins (2:03:46.320)

about it. And this is a trope. People say this, but it was, it's true because I was passionate

Jeff Hawkins (2:03:54.160)

about it. I was able to put up almost so much crap, you know, you know, I was, I was in that,

Jeff Hawkins (2:04:01.040)

you know, I was like person said, you can't do this. I was, I was a graduate student at Berkeley

Jeff Hawkins (2:04:05.200)

when they said, you can't study this problem, you know, no one's can solve this or you can't get

Jeff Hawkins (2:04:09.040)

funded for it. You know, then I went into do mobile computing and it was like, people say,

Jeff Hawkins (2:04:13.120)

you can't do that. You can't build a cell phone, you know? So, but all along I kept being motivated

Jeff Hawkins (2:04:18.880)

because I wanted to work on this problem. I said, I want to understand the brain works. And I got

Jeff Hawkins (2:04:22.720)

myself, you know, I got one lifetime. I'm going to figure it out, do the best I can. So by having

Jeff Hawkins (2:04:28.160)

that, cause you know, it's really, as you pointed out, Lex, it's really hard to do these things.

Jeff Hawkins (2:04:33.440)

People, it just, there's so many downers along the way. So many ways, obstacles to get in your

Jeff Hawkins (2:04:38.320)

way. Yeah. I'm sitting here happy all the time, but trust me, it's not always like that.

Jeff Hawkins (2:04:42.000)

Well, that's, I guess the happiness, the passion is a prerequisite for surviving the whole thing.

Jeff Hawkins (2:04:47.520)

Yeah, I think so. I think that's right. And so I don't want to sit to someone and say, you know,

Jeff Hawkins (2:04:53.120)

you need to find a passion and do it. No, maybe you don't. But if you do find something you're

Jeff Hawkins (2:04:57.920)

passionate about, then you can follow it as far as your passion will let you put up with it.

Lex Fridman (2:05:04.000)

Do you remember how you found it? How the spark happened?

Lex Fridman (2:05:09.200)

Why specifically for me?

Jeff Hawkins (2:05:10.800)

Yeah. Cause you said it's such an interesting, so like almost like later in life, by later,

Jeff Hawkins (2:05:15.200)

I mean like not when you were five, you didn't really know. And then all of a sudden you fell

Jeff Hawkins (2:05:21.120)

in love with that idea. Yeah, yeah. There was two separate events that compounded one another.

Jeff Hawkins (2:05:25.600)

One, when I was probably a teenager, it might've been 17 or 18, I made a list of the most

Jeff Hawkins (2:05:31.520)

interesting problems I could think of. First was why does the universe exist? It seems like

Jeff Hawkins (2:05:36.960)

not existing is more likely. The second one was, well, given it exists, why does it behave the way

Jeff Hawkins (2:05:41.120)

it does? Laws of physics, why is it equal MC squared, not MC cubed? That's an interesting

Jeff Hawkins (2:05:45.680)

question. The third one was like, what's the origin of life? And the fourth one was, what's

Jeff Hawkins (2:05:51.680)

intelligence? And I stopped there. I said, well, that's probably the most interesting one. And I

Jeff Hawkins (2:05:56.240)

put that aside as a teenager. But then when I was 22 and I was reading the, no, excuse me, it was

Jeff Hawkins (2:06:05.680)

1979, excuse me, 1979, I was reading, so I was, at that time I was 22, I was reading the September

Jeff Hawkins (2:06:13.520)

issue of Scientific American, which is all about the brain. And then the final essay was by Francis

Jeff Hawkins (2:06:19.440)

Crick, who of DNA fame, and he had taken his interest to studying the brain now. And he said,

Jeff Hawkins (2:06:25.920)

you know, there's something wrong here. He says, we got all this data, all this fact, this is 1979,

Jeff Hawkins (2:06:33.600)

all these facts about the brain, tons and tons of facts about the brain. Do we need more facts? Or do

Jeff Hawkins (2:06:39.360)

we just need to think about a way of rearranging the facts we have? Maybe we're just not thinking

Jeff Hawkins (2:06:42.800)

about the problem correctly. Cause he says, this shouldn't be like this. So I read that and I said,

Jeff Hawkins (2:06:51.440)

wow. I said, I don't have to become like an experimental neuroscientist. I could just

Jeff Hawkins (2:06:57.360)

take, look at all those facts and try and become a theoretician and try to figure it out. And I said

Jeff Hawkins (2:07:04.320)

that I felt like it was something I would be good at. I said, I wouldn't be a good experimentalist.

Jeff Hawkins (2:07:08.640)

I don't have the patience for it, but I'm a good thinker and I love puzzles. And this is like the

Jeff Hawkins (2:07:14.320)

biggest puzzle in the world. It's the biggest puzzle of all time. And I got all the puzzle

Jeff Hawkins (2:07:18.240)

pieces in front of me. Damn, that was exciting. And there's something obviously you can't

Jeff Hawkins (2:07:23.360)

convert into words that just kind of sparked this passion. And I have that a few times in my life,

Jeff Hawkins (2:07:29.440)

just something just like you, it grabs you. Yeah. I felt it was something that was both

Jeff Hawkins (2:07:37.680)

important and that I could make a contribution to. And so all of a sudden it felt like,

Jeff Hawkins (2:07:41.680)

oh, it gave me purpose in life. I honestly don't think it has to be as big as one of those four

Jeff Hawkins (2:07:46.960)

questions. I think you can find those things in the smallest. Oh, absolutely. David Foster Wallace

Jeff Hawkins (2:07:54.160)

said like the key to life is to be unboreable. I think it's very possible to find that intensity

Jeff Hawkins (2:08:01.040)

of joy in the smallest thing. Absolutely. I'm just, you asked me my story. Yeah. No, but I'm

Jeff Hawkins (2:08:06.000)

actually speaking to the audience. It doesn't have to be those four. You happen to get excited by one

Jeff Hawkins (2:08:10.800)

of the bigger questions of in the universe, but even the smallest things and watching the Olympics

Jeff Hawkins (2:08:18.320)

now, just giving yourself life, giving your life over to the study and the mastery of a particular

Jeff Hawkins (2:08:25.920)

sport is fascinating. And if it sparks joy and passion, you're able to, in the case of the

Jeff Hawkins (2:08:32.720)

Olympics, basically suffer for like a couple of decades to achieve. I mean, you can find joy and

Jeff Hawkins (2:08:37.520)

passion just being a parent. I mean, yeah, the parenting one is funny. So I was, not always,

Lex Fridman (2:08:43.600)

but for a long time, wanted kids and get married and stuff. And especially that has to do with the

Jeff Hawkins (2:08:48.720)

fact that I've seen a lot of people that I respect get a whole nother level of joy from kids. And

Jeff Hawkins (2:08:58.880)

at first is like, you're thinking is, well, like I don't have enough time in the day, right? If I

Jeff Hawkins (2:09:05.920)

have this passion to solve, but like, if I want to solve intelligence, how's this kid situation

Jeff Hawkins (2:09:13.200)

going to help me? But then you realize that, you know, like you said, the things that sparks joy,

Lex Fridman (2:09:22.000)

and it's very possible that kids can provide even a greater or deeper, more meaningful joy than

Jeff Hawkins (2:09:28.640)

those bigger questions when they enrich each other. And that seemed like, obviously when I

Jeff Hawkins (2:09:34.160)

was younger, it's probably a counterintuitive notion because there's only so many hours in the

Lex Fridman (2:09:37.920)

day, but then life is finite and you have to pick the things that give you joy.

