WEBVTT

1
00:00:00.000 --> 00:00:04.800
After many, many decades of people debating this, you might have figured out the reason why we dream.

2
00:00:04.800 --> 00:00:06.600
Yes, and it's a simple answer.

3
00:00:06.600 --> 00:00:13.300
So if you go blind, the visual cortex at the back of the brain gets taken over by hearing and by touch and by other things.

4
00:00:13.300 --> 00:00:17.500
In fact, our colleagues at Harvard did an experiment where they blindfolded and were only sighted people,

5
00:00:17.500 --> 00:00:20.500
and you could start seeing that takeover happening after 60 minutes.

6
00:00:20.500 --> 00:00:29.200
And that's when we realized, wow, the purpose of dreaming is to defend the visual territory from takeover from the other senses.

7
00:00:29.200 --> 00:00:36.200
For what fascinates me about brain plasticity and what I've devoted my career to is figuring out the way that we can be the sculptors of our own brains

8
00:00:36.200 --> 00:00:40.700
and how it gives us an opportunity to become the kind of person we'd like to be.

9
00:00:40.700 --> 00:00:41.700
And can we do that?

10
00:00:41.700 --> 00:00:45.700
Yes, here's the thing, your brain peaked the age of tube.

11
00:00:45.700 --> 00:00:50.700
Okay, so at the beginning, you've got fluid intelligence meaning you could learn anything.

12
00:00:50.700 --> 00:00:56.200
But now that you have grown up in this world, you've got crystallized intelligence meaning you know how to drive a car,

13
00:00:56.200 --> 00:01:01.200
you know how to operate a cell phone, you know how to run a business, and so your brain doesn't require as much change,

14
00:01:01.200 --> 00:01:05.200
which means that the structure of the brain is always degenerating.

15
00:01:05.200 --> 00:01:10.200
So what are the set of actions that will fundamentally change my brain and make me that type of person

16
00:01:10.200 --> 00:01:13.700
who's motivated and disciplined and who has high agency and attacks the world?

17
00:01:13.700 --> 00:01:17.700
So this is something I've started my lab for decades now, and the key is that...

18
00:01:17.700 --> 00:01:21.700
And what about AI and the social media debate as it relates to brain development?

19
00:01:21.700 --> 00:01:25.700
Well, I happen to be a cyber optimist for young people.

20
00:01:25.700 --> 00:01:30.700
I think it's going to make them much smarter than the generation that came before, and here's why.

21
00:01:30.700 --> 00:01:32.700
Interesting.

22
00:01:32.700 --> 00:01:35.700
Guys, I've got a favor to ask before this episode begins.

23
00:01:35.700 --> 00:01:41.700
The algorithm, if you follow a show, will deliver you the best episodes from that show very prominently in your feed.

24
00:01:41.700 --> 00:01:45.700
So when we have our best episodes on this show, the most shared episodes, the most rated episodes,

25
00:01:45.700 --> 00:01:50.700
I would love you to know, and the simple way for you to know that is to hit that follow button.

26
00:01:50.700 --> 00:01:54.700
But also, it's the simple, easy, free thing that you can do to help us make this show better.

27
00:01:54.700 --> 00:01:58.700
I would be hugely grateful if you could take a minute on the app you're listening to this one right now,

28
00:01:58.700 --> 00:02:00.700
and hit that follow button.

29
00:02:00.700 --> 00:02:03.700
Thank you so, so, so much.

30
00:02:03.700 --> 00:02:14.700
Dr. David Eagleman, what made you so fascinated about the brain?

31
00:02:14.700 --> 00:02:17.700
And why should everybody listening be fascinated about the brain as well?

32
00:02:17.700 --> 00:02:18.700
Here's what I think it is.

33
00:02:18.700 --> 00:02:22.700
When I was eight years old, I fell off of the roof of the house that was in our construction,

34
00:02:22.700 --> 00:02:27.700
and I fell 12 feet and broke my nose on the floor below.

35
00:02:27.700 --> 00:02:29.700
But the whole thing seemed to take a long time.

36
00:02:29.700 --> 00:02:35.700
I did the calculation and figured out that it only took 0.6 of a second to get from the top to the bottom,

37
00:02:35.700 --> 00:02:38.700
and I couldn't figure out why it seemed to have taken so long.

38
00:02:38.700 --> 00:02:41.700
So I think that got me really interested in perception,

39
00:02:41.700 --> 00:02:45.700
and the machinery by which we view the world and taken in,

40
00:02:45.700 --> 00:02:50.700
and what is actually real versus what's a construction of the brain,

41
00:02:50.700 --> 00:02:54.700
and that's what I've devoted my career to is figuring out how the brain,

42
00:02:54.700 --> 00:02:58.700
which is locked inside the skull, it's about three pounds,

43
00:02:58.700 --> 00:03:03.700
how it constructs this model of the world and which things we can take as reality,

44
00:03:03.700 --> 00:03:05.700
which things we shouldn't.

45
00:03:05.700 --> 00:03:08.700
I think most people don't even know they have a, there's a brain there almost.

46
00:03:08.700 --> 00:03:12.700
It sounds like a strange thing to say, but we've never really, most of us haven't really seen our own brains at all.

47
00:03:12.700 --> 00:03:15.700
We've never been able to touch our own brains at all.

48
00:03:15.700 --> 00:03:22.700
So it's, it's easy to fall into the trap of thinking that everything I experience is true and is reality.

49
00:03:22.700 --> 00:03:26.700
So I'm wondering how a deeper understanding of all this stuff can help me live a better life.

50
00:03:26.700 --> 00:03:31.700
Yeah, one of the things that I started running about years ago is that I think we're not,

51
00:03:31.700 --> 00:03:35.700
I think we often think of ourselves as individuals,

52
00:03:35.700 --> 00:03:42.700
meaning not divisible into other things, but really you are a team of rivals.

53
00:03:42.700 --> 00:03:45.700
You've got all these neural networks that have different drives,

54
00:03:45.700 --> 00:03:47.700
making different suggestions to you.

55
00:03:47.700 --> 00:03:49.700
What's a neural network?

56
00:03:49.700 --> 00:03:53.700
So in the brain, you've got 86 billion cells called neurons.

57
00:03:53.700 --> 00:03:57.700
And these are communicating with each other at a blindingly fast rate.

58
00:03:57.700 --> 00:03:59.700
Many of these cells are hooked up in networks.