Jeff Hawkins (2:09:44.160)

Yeah. But you also understand you can be patient too. I mean, it's finite, but we do have, you know,

Jeff Hawkins (2:09:50.800)

whatever, 50 years or something. So in my case, I had to give up on my dream of the neuroscience

Jeff Hawkins (2:09:58.480)

because I was a graduate student at Berkeley and they told me I couldn't do this and I couldn't

Jeff Hawkins (2:10:02.240)

get funded. And so I went back in the computing industry for a number of years. I thought it

Jeff Hawkins (2:10:09.440)

would be four, but it turned out to be more. But I said, I'll come back. I'm definitely going to

Jeff Hawkins (2:10:14.880)

come back. I know I'm going to do this computer stuff for a while, but I'm definitely coming back.

Jeff Hawkins (2:10:17.920)

Everyone knows that. And it's like raising kids. Well, yeah, you have to spend a lot of time with

Jeff Hawkins (2:10:22.800)

your kids. It's fun, enjoyable. But that doesn't mean you have to give up on other dreams. It just

Jeff Hawkins (2:10:28.240)

means that you may have to wait a week or two to work on that next idea. Well, you talk about the

Jeff Hawkins (2:10:36.800)

darker side of me, disappointing sides of human nature that we're hoping to overcome so that we

Jeff Hawkins (2:10:42.240)

don't destroy ourselves. I tend to put a lot of value in the broad general concept of love,

Jeff Hawkins (2:10:48.640)

of the human capacity of compassion towards each other, of just kindness, whatever that longing of

Jeff Hawkins (2:10:58.960)

like just the human to human connection. It connects back to our initial discussion. I tend to

Jeff Hawkins (2:11:05.120)

see a lot of value in this collective intelligence aspect. I think some of the magic of human

Jeff Hawkins (2:11:09.360)

civilization happens when there's a party is not as fun when you're alone. I totally agree with

Jeff Hawkins (2:11:16.080)

you on these issues. Do you think from a neocortex perspective, what role does love play in the human

Jeff Hawkins (2:11:24.080)

condition? Well, those are two separate things from a neocortex point of view. It doesn't impact

Jeff Hawkins (2:11:29.600)

our thinking about the neocortex. From a human condition point of view, I think it's core.

Jeff Hawkins (2:11:34.400)

I mean, we get so much pleasure out of loving people and helping people. I'll rack it up to

Jeff Hawkins (2:11:44.720)

old brain stuff and maybe we can throw it under the bus of evolution if you want. That's fine.

Jeff Hawkins (2:11:52.800)

It doesn't impact how I think about how we model the world, but from a humanity point of view,

Jeff Hawkins (2:11:57.840)

I think it's essential. Well, I tend to give it to the new brain and also I tend to give it to

Jeff Hawkins (2:12:03.680)

the old brain. Also, I tend to think that some aspects of that need to be engineered into AI

Jeff Hawkins (2:12:09.120)

systems, both in their ability to have compassion for other humans and their ability to maximize

Jeff Hawkins (2:12:21.440)

love in the world between humans. I'm more thinking about social networks. Whenever there's a deep

Jeff Hawkins (2:12:27.760)

AI systems in humans, specific applications where it's AI and humans, I think that's something that

Jeff Hawkins (2:12:35.120)

often not talked about in terms of metrics over which you try to maximize,

Lex Fridman (2:12:44.480)

like which metric to maximize in a system. It seems like one of the most

Jeff Hawkins (2:12:48.960)

powerful things in societies is the capacity to love.

Jeff Hawkins (2:12:55.120)

It's fascinating. I think it's a great way of thinking about it. I have been thinking more of

Jeff Hawkins (2:13:01.120)

these fundamental mechanisms in the brain as opposed to the social interaction between humans

Lex Fridman (2:13:06.640)

and AI systems in the future. If you think about that, you're absolutely right. That's a complex

Jeff Hawkins (2:13:13.680)

system. I can have intelligent systems that don't have that component, but they're not interacting

Jeff Hawkins (2:13:17.360)

with people. They're just running something or building some place or something. I don't know.

Lex Fridman (2:13:21.600)

But if you think about interacting with humans, yeah, but it has to be engineered in there. I

Lex Fridman (2:13:26.640)

don't think it's going to appear on its own. That's a good question.

Jeff Hawkins (2:13:30.560)

Yeah. Well, we could, we'll leave that open. In terms of, from a reinforcement learning

Jeff Hawkins (2:13:38.000)

perspective, whether the darker sides of human nature or the better angels of our nature win out,

Jeff Hawkins (2:13:46.880)

statistically speaking, I don't know. I tend to be optimistic and hope that love wins out in the end.

Jeff Hawkins (2:13:52.960)

You've done a lot of incredible stuff and your book is driving towards this fourth question that

Jeff Hawkins (2:14:01.520)

you started with on the nature of intelligence. What do you hope your legacy for people reading

Jeff Hawkins (2:14:08.880)

a hundred years from now? How do you hope they remember your work? How do you hope they remember

Jeff Hawkins (2:14:14.560)

this book? Well, I think as an entrepreneur or a scientist or any human who's trying to accomplish

Jeff Hawkins (2:14:21.920)

some things, I have a view that really all you can do is accelerate the inevitable. Yeah. It's like,

Jeff Hawkins (2:14:30.960)

you know, if we didn't figure out, if we didn't study the brain, someone else will study the

Jeff Hawkins (2:14:33.920)

brain. If, you know, if Elon didn't make electric cars, someone else would do it eventually.

Lex Fridman (2:14:38.080)

And if, you know, if Thomas Edison didn't invent a light bulb, we wouldn't be using candles today.

Jeff Hawkins (2:14:42.400)

So, what you can do as an individual is you can accelerate something that's beneficial

Lex Fridman (2:14:48.880)

and make it happen sooner than it would have. That's really it. That's all you can do.

Jeff Hawkins (2:14:53.680)

You can't create a new reality that it wasn't going to happen. So, from that perspective,

Jeff Hawkins (2:15:01.280)

I would hope that our work, not just me, but our work in general, people would look back and said,

Jeff Hawkins (2:15:07.440)

hey, they really helped make this better future happen sooner. They, you know, they helped us

Jeff Hawkins (2:15:14.160)

understand the nature of false beliefs sooner than they might have. Now we're so happy that

Jeff Hawkins (2:15:18.640)

we have these intelligent machines doing these things, helping us that maybe that solved the

Jeff Hawkins (2:15:22.560)

climate change problem and they made it happen sooner. So, I think that's the best I would hope

Jeff Hawkins (2:15:28.320)

for. Some would say those guys just moved the needle forward a little bit in time.

Jeff Hawkins (2:15:33.280)

Well, I do. It feels like the progress of human civilization is not, is there's a lot

Jeff Hawkins (2:15:40.000)

of trajectories. And if you have individuals that accelerate towards one direction that helps steer

Jeff Hawkins (2:15:48.480)

human civilization. So, I think in those long stretch of time, all trajectories will be traveled.

Lex Fridman (2:15:55.200)

But I think it's nice for this particular civilization on earth to travel down one that's

Jeff Hawkins (2:15:59.840)

not. Well, I think you're right. We have to take the whole period of, you know, World War II,

Jeff Hawkins (2:16:03.440)

Nazism or something like that. Well, that was a bad sidestep, right? We've been over there for a

Jeff Hawkins (2:16:07.520)

while. But, you know, there is the optimistic view about life that ultimately it does converge

Jeff Hawkins (2:16:13.680)

in a positive way. It progresses ultimately, even if we have years of darkness. So, yeah. So,

Jeff Hawkins (2:16:21.920)

I think you can perhaps that's accelerating the positive could also mean eliminating some bad

Jeff Hawkins (2:16:27.200)

missteps along the way, too. But I'm an optimistic in that way. Despite we talked about the end of

Jeff Hawkins (2:16:34.560)

civilization, you know, I think we're going to live for a long time. I hope we are. I think our

Jeff Hawkins (2:16:40.080)

society in the future is going to be better. We're going to have less discord. We're going to have

Jeff Hawkins (2:16:42.640)

less people killing each other. You know, we'll make them live in some sort of way that's compatible

Jeff Hawkins (2:16:47.600)

with the carrying capacity of the earth. I'm optimistic these things will happen. And all we

Jeff Hawkins (2:16:53.520)

can do is try to get there sooner. And at the very least, if we do destroy ourselves,

Jeff Hawkins (2:16:57.840)

we'll have a few satellites orbiting that will tell alien civilization that we were once here.

Jeff Hawkins (2:17:05.680)

Or maybe our future, you know, future inhabitants of earth. You know, imagine we,

Jeff Hawkins (2:17:10.560)

you know, the planet of the apes in here. You know, we kill ourselves, you know,

Jeff Hawkins (2:17:13.600)

a million years from now or a billion years from now. There's another species on the planet.

Jeff Hawkins (2:17:16.480)

Curious creatures were once here. Jeff, thank you so much for your work. And thank you so much for

Jeff Hawkins (2:17:23.200)

talking to me once again. Well, actually, it's great. I love what you do. I love your podcast.

Jeff Hawkins (2:17:27.040)

You have the most interesting people, me aside. So it's a real service, I think you do for,

Jeff Hawkins (2:17:35.280)

in a very broader sense for humanity, I think. Thanks, Jeff. All right. It's a pleasure.

Lex Fridman (2:17:40.000)

Thanks for listening to this conversation with Jeff Hawkins. And thank you to

Jeff Hawkins (2:17:43.360)

Codecademy, BioOptimizers, ExpressVPN, Asleep, and Blinkist. Check them out in the description

Jeff Hawkins (2:17:50.960)

to support this podcast. And now, let me leave you with some words from Albert Camus.

Jeff Hawkins (2:17:57.600)

An intellectual is someone whose mind watches itself. I like this, because I'm happy to be

Jeff Hawkins (2:18:04.240)

both halves, the watcher and the watched. Can they be brought together? This is the

Jeff Hawkins (2:18:10.880)

practical question we must try to answer. Thank you for listening. I hope to see you next time.

Lex Fridman (30:02.640)

So it's basically the major, in some sense, the major predictive part of this theory is that we

Jeff Hawkins (30:09.200)

will find these equivalent mechanisms in each column in the neocortex, which tells us that

Jeff Hawkins (30:14.560)

that's what they're doing. They're learning these sensory motor models of the world. So we're pretty

Jeff Hawkins (30:21.600)

confident that would happen, but now we're seeing the evidence. So the evolutionary process, nature

Jeff Hawkins (30:26.000)

does a lot of copy and paste and see what happens. Yeah. Yeah. There's no direction to it. But it

Jeff Hawkins (30:31.920)

just found out like, hey, if I took these elements and made more of them, what happens? And let's hook

Jeff Hawkins (30:37.600)

them up to the eyes and let's hook them to ears. And that seems to work pretty well for us. Again,

Jeff Hawkins (30:43.600)

just to take a quick step back to our conversation of collective intelligence.

Lex Fridman (30:48.960)

Do you sometimes see that as just another copy and paste aspect is copying and pasting

Jeff Hawkins (30:56.160)

these brains and humans and making a lot of them and then creating social structures that then

Jeff Hawkins (31:04.080)

almost operate as a single brain? I wouldn't have said that, but you said it sounded pretty good.

Lex Fridman (31:08.320)

So to you, the brain is its own thing.

Jeff Hawkins (31:15.440)

I mean, our goal is to understand how the neocortex works. We can argue how essential

Jeff Hawkins (31:20.560)

that is to understand the human brain because it's not the entire human brain. You can argue

Lex Fridman (31:25.200)

how essential that is to understanding human intelligence. You can argue how essential this

Jeff Hawkins (31:29.680)

is to sort of communal intelligence. Our goal was to understand the neocortex.

Lex Fridman (31:38.640)

Yeah. So what is the neocortex and where does it fit

Lex Fridman (31:41.680)

in the various aspects of what the brain does? Like how important is it to you?

Jeff Hawkins (31:46.480)

Well, obviously, again, I mentioned again in the beginning, it's about 70 to 75% of the volume of

Jeff Hawkins (31:53.680)

the human brain. So it dominates our brain in terms of size. Not in terms of number of neurons,

Lex Fridman (31:58.640)

but in terms of size.

Jeff Hawkins (32:00.640)

Size isn't everything, Jeff.

Lex Fridman (32:02.400)

I know, but it's not that. We know that all high level vision,

Jeff Hawkins (32:09.040)

hearing, and touch happens in the neocortex. We know that all language occurs and is understood

Jeff Hawkins (32:13.920)

in the neocortex, whether that's spoken language, written language, sign language,

Jeff Hawkins (32:17.280)

whether it's language of mathematics, language of physics, music. We know that all high level

Jeff Hawkins (32:23.360)

planning and thinking occurs in the neocortex. If I were to say, what part of your brain designed

Jeff Hawkins (32:27.840)

a computer and understands programming and creates music? It's all the neocortex.