59
00:03:59.700 --> 00:04:03.700
So they're, you know, this guy is talking to this guy and this guy and they're all in particular networks.

60
00:04:03.700 --> 00:04:07.700
The thing is you can actually get competing networks.

61
00:04:07.700 --> 00:04:12.700
So for example, Stephen, if I drop some chocolate chip cookies in front of you,

62
00:04:12.700 --> 00:04:13.700
party brain wants to eat it.

63
00:04:13.700 --> 00:04:14.700
It's a good energy source.

64
00:04:14.700 --> 00:04:15.700
Party brain says, don't eat it.

65
00:04:15.700 --> 00:04:16.700
I'll gain weight.

66
00:04:16.700 --> 00:04:18.700
Party says, okay, I'll eat one, but I'll go to the gym tonight.

67
00:04:18.700 --> 00:04:22.700
The point is you are arguing with yourself.

68
00:04:22.700 --> 00:04:24.700
You are conflicted.

69
00:04:24.700 --> 00:04:32.700
This is what makes humans so interesting is that we have all these voices trying to drive us to different conclusions about our behavior.

70
00:04:32.700 --> 00:04:39.700
So the way that you are ship of state moves depends on the vote of the neural parliament at any time.

71
00:04:39.700 --> 00:04:44.700
So understanding this, I think, is really critical to navigating our own lives

72
00:04:44.700 --> 00:04:48.700
because all of us do things where retrospectively we regret it.

73
00:04:48.700 --> 00:04:54.700
We say, I shouldn't have eaten that whole bag of chips or done the alcohol or the drugs or whatever.

74
00:04:54.700 --> 00:04:57.700
Everybody has regrets all the time with things.

75
00:04:57.700 --> 00:05:02.700
And it's because you have different voices in charge at different times.

76
00:05:02.700 --> 00:05:03.700
Okay.

77
00:05:03.700 --> 00:05:08.700
Part of what this leads to is what we call the Ulysses contract.

78
00:05:08.700 --> 00:05:14.700
So a Ulysses contract is where you do something now to prevent yourself from behaving badly in the near future.

79
00:05:14.700 --> 00:05:21.700
Just as an example, you know, when people go to alcoholic synonymous, the first thing they're told is clear all the alcohol out of the house.

80
00:05:21.700 --> 00:05:26.700
Because even if you feel like, look, I'm in a moment of sober reflection, I don't want to ever drink again.

81
00:05:26.700 --> 00:05:33.700
If you have alcohol in the house, you're going to bust into that cabinet at some point on a festive Saturday night or a lonely Sunday night or whatever.

82
00:05:33.700 --> 00:05:40.700
So what you do is you constrain your future behavior by setting things up in the right way.

83
00:05:40.700 --> 00:05:43.700
So your future, the future you can't behave badly.

84
00:05:43.700 --> 00:05:46.700
We naively think, okay, well, I know who I am.

85
00:05:46.700 --> 00:05:49.700
I'm just one person, but you're not.

86
00:05:49.700 --> 00:05:54.700
And then our different circumstances, you're tempted by different things and you'll do different kinds of behavior.

87
00:05:54.700 --> 00:06:04.700
So having a sense of what's going on under the hood gives us an opportunity to be more closely aligned with the kind of person we would like to be.

88
00:06:04.700 --> 00:06:12.700
Because it feels like there's just one, well, I do argue with myself in my head sometimes, but it feels like there is just one me.

89
00:06:12.700 --> 00:06:15.700
And so when I hear that voice say, Steve, you should have that cookie and it's one a.m.

90
00:06:15.700 --> 00:06:20.700
And then the other voice says, no, you shouldn't. I think it's kind of the same person just tussling with himself.

91
00:06:20.700 --> 00:06:25.700
Right. Well, but that tussling with himself implies different political parties that are all battling it out.

92
00:06:25.700 --> 00:06:29.700
You know, when you look at a parliament, you've got all these political parties that all love their country.

93
00:06:29.700 --> 00:06:32.700
They just have different ideas of how to steer it.

94
00:06:32.700 --> 00:06:36.700
And this is what's going on in the brain all the time.

95
00:06:36.700 --> 00:06:41.700
So what does one do about that? How do I make? Do I have to make a Lissia's contract?

96
00:06:41.700 --> 00:06:46.700
I think it's very useful to make that sort of thing, but also just understanding what's.

97
00:06:46.700 --> 00:06:50.700
Part of the, you know, there was this Greek admonition to know thyself.

98
00:06:50.700 --> 00:06:54.700
This was a sign they had in various places, various temples and stuff.

99
00:06:54.700 --> 00:06:57.700
But I think that becomes know thyselves.

100
00:06:57.700 --> 00:07:03.700
And the better we know ourselves, the more we can get rid of the illusion that we are one person.

101
00:07:03.700 --> 00:07:09.700
Because all any of us need to do is look back on our behavior to say, oh, yeah, in some circumstances, I would do that.

102
00:07:09.700 --> 00:07:11.700
Other circumstances, I think it's a terrible idea.

103
00:07:11.700 --> 00:07:15.700
So this is all to the goal of understanding who you are.

104
00:07:15.700 --> 00:07:20.700
One of the big misconceptions about the brain that people have gone through their life believing.

105
00:07:20.700 --> 00:07:21.700
I mean, that's one of them.

106
00:07:21.700 --> 00:07:29.700
Something that is true that kind of could fall in place of that is just this fundamental idea that our brains are plastic or adaptable.

107
00:07:29.700 --> 00:07:34.700
Because when I found out that I could change my brain by what I do, I found that to be really, really inspiring.

108
00:07:34.700 --> 00:07:36.700
Yes, that's exactly right.

109
00:07:36.700 --> 00:07:41.700
So brain plasticity, if someone hasn't heard that term before, it sounds like a weird term.

110
00:07:41.700 --> 00:07:47.700
But the reason it came about a hundred years ago is because the great psychologist William James pointed out that,

111
00:07:47.700 --> 00:07:54.700
you know, if you take a piece of plastic, what we like about that material that we call plastic is that you can mold it into a shape and it'll hold that shape.

112
00:07:54.700 --> 00:07:55.700
And that's what your brain does.

113
00:07:55.700 --> 00:08:02.700
So if I ask you the name of your third grade teacher, you can remember that name, even though it's been a long time,

114
00:08:02.700 --> 00:08:07.700
because your neural networks changed and held on to that piece of information.

115
00:08:07.700 --> 00:08:08.700
Okay.