Lex Fridman (32:33.040)

So then that's an undeniable fact. But then there's other parts of our brain are important too,

Jeff Hawkins (32:39.920)

right? Our emotional states, our body regulating our body. So the way I like to look at it is,

Lex Fridman (32:48.400)

can you understand the neocortex without the rest of the brain? And some people say you can't,

Lex Fridman (32:53.200)

and I think absolutely you can. It's not that they're not interacting, but you can understand.

Lex Fridman (32:58.480)

Can you understand the neocortex without understanding the emotions of fear? Yes,

Jeff Hawkins (33:01.920)

you can. You can understand how the system works. It's just a modeling system. I make the analogy

Jeff Hawkins (33:06.480)

in the book that it's like a map of the world, and how that map is used depends on who's using it.

Lex Fridman (33:12.720)

So how our map of our world in our neocortex, how we manifest as a human depends on the rest of our

Lex Fridman (33:19.680)

brain. What are our motivations? What are my desires? Am I a nice guy or not a nice guy?

Lex Fridman (33:23.760)

Am I a cheater or not a cheater? How important different things are in my life?

Lex Fridman (33:33.840)

But the neocortex can be understood on its own. And I say that as a neuroscientist,

Jeff Hawkins (33:39.760)

I know there's all these interactions, and I don't want to say I don't know them and we

Jeff Hawkins (33:43.840)

don't think about them. But from a layperson's point of view, you can say it's a modeling system.

Jeff Hawkins (33:47.840)

I don't generally think too much about the communal aspect of intelligence, which you brought up a

Lex Fridman (33:51.680)

number of times already. So that's not really been my concern.

Jeff Hawkins (33:55.040)

I just wonder if there's a continuum from the origin of the universe, like

Jeff Hawkins (34:00.320)

this pockets of complexities that form living organisms. I wonder if we're just,

Jeff Hawkins (34:08.480)

if you look at humans, we feel like we're at the top. And I wonder if there's like just,

Jeff Hawkins (34:13.120)

I wonder if there's like just where everybody probably every living type pocket of complexity

Jeff Hawkins (34:20.800)

probably thinks they're the, pardon the French, they're the shit. They're at the top of the

Lex Fridman (34:26.240)

pyramid. Well, if they're thinking. Well, then what is thinking? In this sense,

Jeff Hawkins (34:32.240)

the whole point is in their sense of the world, their sense is that they're at the top of it.

Lex Fridman (34:40.560)

I think what is a turtle, but you're, you're, you're bringing up, you know,

Jeff Hawkins (34:44.320)

the problems of complexity and complexity theory are, you know, it's a huge,

Jeff Hawkins (34:48.880)

interesting problem in science. Um, and you know, I think we've made surprisingly little progress

Lex Fridman (34:55.280)

and understanding complex systems in general. Um, and so, you know, the Santa Fe Institute was

Jeff Hawkins (35:01.120)

founded to study this and even the scientists there will say, it's really hard. We haven't

Jeff Hawkins (35:05.200)

really been able to figure out exactly, you know, that science hasn't really congealed yet. We're

Jeff Hawkins (35:10.560)

still trying to figure out the basic elements of that science. Uh, what, you know, where does

Jeff Hawkins (35:15.360)

complexity come from and what is it and how you define it, whether it's DNA creating bodies or

Jeff Hawkins (35:20.000)

phenotypes or it's individuals creating societies or ants and, you know, markets and so on. It's,

Jeff Hawkins (35:26.480)

it's a very complex thing. I'm not a complexity theorist person, right? Um, and I, I think you

Jeff Hawkins (35:32.800)

should ask, well, the brain itself is a complex system. So can we understand that? Um, I think

Jeff Hawkins (35:38.000)

we've made a lot of progress understanding how the brain works. So, uh, but I haven't

Jeff Hawkins (35:42.640)

brought it out to like, oh, well, where are we on the complexity spectrum? You know, it's like,

Jeff Hawkins (35:47.520)

um, it's a great question. I'd prefer for that answer to be we're not special. It seems like

Jeff Hawkins (35:55.680)

if we're honest, most likely we're not special. So if there is a spectrum or probably not in some

Jeff Hawkins (36:01.680)

kind of significant place, there's one thing we could say that we are special. And again,

Jeff Hawkins (36:06.480)

only here on earth, I'm not saying is that if we think about knowledge, what we know,

Jeff Hawkins (36:14.080)

um, we clearly human brains have, um, the only brains that have a certain types of knowledge.

Jeff Hawkins (36:21.040)

We're the only brains on this earth to understand, uh, what the earth is, how old it is,

Jeff Hawkins (36:25.920)

that the universe is a picture as a whole with the only organisms understand DNA and

Jeff Hawkins (36:30.400)

the origins of, you know, of species. Uh, no other species on, on this planet has that knowledge.

Lex Fridman (36:37.200)

So if we think about, I like to think about, you know, one of the endeavors of humanity is to

Jeff Hawkins (36:43.440)

understand the universe as much as we can. Um, I think our species is further along in that

Jeff Hawkins (36:49.920)

undeniably, um, whether our theories are right or wrong, we can debate, but at least we have

Jeff Hawkins (36:54.480)

theories. You know, we, we know that what the sun is and how its fusion is and how what black holes

Jeff Hawkins (36:59.760)

are and, you know, we know general theory of relativity and no other animal has any of this

Jeff Hawkins (37:04.400)

knowledge. So in that sense that we're special, uh, are we special in terms of the hierarchy of

Jeff Hawkins (37:10.800)

complexity in the universe? Probably not. Can we look at a neuron? Yeah. You say that prediction

Jeff Hawkins (37:20.960)

happens in the neuron. What does that mean? So the neuron traditionally is seen as the

Jeff Hawkins (37:24.960)

basic element of the brain. So we, I mentioned this earlier that prediction was our research agenda.

Jeff Hawkins (37:31.760)

Yeah. We said, okay, um, how does the brain make a prediction? Like I I'm about to grab this water

Jeff Hawkins (37:37.840)

bottle and my brain is predicting what I'm going to feel on, on all my parts of my fingers. If I

Jeff Hawkins (37:42.720)

felt something really odd on any part here, I'd notice it. So my brain is predicting what it's

Jeff Hawkins (37:46.560)

going to feel as I grab this thing. So what does that, how does that manifest itself in neural

Jeff Hawkins (37:51.360)

tissue? Right. We got brains made of neurons and there's chemicals and there's neurons and there's

Jeff Hawkins (37:57.360)

spikes and the connect, you know, where, where is the prediction going on? And one argument could be

Jeff Hawkins (38:03.600)

that, well, when I'm predicting something, um, a neuron must be firing in advance. It's like, okay,

Jeff Hawkins (38:09.360)

this neuron represents what you're going to feel and it's firing. It's sending a spike.