116
00:08:08.700 --> 00:08:12.700
Well, our whole lives, our brains are changing every moment.

117
00:08:12.700 --> 00:08:16.700
So now we have certain doors that close at different times.

118
00:08:16.700 --> 00:08:21.700
So just as an example, you need to learn language in the first several years of your life.

119
00:08:21.700 --> 00:08:26.700
If you don't learn language, you can never get the concept of language.

120
00:08:26.700 --> 00:08:27.700
Your brain will never figure that out.

121
00:08:27.700 --> 00:08:29.700
You're not saying you can't learn a new language as an adult.

122
00:08:29.700 --> 00:08:35.700
You're saying the concept of language, the concept that I can name things and I can ask for things and so on.

123
00:08:35.700 --> 00:08:37.700
That never clicks in the brain.

124
00:08:37.700 --> 00:08:45.700
For example, in Romania, at the fall of Chichescu, there were tens of thousands of kids in the orphanages because their parents had been killed.

125
00:08:45.700 --> 00:08:47.700
It was too many kids.

126
00:08:47.700 --> 00:08:52.700
And so the staff there said, look, the kids will get clingy if you pay too much attention to them.

127
00:08:52.700 --> 00:08:53.700
So here's what we're going to do.

128
00:08:53.700 --> 00:08:54.700
We're going to feed the kids.

129
00:08:54.700 --> 00:08:55.700
We're not going to hold them.

130
00:08:55.700 --> 00:08:57.700
We're not going to talk to them.

131
00:08:57.700 --> 00:09:01.700
And all these children grew up with real cognitive deficits as a result.

132
00:09:01.700 --> 00:09:04.700
Here's the thing about brain plasticity.

133
00:09:04.700 --> 00:09:09.700
Human beings have a similar brain to all our neighbors in the animal kingdom.

134
00:09:09.700 --> 00:09:12.700
If you compare our brain to a horse brain, a dog brain, anything like that.

135
00:09:12.700 --> 00:09:14.700
It's the same general structures and stuff.

136
00:09:15.700 --> 00:09:19.700
But what we have is much more of the wrinkly outer bit called the cortex.

137
00:09:19.700 --> 00:09:21.700
It's the outer three millimeters.

138
00:09:21.700 --> 00:09:24.700
And maybe we'll come back to why that matters so much.

139
00:09:24.700 --> 00:09:28.700
But the other thing that Mother Nature tweaked with us, it's small genetic tweaks.

140
00:09:28.700 --> 00:09:33.700
But we have much more plasticity, adaptability, such that.

141
00:09:33.700 --> 00:09:37.700
When a horse drops into the world, it's doing the same thing that horses did a hundred thousand years ago.

142
00:09:37.700 --> 00:09:39.700
It's just, you know, eat, mate.

143
00:09:39.700 --> 00:09:44.700
But when a human drops in the world, we learn everything that's happened before us.

144
00:09:44.700 --> 00:09:46.700
And then we springboard off the top of that.

145
00:09:46.700 --> 00:09:50.700
So we living in the 21st century, we say, oh, great.

146
00:09:50.700 --> 00:09:53.700
You know, physics math, this, that are great.

147
00:09:53.700 --> 00:09:55.700
We got everything that's happened before us.

148
00:09:55.700 --> 00:09:56.700
Now let's do our own thing.

149
00:09:56.700 --> 00:10:01.700
And that's what's so special about the plasticity of the human brain, the adaptability of it.

150
00:10:01.700 --> 00:10:08.700
The downside, the gamble is that Mother Nature drops human brains into the world kind of half-baked.

151
00:10:08.700 --> 00:10:11.700
And we then get to absorb everything.

152
00:10:11.700 --> 00:10:15.700
But in the rare circumstance where you're not getting the right input,

153
00:10:15.700 --> 00:10:19.700
then that ends up really in trouble because it's only half-baked.

154
00:10:19.700 --> 00:10:23.700
So when it comes to language, we can learn multiple languages when we're young.

155
00:10:23.700 --> 00:10:26.700
That's very easy, but it gets harder and harder as that goes along.

156
00:10:26.700 --> 00:10:28.700
And various other things become harder.

157
00:10:28.700 --> 00:10:29.700
And here's why.

158
00:10:29.700 --> 00:10:35.700
It's because, I mentioned this earlier, but the job of the brain is to make a model of the world

159
00:10:35.700 --> 00:10:38.700
so it can operate within it.

160
00:10:38.700 --> 00:10:42.700
So for example, you're an entrepreneur and you love doing business.

161
00:10:42.700 --> 00:10:43.700
So you get it.

162
00:10:43.700 --> 00:10:46.700
Okay, here's how, you know, here's a structure business.

163
00:10:46.700 --> 00:10:47.700
Here's how you hire well.

164
00:10:47.700 --> 00:10:49.700
Here's how you set up a board well.

165
00:10:49.700 --> 00:10:54.700
You're doing everything because you've got a really rich internal model of how to structure a business.

166
00:10:54.700 --> 00:10:58.700
That's what the brain wants to do is get that stuff right.

167
00:10:58.700 --> 00:11:04.700
As a result, if you suddenly ended up, you know, taking a trip to Mars,

168
00:11:04.700 --> 00:11:08.700
and there's a whole very different society there that does businesses very differently,

169
00:11:08.700 --> 00:11:11.700
you would have to relearn stuff really quickly.

170
00:11:11.700 --> 00:11:13.700
So here's the thing.

171
00:11:13.700 --> 00:11:18.700
You went from having a brain that had high fluid intelligence

172
00:11:18.700 --> 00:11:21.700
to now having a brain that has high crystallized intelligence.

173
00:11:21.700 --> 00:11:24.700
What that means is at the beginning, you can learn anything.

174
00:11:24.700 --> 00:11:27.700
You could learn any language you could have dropped into any era.

175
00:11:27.700 --> 00:11:30.700
You could have dropped into 13th century Japan when I was young.

176
00:11:30.700 --> 00:11:34.700
When you were young, when you were a baby, if you had dropped out of the womb in, you know,

177
00:11:34.700 --> 00:11:38.700
10th century Mongolia, you would have said like, okay, cool, learn a language.

178
00:11:38.700 --> 00:11:41.700
You would be a 10th century Mongolian.

179
00:11:41.700 --> 00:11:46.700
But as it happens, you dropped into this era, you know, a certain place and time,

180
00:11:46.700 --> 00:11:48.700
a neighborhood and culture and family.

181
00:11:48.700 --> 00:11:49.700
And so you learn that.

182
00:11:49.700 --> 00:11:52.700
That's who you become is that person.

183
00:11:52.700 --> 00:11:55.700
We often think that plasticity diminishes as you age.

184
00:11:55.700 --> 00:11:57.700
But it's not simply that it's diminishing.

185
00:11:57.700 --> 00:12:02.700
It's that you are getting the right answers about how to operate in the world.

186
00:12:02.700 --> 00:12:05.700
And so you don't have to change as much.

187
00:12:05.700 --> 00:12:07.700
Your brain doesn't require as much change.

188
00:12:07.700 --> 00:12:09.700
What if I want to change?

189
00:12:09.700 --> 00:12:10.700
Yes.

190
00:12:10.700 --> 00:12:12.700
So it turns out you still can change.

191
00:12:12.700 --> 00:12:17.700
That's the key is that the reason brains change us less is because they don't have to.

192
00:12:17.700 --> 00:12:21.700
But when things get upside down, just as one example,

193
00:12:21.700 --> 00:12:24.700
everything about the pandemic really stunk.

194
00:12:24.700 --> 00:12:32.700
Except for one thing, I think the tiny silver lining is that all of us had to reassess,

195
00:12:32.700 --> 00:12:34.700
oh my gosh, wait, how is the world working?

196
00:12:34.700 --> 00:12:36.700
I thought I knew how the world worked.

197
00:12:36.700 --> 00:12:39.700
But now I don't know if there's going to be toilet paper at the store.

198
00:12:39.700 --> 00:12:41.700
I don't know if the bank's going to be open.

199
00:12:41.700 --> 00:12:43.700
I don't know if I can get coffee at the coffee shop.

200
00:12:43.700 --> 00:12:44.700
Like everything was different.

201
00:12:44.700 --> 00:12:50.700
As awful as it was, it's really useful to challenge your internal model of the world

202
00:12:50.700 --> 00:12:52.700
and get to do that as an adult.

203
00:12:52.700 --> 00:12:54.700
We don't usually get to.

204
00:12:54.700 --> 00:12:57.700
So if I want to change, what would you recommend that I do?

205
00:12:57.700 --> 00:13:01.700
If I want to change who I am, so I'm stubborn, I'm not motivated.

206
00:13:01.700 --> 00:13:03.700
And I want to be a different person.

207
00:13:03.700 --> 00:13:04.700
The key is challenge.

208
00:13:04.700 --> 00:13:06.700
The key is seeking challenge.

209
00:13:06.700 --> 00:13:12.700
So it turns out that where we always want to be is in between the levels of frustrating but achievable.

210
00:13:12.700 --> 00:13:14.700
And you want to take on new tasks.

211
00:13:14.700 --> 00:13:17.700
You want to seek novelty to find yourself in that zone.

212
00:13:18.700 --> 00:13:21.700
And push yourself to do things that you just haven't done before.

213
00:13:21.700 --> 00:13:25.700
And one of the things that's so wonderful about the modern world,

214
00:13:25.700 --> 00:13:28.700
you know, everyone's got complaints about the internet and social media and stuff like that.

215
00:13:28.700 --> 00:13:34.700
But the good news is it exposes you to so much more than you ever even knew was out there.

216
00:13:34.700 --> 00:13:38.700
The key is to actively seek those challenges and seek new things

217
00:13:38.700 --> 00:13:41.700
and seek to become expert in various sorts of fields.

218
00:13:41.700 --> 00:13:45.700
And I think the key is that once you become good at something,

219
00:13:45.700 --> 00:13:48.700
you have to drop that and take on something you're not good at.

220
00:13:48.700 --> 00:13:51.700
This is the best thing you can do for your brain.

221
00:13:51.700 --> 00:13:56.700
The reason is because what you're doing is you're constantly building new roadways and pathways in the brain.

222
00:13:56.700 --> 00:14:01.700
There's a study that's been going on for decades now called the Religious Order study

223
00:14:01.700 --> 00:14:07.700
where a bunch of Catholic nuns agreed to donate their brains for autopsy when they passed away.

224
00:14:07.700 --> 00:14:11.700
What the researchers discovered when they look at the brain carefully

225
00:14:11.700 --> 00:14:15.700
is that some fraction of these nuns had Alzheimer's disease.

226
00:14:15.700 --> 00:14:20.700
Their brains were physically degenerating with the ravages of the dementia.

227
00:14:20.700 --> 00:14:25.700
But they didn't show any of the cognitive deficits that one normally has.

228
00:14:25.700 --> 00:14:28.700
They didn't seem to be having memory problems and so on.

229
00:14:28.700 --> 00:14:33.700
It turns out it's because all these nuns lived in these convents till the day they died.

230
00:14:33.700 --> 00:14:37.700
They had social challenges and they had fights with their fellow sisters

231
00:14:37.700 --> 00:14:41.700
and they played games with their fellow sisters and they had chores and responsibilities

232
00:14:41.700 --> 00:14:43.700
and they were doing stuff.

233
00:14:43.700 --> 00:14:47.700
What that means is even as the tissue, the brain tissue was physically degenerating,

234
00:14:47.700 --> 00:14:51.700
they were making new roadways and bridges all the time.

235
00:14:51.700 --> 00:14:54.700
And so that's what kept them cognitively healthy.

236
00:14:54.700 --> 00:14:56.700
We call that cognitive reserve.

237
00:14:56.700 --> 00:15:01.700
Contrast this with people who retire at 65 and they go home and they watch television

238
00:15:01.700 --> 00:15:04.700
and their social circle shrink and so on.

239
00:15:04.700 --> 00:15:08.700
That's when you've really got concerns because you're not building the new pathways.

240
00:15:08.700 --> 00:15:14.700
Is that data to support that when you retire if you retire early or if you retire in your 60s,

241
00:15:14.700 --> 00:15:19.700
it increases your probability of an earlier death or cognitive decline?

242
00:15:19.700 --> 00:15:23.700
Almost certainly with cognitive decline because you're just not getting the challenge.

243
00:15:23.700 --> 00:15:27.700
At that point you're just coasting on your internal model.

244
00:15:27.700 --> 00:15:32.700
It's tragic but what happens often is that people's hearing gets worse and so by the time they retire

245
00:15:32.700 --> 00:15:37.700
let's say in the mid 60s, it's not really that fun for them to go out to parties and restaurants anymore

246
00:15:37.700 --> 00:15:41.700
because they can't quite hear and so they're all these converging reasons

247
00:15:41.700 --> 00:15:43.700
why their social lives shrink.