Lex Fridman (38:13.600)

And certainly that happens to some extent, but our predictions are so ubiquitous

Jeff Hawkins (38:17.760)

that we're making so many of them, which we're totally unaware of just the vast majority of me

Jeff Hawkins (38:21.360)

have no idea that you're doing this. Um, that it, there wasn't really, we were trying to figure,

Lex Fridman (38:27.120)

how could this be? Where, where are these, where are these happening? Right. And I won't walk you

Jeff Hawkins (38:31.920)

through the whole story unless you insist upon it. But we came to the realization that most of your

Jeff Hawkins (38:38.880)

predictions are occurring inside individual neurons, especially these, the most common

Jeff Hawkins (38:43.440)

are in the parameter cells. And there are, there's a property of neurons. We, everyone knows,

Jeff Hawkins (38:49.120)

or most people know that a neuron is a cell and it has this spike called an action potential,

Lex Fridman (38:53.280)

and it sends information. But we now know that there's these spikes internal to the neuron,

Jeff Hawkins (38:58.160)

they're called dendritic spikes. They travel along the branches of the neuron and they don't leave

Jeff Hawkins (39:03.200)

the neuron. They're just internal only. There's far more dendritic spikes than there are action

Jeff Hawkins (39:08.240)

potentials, far more. They're happening all the time. And what we came to understand that those

Jeff Hawkins (39:14.240)

dendritic spikes, the ones that are occurring are actually a form of prediction. They're telling the

Jeff Hawkins (39:18.880)

neuron, the neuron is saying, I expect that I might become active shortly. And that internal,

Lex Fridman (39:25.360)

so the internal spike is a way of saying, you're going to, you might be generating external spikes

Jeff Hawkins (39:30.240)

soon. I predicted you're going to become active. And, and we've, we've, we wrote a paper in 2016

Jeff Hawkins (39:36.640)

which explained how this manifests itself in neural tissue and how it is that this all works

Jeff Hawkins (39:42.480)

together. But the vast majority, we think it's, there's a lot of evidence supporting it. So we,

Jeff Hawkins (39:48.320)

that's where we think that most of these predictions are internal. That's why you can't

Jeff Hawkins (39:51.360)

be, they're internal to the neuron, you can't perceive them.

Jeff Hawkins (39:54.160)

Well, from understanding the prediction mechanism of a single neuron, do you think there's deep

Jeff Hawkins (40:00.080)

insights to be gained about the prediction capabilities of the mini brains of the neural

Lex Fridman (40:05.520)

brain? Of the mini brains and then the bigger brain and the brain?

Jeff Hawkins (40:08.160)

Oh yeah. Yeah. Yeah. So having a prediction side of their individual neuron is not that useful.

Lex Fridman (40:12.720)

So what? The way it manifests itself in neural tissue is that when a neuron, a neuron emits these

Jeff Hawkins (40:22.320)

spikes are a very singular type event. If a neuron is predicting that it's going to be active, it

Jeff Hawkins (40:27.440)

emits its spike very, a little bit sooner, just a few milliseconds sooner than it would have

Jeff Hawkins (40:31.840)

been. It's like, I give the analogy of the book is like a sprinter on a, on a starting blocks in a,

Jeff Hawkins (40:36.240)

in a race. And if someone says, get ready, set, you get up and you're ready to go. And then when

Jeff Hawkins (40:42.480)

your race starts, you get a little bit earlier start. So that it's that, that ready set is like

Jeff Hawkins (40:46.320)

the prediction and the neurons like ready to go quicker. And what happens is when you have a whole

Jeff Hawkins (40:50.800)

bunch of neurons together and they're all getting these inputs, the ones that are in the predictive

Jeff Hawkins (40:55.520)

state, the ones that are anticipating to become active, if they do become active, they, they

Jeff Hawkins (40:59.920)

sooner, they disable everything else. And it leads to different representations in the brain. So

Jeff Hawkins (41:04.240)

you have to, it's not isolated just to the neuron, the prediction occurs with the neuron,

Lex Fridman (41:09.600)

but the network behavior changes. So what happens under different predictions, different inputs

Jeff Hawkins (41:14.880)

have different representations. So how I, what I predict is going to be different under different

Jeff Hawkins (41:20.800)

contexts, you know, what my input will be is different under different contexts. So this is,

Jeff Hawkins (41:24.960)

this is a key to the whole theory, how this works. So the theory of the thousand brains,

Jeff Hawkins (41:30.560)

if you were to count the number of brains, how would you do it? The thousand brain theory says

Jeff Hawkins (41:35.920)

that basically every cortical column in the, in your, in your cortex is a complete modeling system.

Lex Fridman (41:42.320)

And that when I ask, where do I have a model of something like a coffee cup? It's not in one of

Jeff Hawkins (41:46.800)

those models. It's in thousands of those models. There's thousands of models of coffee cups. That's

Lex Fridman (41:51.040)

what the thousand brains, then there's a voting mechanism, which you lead, which you're, which is

Jeff Hawkins (41:56.160)

the thing you're, which you're conscious of, which leads to your singular perception. That's why you,

Jeff Hawkins (42:01.520)

you perceive something. So that's the thousand brains theory. The details, how we got to that

Jeff Hawkins (42:07.200)

theory are complicated. It wasn't, we just thought of it one day. And one of those details that we

Jeff Hawkins (42:13.440)

had to ask, how does a model make predictions? And we've talked about just these predictive neurons.

Jeff Hawkins (42:18.160)

That's part of this theory. It's like saying, Oh, it's a detail, but it was like a crack in the

Jeff Hawkins (42:22.320)

door. It's like, how are we going to figure out how these neurons built through this? You know,

Lex Fridman (42:24.960)

what is going on here? So we just looked at prediction as like, well, we know that's ubiquitous.

Jeff Hawkins (42:30.080)

We know that every part of the cortex is making predictions. Therefore, whatever the predictive

Jeff Hawkins (42:34.400)

system is, it's going to be everywhere. We know there's a gazillion predictions happening at once.

Lex Fridman (42:39.040)

So this is where we can start teasing apart, you know, ask questions about, you know, how could

Jeff Hawkins (42:44.000)

neurons be making these predictions? And that sort of built up to now what we have this thousand

Jeff Hawkins (42:48.640)

brains theory, which is complex. You know, it's just, I can state it simply, but we just didn't

Jeff Hawkins (42:53.200)

think of it. We had to get there step by step, very, it took years to get there.

Lex Fridman (42:59.200)

And where does reference frames fit in? So, yeah.