248
00:15:43.700 --> 00:15:48.700
But it turns out social life is one of the most important things that we can do for our brains

249
00:15:48.700 --> 00:15:54.700
because there's an expression we sometimes use in neuroscience which is that nothing is as hard for the brain as other people

250
00:15:54.700 --> 00:15:59.700
because you never know what the other person is going to say and do and how they'll react emotionally and so on.

251
00:15:59.700 --> 00:16:06.700
You're constantly on your toes with other people and if you're not doing that anymore that ends up being a problem.

252
00:16:06.700 --> 00:16:08.700
Interesting.

253
00:16:08.700 --> 00:16:16.700
I'm 33 years old so if you were to pull up where my brain is on like a graph of decline,

254
00:16:16.700 --> 00:16:21.700
is it the case that I should be doing as much as I can now to build as many pathways as I can

255
00:16:21.700 --> 00:16:26.700
so that when I'm 80, my decline sort of levels out in a better place?

256
00:16:26.700 --> 00:16:31.700
Oh yeah, for sure, but this is true for many reasons actually.

257
00:16:31.700 --> 00:16:39.700
The truth is your brain peaked at two at the age of two because that's when you get the most connections between neurons,

258
00:16:39.700 --> 00:16:41.700
between these cells in the brain.

259
00:16:41.700 --> 00:16:46.700
At first you're born with these 86 billion neurons and they're connecting, connecting, connecting.

260
00:16:46.700 --> 00:16:50.700
And it finally becomes like an overgrown garden at the age of two and from there you're pruning,

261
00:16:50.700 --> 00:16:53.700
from there you're taking connections away.

262
00:16:53.700 --> 00:17:00.700
Now it happens that that's not a bad thing, that's a good thing because that's how you're resonating with the world that you are in.

263
00:17:00.700 --> 00:17:12.700
21st century London and LA versus 10th century Mongolia because you're just strengthening those pathways that resonate and you're getting rid of everything else.

264
00:17:12.700 --> 00:17:18.700
But over time your brain cells die, every time you hit your head on something or whatever your brain cells are going down.

265
00:17:19.700 --> 00:17:26.700
So in that sense you've peaked, but your crystallized intelligence that you've been building your whole life, that keeps going.

266
00:17:26.700 --> 00:17:29.700
And you'll have decades ahead of you where you can start doing stuff.

267
00:17:29.700 --> 00:17:36.700
But yes, the reason to learn everything you can is because all that stuff cashes out at various points in your life.

268
00:17:36.700 --> 00:17:43.700
When you're starting your next business or you're wanting to do the next great thing where you're surfing the web of AI,

269
00:17:43.700 --> 00:17:49.700
you'll say, oh, I learned this thing when I was 16, I learned this thing when I was 22 and these are paying off now.

270
00:17:49.700 --> 00:17:58.700
I think I heard Andrew Heuberman say that one of the most fascinating discoveries of the last century is a particular part of the brain called the anterior mid-singular cortex.

271
00:17:58.700 --> 00:18:04.700
And it links to what you were saying a second ago about challenge and doing things that are difficult.

272
00:18:04.700 --> 00:18:05.700
Yeah.

273
00:18:05.700 --> 00:18:09.700
It turns out that area of the brain is involved and other networks as well.

274
00:18:09.700 --> 00:18:18.700
Because when you're doing something new and challenging and difficult, you have stress and anxiety, your whole brain is active.

275
00:18:18.700 --> 00:18:24.700
Let's say I measured your brain, even with something like EEG, Electroencephalography, that's where I stick electrodes on the outside.

276
00:18:24.700 --> 00:18:26.700
Let's say I measure your brain in my brain.

277
00:18:26.700 --> 00:18:31.700
We're doing something that let's say you're an expert at what something you're really good at.

278
00:18:31.700 --> 00:18:32.700
Juggling.

279
00:18:32.700 --> 00:18:34.700
Let's go for juggling.

280
00:18:34.700 --> 00:18:37.700
Okay, let's say you're an expert juggler, so I've never juggled.

281
00:18:37.700 --> 00:18:40.700
Okay, if we're both juggling, you're going to be much better than I am.

282
00:18:40.700 --> 00:18:43.700
But your brain will be less active.

283
00:18:43.700 --> 00:18:45.700
You won't have as much activity in your brain.

284
00:18:45.700 --> 00:18:48.700
All my brain is on fire with activity because why?

285
00:18:48.700 --> 00:18:51.700
I'm trying to figure out, okay, where do I put my hand?

286
00:18:51.700 --> 00:18:52.700
How do I throw this?

287
00:18:52.700 --> 00:18:56.700
So when I'm in novice at something, my brain is using much more activity.

288
00:18:57.700 --> 00:19:02.700
Not just the anterior, maybe singular, but tons of activity all over because I'm trying to figure out the rules.

289
00:19:02.700 --> 00:19:03.700
I'm trying to figure out what's going on.

290
00:19:03.700 --> 00:19:06.700
You as an expert, you know, you got it.

291
00:19:06.700 --> 00:19:07.700
You don't need to burn much activity.

292
00:19:07.700 --> 00:19:13.700
This is what the brain's goal is, is to say, hey, once I've practiced something a lot, once I get something about the world,

293
00:19:13.700 --> 00:19:17.700
I'm going to burn it deeper and deeper into the circuitry, so I don't have to burn a lot of energy on it.

294
00:19:17.700 --> 00:19:20.700
On this part of the brain, the anterior and the singular cortex,

295
00:19:20.700 --> 00:19:24.700
the anterior human was saying it's larger in people that do things that they basically don't want to do.

296
00:19:24.700 --> 00:19:25.700
Hard things.

297
00:19:25.700 --> 00:19:28.700
If you spend your life doing things you don't want to do, then it happens to be bigger.

298
00:19:28.700 --> 00:19:31.700
And so people have now thought of this part of the brain, almost like the willpower muscle.

299
00:19:31.700 --> 00:19:37.700
Because for some reason, those that are doing hard things have big ones and those that are not have smaller ones.

300
00:19:37.700 --> 00:19:40.700
I mean, it wouldn't be so much the willpower muscle.

301
00:19:40.700 --> 00:19:45.700
It would be some indication retrospectively of how hard you have worked.

302
00:19:45.700 --> 00:19:49.700
Look, the fact is you can see changes in brain size with lots of things.

303
00:19:49.700 --> 00:19:50.700
I'll give you an example.

304
00:19:50.700 --> 00:19:57.700
If you are a pianist, if you play piano, then we can actually see physical changes in your motor cortex.

305
00:19:57.700 --> 00:20:02.700
This is the part of the brain essentially underneath where you would wear headphones for those who are looking visually.

306
00:20:02.700 --> 00:20:03.700
It's this red part here.

307
00:20:03.700 --> 00:20:10.700
You actually get a bigger loop of tissue here than you do in a normal brain.

308
00:20:10.700 --> 00:20:15.700
Why? Because you're doing so much fine motor activity with your fingers with both hands.

309
00:20:16.700 --> 00:20:22.700
In contrast, if you're a violinist, you're only really doing that kind of detail activity with one hand.

310
00:20:22.700 --> 00:20:23.700
The other hand is just bowing.

311
00:20:23.700 --> 00:20:28.700
And so you only get that activity here in one half of the brain for violinists.

312
00:20:28.700 --> 00:20:33.700
So I can look at a brain and tell you, hey, is the person a pianist or violinist or neither?

313
00:20:33.700 --> 00:20:40.700
I can tell just by looking at the visual cortex because you see changes in the brain based on what you do.

314
00:20:40.700 --> 00:20:45.700
For example, jugglers, people who play music, even you can tell it's with medical students who study for final exams.

315
00:20:45.700 --> 00:20:51.700
You actually see changes in the distribution of their cortex. Why would it be getting bigger?

316
00:20:51.700 --> 00:20:55.700
The reason is the brain is devoting more real estate to that.

317
00:20:55.700 --> 00:21:02.700
In this case, let's say we're talking about fingers on a piano or a violin, the brain is devoting more, there's more relevance to that.

318
00:21:02.700 --> 00:21:07.700
And so it more real estate so that you can do it better in the future.

319
00:21:07.700 --> 00:21:08.700
Exactly.

320
00:21:08.700 --> 00:21:13.700
The thing about the cortex, this wrinkly outer part is that it is a one trick pony.

321
00:21:13.700 --> 00:21:22.700
This is often overlooked because even this brain that I'm holding here is color coded so that we think, oh, okay, that's clear label this, that's clear label that and so on.

322
00:21:22.700 --> 00:21:25.700
But in fact, it's all the same stuff and it can change.

323
00:21:25.700 --> 00:21:34.700
So for instance, if you are born blind, then this area that we normally call the visual cortex gets taken over by the rest of the brain.

324
00:21:34.700 --> 00:21:40.700
If you're born deaf, then this part that we call the auditory cortex gets taken over it gets devoted to other tasks.

325
00:21:40.700 --> 00:21:43.700
And so this whole system is very, very fluid.

326
00:21:43.700 --> 00:21:58.700
And this is what fascinates me about brain plasticity is the way that we can be the sculptors of our own brains because we can devote ourselves to particular things and have the brains real estate get involved in that.

327
00:21:58.700 --> 00:22:08.700
So if I was currently someone that couldn't get out of bed, I didn't have a lot of discipline or motivation and I wasn't very good at committing myself to hard things.

328
00:22:08.700 --> 00:22:21.700
With everything you know about the brain, is it possible to take a set of actions that will fundamentally change my brain and make me that type of person who runs marathons, who does hard things, who's motivated and disciplines and who has high agency and attacks the world.

329
00:22:21.700 --> 00:22:22.700
Yes.

330
00:22:22.700 --> 00:22:31.700
But it's much more than simply resolve because I mean, just looking at New Year's resolutions, yeah, by February, most people have dropped most of them.

331
00:22:31.700 --> 00:22:37.700
So it's really a psychology problem about figuring out, okay, what are the things that motivate me?

332
00:22:37.700 --> 00:22:44.700
So let's say you want to become a marathon runner, you've got that distant dream, you figure out like what actually motivates me in the short term?

333
00:22:44.700 --> 00:22:49.700
Who am I trying to impress? What am I trying to accomplish in my life?

334
00:22:49.700 --> 00:23:03.700
How can I structure things like this Ulysses contract that I talked about earlier where I'm actually locking myself into a contract like, you know, I call Bob and I say, I will meet you every morning at 7 and we're going to run until we drop.

335
00:23:03.700 --> 00:23:09.700
Like once I've committed to those sorts of things, that's how you set things up so that you do the right thing.

336
00:23:09.700 --> 00:23:14.700
I'm going to have a cycle right because then my brain will adapt and then presumably that will make it easier for me to run.

337
00:23:14.700 --> 00:23:15.700
Yeah.

338
00:23:15.700 --> 00:23:17.700
And then I'll run more and then my brain will adapt.

339
00:23:17.700 --> 00:23:18.700
That's right.

340
00:23:18.700 --> 00:23:19.700
And the cycle continues.

341
00:23:19.700 --> 00:23:23.700
And it's not just your brain, of course, in this case, it's your body, you're getting better, you're getting stronger, you don't get as out of breath.

342
00:23:23.700 --> 00:23:25.700
And so all these things help exactly.

343
00:23:25.700 --> 00:23:35.700
But in order to keep the cycle going, you need to figure out what is spinning this flywheel and what are the all the other things in your life, whether good motivations are bad.

344
00:23:35.700 --> 00:23:36.700
It doesn't matter.

345
00:23:36.700 --> 00:23:39.700
You just figure out what it is that you can do to get there.

346
00:23:39.700 --> 00:23:45.700
All that's on physical exercises that are particularly good for the brain from what you've understood.

347
00:23:45.700 --> 00:23:48.700
The general story is exercise is really important for the brain.

348
00:23:48.700 --> 00:23:56.700
I'll give you just one example of that, which is there's still this debate going on about whether we get new neurons in the brain.

349
00:23:56.700 --> 00:24:01.700
The general story has always been you're born with 86 billion neurons and those slowly die with time.

350
00:24:01.700 --> 00:24:08.700
But in rats, for example, there is a little trickle of new cells, new brain cells.

351
00:24:08.700 --> 00:24:12.700
And there's been a debate for a long time about whether that little trickle happens in humans or not.

352
00:24:12.700 --> 00:24:13.700
Still unresolved.

353
00:24:13.700 --> 00:24:18.700
But in rats, what you can see is that exercise causes the trickle to increase.

354
00:24:18.700 --> 00:24:24.700
If you stick the rat on the wheel and is doing physical exercise, you get more new brain cells.