Jeff Hawkins (43:04.560)

Okay. So again, a reference frame, I mentioned earlier about the model of a house. And I said,

Jeff Hawkins (43:11.200)

if you're going to build a model of a house in a computer, they have a reference frame. And you

Jeff Hawkins (43:14.560)

can think of reference frame like Cartesian coordinates, like X, Y, and Z axes. So I could

Jeff Hawkins (43:19.600)

say, oh, I'm going to design a house. I can say, well, the front door is at this location, X, Y,

Jeff Hawkins (43:24.400)

Z, and the roof is at this location, X, Y, Z, and so on. That's a type of reference frame.

Lex Fridman (43:29.440)

So it turns out for you to make a prediction, and I walk you through the thought experiment in the

Jeff Hawkins (43:33.600)

book where I was predicting what my finger was going to feel when I touched a coffee cup.

Jeff Hawkins (43:37.360)

It was a ceramic coffee cup, but this one will do. And what I realized is that to make a prediction

Jeff Hawkins (43:45.200)

of what my finger's going to feel, like it's going to feel different than this, what's it feel

Jeff Hawkins (43:48.240)

different if I touch the hole or this thing on the bottom, make that prediction. The cortex needs to

Jeff Hawkins (43:53.280)

know where the finger is, the tip of the finger, relative to the coffee cup. And exactly relative

Jeff Hawkins (43:59.360)

to the coffee cup. And to do that, I have to have a reference frame for the coffee cup. It has to

Jeff Hawkins (44:03.440)

have a way of representing the location of my finger to the coffee cup. And then we realized,

Jeff Hawkins (44:08.160)

of course, every part of your skin has to have a reference frame relative to things that touch.

Lex Fridman (44:11.360)

And then we did the same thing with vision. So the idea that a reference frame is necessary

Jeff Hawkins (44:16.240)

to make a prediction when you're touching something or when you're seeing something

Lex Fridman (44:20.080)

and you're moving your eyes or you're moving your fingers, it's just a requirement

Jeff Hawkins (44:24.000)

to predict. If I have a structure, I'm going to make a prediction. I have to know where it is I'm

Jeff Hawkins (44:29.200)

looking or touching it. So then we said, well, how do neurons make reference frames? It's not obvious.

Jeff Hawkins (44:36.160)

X, Y, Z coordinates don't exist in the brain. It's just not the way it works. So that's when we

Jeff Hawkins (44:40.480)

looked at the older part of the brain, the hippocampus and the anterior cortex, where we knew

Jeff Hawkins (44:45.120)

that in that part of the brain, there's a reference frame for a room or a reference frame for an

Jeff Hawkins (44:49.920)

environment. Remember, I talked earlier about how you could make a map of this room. So we said,

Jeff Hawkins (44:55.200)

oh, they are implementing reference frames there. So we knew that reference frames needed to exist

Jeff Hawkins (45:01.440)

in every quarter of a column. And so that was a deductive thing. We just deduced it. It has to

Jeff Hawkins (45:07.680)

exist. So you take the old mammalian ability to know where you are in a particular space

Lex Fridman (45:15.920)

and you start applying that to higher and higher levels.

Jeff Hawkins (45:18.320)

Yeah. First you apply it to like where your finger is. So here's what I think about it.

Jeff Hawkins (45:22.560)

The old part of the brain says, where's my body in this room? The new part of the brain says,

Jeff Hawkins (45:26.720)

where's my finger relative to this object? Where is a section of my retina relative to

Lex Fridman (45:34.720)

this object? I'm looking at one little corner. Where is that relative to this patch of my retina?

Lex Fridman (45:40.800)

And then we take the same thing and apply it to concepts, mathematics, physics, humanity,

Jeff Hawkins (45:47.280)

whatever you want to think about. And eventually you're pondering your own mortality.

Jeff Hawkins (45:50.240)

Well, whatever. But the point is when we think about the world, when we have knowledge about

Jeff Hawkins (45:55.520)

the world, how is that knowledge organized, Lex? Where is it in your head? The answer is it's in

Jeff Hawkins (46:00.560)

reference frames. So the way I learned the structure of this water bottle where the

Jeff Hawkins (46:05.920)

features are relative to each other, when I think about history or democracy or mathematics,

Jeff Hawkins (46:11.200)

the same basic underlying structure is happening. There's reference frames for where the knowledge

Jeff Hawkins (46:15.680)

that you're assigning things to. So in the book, I go through examples like mathematics

Lex Fridman (46:19.200)

and language and politics. But the evidence is very clear in the neuroscience. The same mechanism

Jeff Hawkins (46:25.920)

that we use to model this coffee cup, we're going to use to model high level thoughts.

Jeff Hawkins (46:30.160)

Your demise of humanity, whatever you want to think about.

Jeff Hawkins (46:34.160)

It's interesting to think about how different are the representations of those higher dimensional

Jeff Hawkins (46:38.960)

concepts, higher level concepts, how different the representation there is in terms of reference

Jeff Hawkins (46:45.680)

frames versus spatial. But the interesting thing, it's a different application, but it's the exact

Jeff Hawkins (46:52.080)

same mechanism. But isn't there some aspect to higher level concepts that they seem to be

Jeff Hawkins (46:59.680)

hierarchical? Like they just seem to integrate a lot of information into them. So is our physical

Jeff Hawkins (47:05.600)

objects. So take this water bottle. I'm not particular to this brand, but this is a Fiji

Jeff Hawkins (47:12.160)

water bottle and it has a logo on it. I use this example in my book, our company's coffee cup has

Jeff Hawkins (47:18.880)

a logo on it. But this object is hierarchical. It's got like a cylinder and a cap, but then it

Jeff Hawkins (47:25.520)

has this logo on it and the logo has a word, the word has letters, the letters have different

Jeff Hawkins (47:29.360)

features. And so I don't have to remember, I don't have to think about this. So I say,

Jeff Hawkins (47:33.840)

oh, there's a Fiji logo on this water bottle. I don't have to go through and say, oh, what is the

Jeff Hawkins (47:37.920)

Fiji logo? It's the F and I and the J and I, and there's a hibiscus flower. And, oh, it has the

Jeff Hawkins (47:43.920)

statement on it. I don't have to do that. I just incorporate all of that in some sort of hierarchical

Jeff Hawkins (47:47.760)

representation. I say, put this logo on this water bottle. And then the logo has a word

Lex Fridman (47:55.040)

and the word has letters, all hierarchical. All that stuff is big. It's amazing that the

Jeff Hawkins (47:59.520)

brain instantly just does all that. The idea that there's water, it's liquid and the idea that you

Jeff Hawkins (48:04.960)

can drink it when you're thirsty, the idea that there's brands and then there's like all of that

Jeff Hawkins (48:11.920)

information is instantly like built into the whole thing once you proceed. So I wanted to

Jeff Hawkins (48:17.120)

get back to your point about hierarchical representation. The world itself is hierarchical,

Jeff Hawkins (48:21.680)

right? And I can take this microphone in front of me. I know inside there's going to be some

Jeff Hawkins (48:25.200)

electronics. I know there's going to be some wires and I know there's going to be a little

Jeff Hawkins (48:28.080)

diaphragm that moves back and forth. I don't see that, but I know it. So everything in the world

Jeff Hawkins (48:33.920)

is hierarchical. You just go into a room. It's composed of other components. The kitchen has a

Jeff Hawkins (48:37.840)

refrigerator. The refrigerator has a door. The door has a hinge. The hinge has screws and pin.