355
00:24:24.700 --> 00:24:31.700
Now, we don't know for sure that this happens in humans, but lots of things about physical fitness and exercise matter a lot to the brain.

356
00:24:31.700 --> 00:24:34.700
This is nothing new, exercise, sleep, diet.

357
00:24:34.700 --> 00:24:37.700
These are really important things for keeping the health of this organ.

358
00:24:37.700 --> 00:24:46.700
Is there anything else that's important to know for someone that is trying to change and improve and keep that brain in a healthy state as they age that we haven't touched on?

359
00:24:46.700 --> 00:24:52.700
There is something that all of us are thinking about, which is about social media and the internet in general.

360
00:24:52.700 --> 00:25:01.700
I do think one of the interesting things about the internet and social media is that if we were growing up in a village 500 years ago,

361
00:25:01.700 --> 00:25:04.700
you just know the people in the village and what they can do and so on.

362
00:25:04.700 --> 00:25:08.700
But let's say no one in the village was an entrepreneur or a neuroscientist.

363
00:25:09.700 --> 00:25:14.700
And so we can't even picture that as a thing. We don't know anything about that.

364
00:25:14.700 --> 00:25:21.700
One thing that the internet has done for kids growing up in the digital age is that you get a lot more exposure to things.

365
00:25:21.700 --> 00:25:22.700
You have so much more exposure.

366
00:25:22.700 --> 00:25:29.700
And I actually think this is one of the positive things that I would say about social media is that you not only get exposure.

367
00:25:29.700 --> 00:25:35.700
Wow, that kind of thing is possible and that kind of thing is possible, but you also have people teaching you how to get there.

368
00:25:35.700 --> 00:25:42.700
They say like, hey, I'm a fitness influencer. I'm going to show you exactly how to do the thing or you say, hey, here's exactly how you start a business.

369
00:25:42.700 --> 00:25:47.700
Or I say, hey, here's the route that you go through undergrad and grad school to become a neuroscientist.

370
00:25:47.700 --> 00:25:55.700
And that's great. I mean, there's so much more of a talent window now that everyone gets exposed to. So I think that makes a better brain.

371
00:25:55.700 --> 00:26:01.700
What are we doing to all the children that you think we probably shouldn't be doing as it relates to brain development?

372
00:26:01.700 --> 00:26:06.700
And here's the thing that's really important about this debate is that nobody really knows and I'll tell you why.

373
00:26:06.700 --> 00:26:13.700
It's because to do anything in science when you're saying something about a group, you need to have a control group that you're comparing against.

374
00:26:13.700 --> 00:26:21.700
And when it comes to asking the question of, hey, kids growing up now with social media or the internet, how do they compare to other brains of kids who don't group that?

375
00:26:21.700 --> 00:26:30.700
Well, we don't have a control group unless you look at kids who are incredibly impoverished or let's say quakers who don't believe in technology.

376
00:26:30.700 --> 00:26:37.700
And with both those groups, there's a hundred other important differences. So you can't just say, oh, look, I'm comparing to this kid who grew up without food.

377
00:26:37.700 --> 00:26:41.700
And I'm going to say there's this difference. Who the heck knows why the difference is there?

378
00:26:41.700 --> 00:26:47.700
Even a generation ago, there's so many differences in terms of diet, pollution, and politics, and blah, blah.

379
00:26:47.700 --> 00:26:53.700
But everything that you can't do it. So I only mention this because I think it's very important.

380
00:26:53.700 --> 00:26:59.700
A lot of people pipe off with things about, oh, the younger generation, their brain, this, that, but we don't actually know.

381
00:26:59.700 --> 00:27:07.700
And I will tell you that I happen to be a cyber optimist on this point about what growing up with the internet does for young people.

382
00:27:07.700 --> 00:27:11.700
I think it's going to make them much smarter than the generation that came before.

383
00:27:11.700 --> 00:27:17.700
And here's why it has to do with the size of the intellectual diet that they can bring in.

384
00:27:17.700 --> 00:27:25.700
When I was a kid, I grew up pre-internet. You know, I wanted to know stuff so my mom would drive me to the library, which was 25 minutes away.

385
00:27:25.700 --> 00:27:31.700
And I would pick up the encyclopedia Britannica, and I would flip through it and hope they had an article about the thing that I wanted to know about.

386
00:27:31.700 --> 00:27:35.700
And that's how I was able to get my little straw of knowledge.

387
00:27:35.700 --> 00:27:43.700
But now, kids are growing up with access to anything they're interested in. And this is so good for the brain.

388
00:27:43.700 --> 00:27:51.700
And from a plasticity point of view, the reason this matters is because change happens in the brain when you are curious about something.

389
00:27:51.700 --> 00:28:00.700
So when a kid asks a question to Alexa or Sierra or whatever, and they get the answer, that sticks because they have the right cocktail of chemicals going on in their head.

390
00:28:00.700 --> 00:28:04.700
In contrast, when I grew up, I learned tons of just in case knowledge.

391
00:28:04.700 --> 00:28:09.700
I mean, that's all that the teachers could teach us is just in case you ever need to know this fact. Here it is.

392
00:28:09.700 --> 00:28:22.700
But kids are in a really great situation now. So there are pros and cons to all this stuff, but I think I'm very optimistic about what this means for the warehouse of knowledge that kids can build up now.

393
00:28:22.700 --> 00:28:31.700
And by the way, I saw an interview with Isaac Asimov in 1988. He was the great science fiction writer who wrote foundation in so many other books.

394
00:28:31.700 --> 00:28:44.700
And he was saying on the show in 1988, he said, look, I envision a day when there will be one central supercomputer and every house will have a cable running to that supercomputer.

395
00:28:44.700 --> 00:28:50.700
And you can ask any question you want. And it knows the entirety of humankind's knowledge on that computer.

396
00:28:50.700 --> 00:28:59.700
You know, when he was foreseen here was the internet, he got the details wrong. It doesn't matter. But the idea is he saw how this would be so incredible for education.

397
00:28:59.700 --> 00:29:14.700
Because he pointed out, look, in any classroom, it's going too fast for half the kids to slow for the other half of the kids. And if you could just pursue the sphere of humankind's knowledge, if you could enter in whatever door you wanted to, that's the way to do it.

398
00:29:14.700 --> 00:29:23.700
And because you'll be motivated now, he wasn't talking about brain plasticity or anything, but this is exactly what I'm saying from a brain plasticity point of view really matters.