Lex Fridman (48:43.200)

So anyway, the modeling system that exists in every cortical column learns the hierarchical

Jeff Hawkins (48:49.360)

structure of objects. So it's a very sophisticated modeling system in this grain of rice. It's hard

Jeff Hawkins (48:54.720)

to imagine, but this grain of rice can do really sophisticated things. It's got 100,000 neurons in

Jeff Hawkins (48:58.800)

it. It's very sophisticated. So that same mechanism that can model a water bottle or a coffee cup

Jeff Hawkins (49:07.440)

can model conceptual objects as well. That's the beauty of this discovery that this guy,

Jeff Hawkins (49:13.600)

Vernon Malmkastel, made many, many years ago, which is that there's a single cortical algorithm

Lex Fridman (49:18.720)

underlying everything we're doing. So common sense concepts and higher

Lex Fridman (49:23.840)

level concepts are all represented in the same way?

Jeff Hawkins (49:26.720)

They're set in the same mechanisms, yeah. It's a little bit like computers. All computers are

Jeff Hawkins (49:31.520)

universal Turing machines. Even the little teeny one that's in my toaster and the big one that's

Jeff Hawkins (49:37.520)

running some cloud server someplace. They're all running on the same principle. They can

Jeff Hawkins (49:41.680)

apply different things. So the brain is all built on the same principle. It's all about

Jeff Hawkins (49:46.080)

learning these structured models using movement and reference frames. And it can be applied to

Jeff Hawkins (49:53.120)

something as simple as a water bottle and a coffee cup. And it can be applied to thinking

Jeff Hawkins (49:56.400)

what's the future of humanity and why do you have a hedgehog on your desk? I don't know.

Jeff Hawkins (50:02.800)

Nobody knows. Well, I think it's a hedgehog. That's right. It's a hedgehog in the fog.

Jeff Hawkins (50:09.280)

It's a Russian reference. Does it give you any inclination or hope about how difficult

Lex Fridman (50:16.240)

it is to engineer common sense reasoning? So how complicated is this whole process?

Lex Fridman (50:21.840)

So looking at the brain, is this a marvel of engineering or is it pretty dumb stuff

Lex Fridman (50:28.640)

stuck on top of each other over? Can it be both? Can it be both, right?

Jeff Hawkins (50:35.600)

I don't know if it can be both because if it's an incredible engineering job, that means it's

Lex Fridman (50:43.040)

so evolution did a lot of work. Yeah, but then it just copied that.

Jeff Hawkins (50:48.320)

Yeah. Right. So as I said earlier, figuring out how to model something like a space is really hard

Lex Fridman (50:55.760)

and evolution had to go through a lot of trick. And these cells I was talking about,

Jeff Hawkins (50:59.760)

these grid cells and place cells, they're really complicated. This is not simple stuff.

Jeff Hawkins (51:03.040)

This neural tissue works on these really unexpected, weird mechanisms.

Lex Fridman (51:08.720)

But it did it. It figured it out. But now you could just make lots of copies of it.

Lex Fridman (51:13.120)

But then finding, yeah, so it's a very interesting idea that's a lot of copies

Jeff Hawkins (51:18.320)

of a basic mini brain. But the question is how difficult it is to find that mini brain

Jeff Hawkins (51:25.360)

that you can copy and paste effectively. Today, we know enough to build this.

Jeff Hawkins (51:33.920)

I'm sitting here with, I know the steps we have to go. There's still some engineering problems

Jeff Hawkins (51:37.920)

to solve, but we know enough. And this is not like, oh, this is an interesting idea. We have

Jeff Hawkins (51:43.760)

to go think about it for another few decades. No, we actually understand it pretty well in details.

Lex Fridman (51:48.160)

So not all the details, but most of them. So it's complicated, but it is an engineering problem.

Lex Fridman (51:55.360)

So in my company, we are working on that. We are basically a roadmap of how we do this.

Jeff Hawkins (52:01.360)

It's not going to take decades. It's a matter of a few years optimistically,

Lex Fridman (52:06.880)

but I think that's possible. It's, you know, complex things. If you understand them,

Lex Fridman (52:11.840)

you can build them. So in which domain do you think it's best to build them?

Jeff Hawkins (52:17.200)

Are we talking about robotics, like entities that operate in the physical world that are

Lex Fridman (52:23.440)

able to interact with that world? Are we talking about entities that operate in the digital world?

Jeff Hawkins (52:27.920)

Are we talking about something more like more specific, like it's done in the machine learning

Lex Fridman (52:33.840)

community where you look at natural language or computer vision? Where do you think is easiest?

Lex Fridman (52:41.120)

It's the first, it's the first two more than the third one, I would say.

Jeff Hawkins (52:46.560)

Again, let's just use computers as an analogy. The pioneers in computing, people like John

Jeff Hawkins (52:52.320)

Van Norman and Alan Turing, they created this thing, you know, we now call the universal

Lex Fridman (52:56.800)

Turing machine, which is a computer, right? Did they know how it was going to be applied?

Jeff Hawkins (53:00.800)

Where it was going to be used? Could they envision any of the future? No. They just said,

Jeff Hawkins (53:04.960)

this is like a really interesting computational idea about algorithms and how you can implement

Jeff Hawkins (53:11.120)

them in a machine. And we're doing something similar to that today. Like we are building this

Jeff Hawkins (53:18.400)

sort of universal learning principle that can be applied to many, many different things.

Lex Fridman (53:24.480)

But the robotics piece of that, the interactive...

Jeff Hawkins (53:27.600)

Okay. All right. Let's be just specific. You can think of this cortical column as

Lex Fridman (53:31.360)

what we call a sensory motor learning system. It has the idea that there's a sensor

Lex Fridman (53:35.120)

and then it's moving. That sensor can be physical. It could be like my finger

Lex Fridman (53:39.520)

and it's moving in the world. It could be like my eye and it's physically moving.

Jeff Hawkins (53:43.440)

It can also be virtual. So, it could be, an example would be, I could have a system that

Jeff Hawkins (53:50.160)

lives in the internet that actually samples information on the internet and moves by

Jeff Hawkins (53:55.360)

following links. That's a sensory motor system. Something that echoes the process of a finger

Lex Fridman (54:02.240)

moving along a cortical... But in a very, very loose sense. It's like,

Jeff Hawkins (54:06.640)

again, learning is inherently about discovering the structure of the world and discover the

Jeff Hawkins (54:10.720)

structure of the world, you have to move through the world. Even if it's a virtual world, even if

Jeff Hawkins (54:14.720)

it's a conceptual world, you have to move through it. It doesn't exist in one... It has some structure

Lex Fridman (54:20.480)

to it. So, here's a couple of predictions at getting what you're talking about.

Jeff Hawkins (54:27.040)

In humans, the same algorithm does robotics. It moves my arms, my eyes, my body.

Lex Fridman (54:34.560)

And so, in the future, to me, robotics and AI will merge. They're not going to be separate fields

Jeff Hawkins (54:40.000)

because the algorithms for really controlling robots are going to be the same algorithms we

Jeff Hawkins (54:45.200)

have in our brain, these sensory motor algorithms. Today, we're not there, but I think that's going

Jeff Hawkins (54:50.000)

to happen. But not all AI systems will have to be robotics. You can have systems that have very

Jeff Hawkins (54:58.880)

different types of embodiments. Some will have physical movements, some will not have physical

Jeff Hawkins (55:02.480)

movements. It's a very generic learning system. Again, it's like computers. The Turing machine,

Jeff Hawkins (55:08.560)

it doesn't say how it's supposed to be implemented, it doesn't tell you how big it is,

Jeff Hawkins (55:11.680)

it doesn't tell you what you can apply it to, but it's a computational principle.

Jeff Hawkins (55:15.440)

The cortical column equivalent is a computational principle about learning. It's about how you

Jeff Hawkins (55:20.640)

learn and it can be applied to a gazillion things. I think this impact of AI is going to be as large,

Jeff Hawkins (55:27.440)

if not larger, than computing has been in the last century, by far, because it's getting at

Jeff Hawkins (55:33.200)

a fundamental thing. It's not a vision system or a learning system. It's not a vision system or

Jeff Hawkins (55:37.600)

a hearing system. It is a learning system. It's a fundamental principle, how you learn the structure

Jeff Hawkins (55:41.600)

in the world, how you can gain knowledge and be intelligent. That's what the thousand brains says

Jeff Hawkins (55:46.400)

was going on. We have a particular implementation in our head, but it doesn't have to be like that

Jeff Hawkins (55:49.680)

at all. Do you think there's going to be some kind of impact? Okay, let me ask it another way.

Lex Fridman (55:56.800)

What do increasingly intelligent AI systems do with us humans in the following way? How hard is

Jeff Hawkins (56:05.360)

the human in the loop problem? How hard is it to interact? The finger on the coffee cup equivalent

Lex Fridman (56:13.040)

of having a conversation with a human being. How hard is it to fit into our little human world?

Jeff Hawkins (56:20.880)

I think it's a lot of engineering problems. I don't think it's a fundamental problem.

Lex Fridman (56:25.200)

I could ask you the same question. How hard is it for computers to fit into a human world?

Jeff Hawkins (56:28.880)

Right. That's essentially what I'm asking. How elitist are we as humans? We try to keep out

Jeff Hawkins (56:40.720)

systems. I don't know. I'm not sure that's the right question. Let's look at computers as an

Jeff Hawkins (56:48.240)

analogy. Computers are a million times faster than us. They do things we can't understand.

Jeff Hawkins (56:52.480)

Most people have no idea what's going on when they use computers. How do we integrate them

Jeff Hawkins (56:57.120)

in our society? Well, we don't think of them as their own entity. They're not living things.

Jeff Hawkins (57:04.160)

We don't afford them rights. We rely on them. Our survival as seven billion people or something

Jeff Hawkins (57:12.800)

like that is relying on computers now. Don't you think that's a fundamental problem

Lex Fridman (57:18.320)

that we see them as something we don't give rights to?

Jeff Hawkins (57:22.480)

Computers? Yeah, computers. Robots,

Jeff Hawkins (57:25.600)

computers, intelligence systems. It feels like for them to operate successfully,

Jeff Hawkins (57:29.920)

they would need to have a lot of the elements that we would start having to think about.

Lex Fridman (57:37.760)

Should this entity have rights? I don't think so. I think

Jeff Hawkins (57:42.560)

it's tempting to think that way. First of all, hardly anyone thinks that for computers today.

Jeff Hawkins (57:47.680)

No one says, oh, this thing needs a right. I shouldn't be able to turn it off. If I throw it

Jeff Hawkins (57:52.320)

in the trash can and hit it with a sledgehammer, it might form a criminal act. No one thinks that.

Lex Fridman (57:59.360)

Now we think about intelligent machines, which is where you're going.

Jeff Hawkins (58:05.600)

All of a sudden, you're like, well, now we can't do that. I think the basic problem we have here

Jeff Hawkins (58:10.080)

is that people think intelligent machines will be like us. They're going to have the same emotions

Jeff Hawkins (58:14.000)

as we do, the same feelings as we do. What if I can build an intelligent machine that absolutely

Jeff Hawkins (58:19.120)

could care less about whether it was on or off or destroyed or not? It just doesn't care. It's

Jeff Hawkins (58:23.040)

just like a map. It's just a modeling system. There's no desires to live. Nothing.

Lex Fridman (58:28.400)

Is it possible to create a system that can model the world deeply and not care

Jeff Hawkins (58:35.280)

about whether it lives or dies? Absolutely. No question about it.

Jeff Hawkins (58:38.640)

To me, that's not 100% obvious. It's obvious to me. We can debate it if we want.

Jeff Hawkins (58:43.920)

Where does your desire to live come from? It's an old evolutionary design. We could argue,

Jeff Hawkins (58:52.560)

does it really matter if we live or not? Objectively, no. We're all going to die eventually.

Jeff Hawkins (59:00.720)

Evolution makes us want to live. Evolution makes us want to fight to live. Evolution makes us want

Jeff Hawkins (59:05.840)

to care and love one another and to care for our children and our relatives and our family and so

Jeff Hawkins (59:11.840)

on. Those are all good things. They come about not because we're smart, because we're animals

Jeff Hawkins (59:18.880)

that grew up. The hummingbird in my backyard cares about its offspring. Every living thing

Jeff Hawkins (59:25.280)

in some sense cares about surviving. When we talk about creating intelligent machines,

Jeff Hawkins (59:30.720)

we're not creating life. We're not creating evolving creatures. We're not creating living

Jeff Hawkins (59:35.360)

things. We're just creating a machine that can learn really sophisticated stuff. That machine,

Jeff Hawkins (59:40.400)

it may even be able to talk to us. It's not going to have a desire to live unless somehow we put it

Jeff Hawkins (59:47.120)

into that system. Well, there's learning, right? The thing is... But you don't learn to want to

Jeff Hawkins (59:52.720)

live. It's built into you. It's part of your DNA. People like Ernest Becker argue,

Jeff Hawkins (59:59.600)

there's the fact of finiteness of life. The way we think about it is something we learned,