399
00:29:23.700 --> 00:29:32.700
I'll just mention something, which is a lot of people are concerned that, oh, with with AI, we're going to get lazy. We won't, you know, know how to do anything anymore because we can outsource it.

400
00:29:32.700 --> 00:29:40.700
It just so happens that I love doing home improvement. I'm always fixing my house. I have three acts to myself in the last half year because of AI.

401
00:29:40.700 --> 00:29:48.700
Because I take a picture of something I say, hey, I've never seen this kind of thing before. How does this work? Whatever. And chat GPT says, oh, you do this and you take this out. Here's the bolt and blah, blah.

402
00:29:48.700 --> 00:29:58.700
And it's not me outsourcing it. It's me being curious about something. And so I remember how to do everything now. I know how to do much more than I used to because I like it.

403
00:29:58.700 --> 00:30:10.700
What about the, you know, there's been a couple of studies that have come out that say things like your brain's going to atrophy. If you don't continue to write or if you just defer all of your learning to things like chat GPT or other AI models.

404
00:30:10.700 --> 00:30:22.700
I guess one of the areas that I think in one of the studies, was it a Stanford study that everyone was talking about, where the participants used Google and AI and then they'd learnt something themselves.

405
00:30:22.700 --> 00:30:32.700
But one of the things I've wondered is if I'm going through my business life and I'm encountering hard problems and every time I get encounter a hard problem, I drop it into an AI.

406
00:30:32.700 --> 00:30:38.700
The AI spits out a text based on a copy and paste that and send it as my response.

407
00:30:38.700 --> 00:30:46.700
Presumably there's some kind of important part of the learning cycle of the neurological development that I'm like for going there and missing.

408
00:30:46.700 --> 00:30:54.700
That I probably should, you know, said earlier about doing hard things. What I'm doing there as I'm avoiding that thing, which is like thinking about it and trying to understand it.

409
00:30:54.700 --> 00:31:02.700
Yeah. Here's I think the really important distinction. There's vicious friction in our lives and there's virtuous friction.

410
00:31:02.700 --> 00:31:12.700
So vicious friction is all the stupid stuff that you have to do like, hey, Steven, for your business, I need you to copy this spreadsheet over here and fill in all these cells and do your taxes and whatever.

411
00:31:12.700 --> 00:31:18.700
Okay, that if we can push that off to AI is massively important for improving human lives.

412
00:31:18.700 --> 00:31:28.700
There's really not benefit in vicious friction, but virtuous friction is, hey, Steven, I really want you to think about what is the optimal way to do this business?

413
00:31:28.700 --> 00:31:34.700
What is the best structure for this? How do we actually go D to see? How do we go B to B on this?

414
00:31:34.700 --> 00:31:40.700
What's the what's the approach here that we're going to take that you haven't done before that would be amazing?

415
00:31:40.700 --> 00:31:45.700
That's virtuous friction because you're really using your brain to learn stuff that way.

416
00:31:45.700 --> 00:31:51.700
So that's the first distinction that matters is get rid of all the busy work. There's no honor in that.

417
00:31:51.700 --> 00:31:58.700
I mean, I'll just mention in the 1990s, there was this big debate about whether we should have kids use desk calculators or not.

418
00:31:59.700 --> 00:32:08.700
And thank God that finally guy was talking, we let kids use calculators so that we can learn, you know, we can spend a couple of days learning long division, but you don't have to spend six months on it because who cares?

419
00:32:08.700 --> 00:32:21.700
With the virtuous friction, there's real opportunity to surf the wave of AI so that you are figuring out these tough problems with the aid of somebody who cares about your problem

420
00:32:21.700 --> 00:32:25.700
and is willing to talk with you 24 seven and never gets tired of talking to you about it.

421
00:32:25.700 --> 00:32:38.700
And so you are not just copying and pacing, but you're working with the AI to come up with ideas that were beyond what you would have come up with because I mentioned earlier about internal models, we have pretty narrow fence lines.

422
00:32:38.700 --> 00:32:47.700
And you can think of all these things, but you don't even know what you don't know. So if you can have somebody who's willing to talk with you, an expert in all of humankind's knowledge,

423
00:32:47.700 --> 00:32:58.700
willing to talk to you about it as much as you want, there's a real opportunity there to have a synergy where collectively you both come up with a better idea than either of you could have alone.

424
00:32:58.700 --> 00:33:09.700
But is there a way for that relationship to take place so that I actually benefit because in the example I gave, I take the question I was asked, I put it into an AI, it gives me an answer, I copy and paste it back to the person that asked me the question.

425
00:33:09.700 --> 00:33:15.700
That would happen if you really didn't care about the person asking you the question or the question. I mean, this is what a lot of people are doing.

426
00:33:15.700 --> 00:33:27.700
I get so many, because we interview a lot of candidates who join the business. And so I see tens of thousands of emails. Sometimes a week that I mean, I don't see all of them, but the ones that I see I often know that, you know, because we've sent them five questions or a task.

427
00:33:27.700 --> 00:33:38.700
And I look at it and go, this is, I can almost predict the exact model that sent it to me, because they will have a different personality. So I've got all this one, the person put into Gemini, all this one, the person put it into chat, GPT.

428
00:33:38.700 --> 00:33:50.700
Yeah, exactly. And as full of contrast of construction, like, and it's not this, it's that. Yeah. And then the M dash is exactly, I'm really asking, like, is the person that did that benefiting from it?

429
00:33:50.700 --> 00:34:01.700
No. Well, no, but for a couple of reasons. One is that you know, yeah. And it triggers your red flag. And so that does not do anyone any good. I see so many my colleagues posting on LinkedIn.

430
00:34:01.700 --> 00:34:14.700
He's very obvious AI things. And it irritates me because I feel like I'm not going to spend my time reading that because of, I call this the effort phenomenon, which is in psychology.

431
00:34:14.700 --> 00:34:22.700
We care a lot about things that seem like they took a lot of effort. And there's something about seeing an AI post that's just irritating because it's so obviously AI.

432
00:34:22.700 --> 00:34:35.700
That's a really interesting idea, the effort phenomenon. Yeah. I mean, I've been writing about this for a while because it turns out there are psychology studies where if I offer you two pieces of art and one of them looks like, you know, let's say it's a red dot in the middle of a white canvas.

433
00:34:35.700 --> 00:34:45.700
And the other one is, you know, bottle caps stacked up and glued in this great shape or whatever. You'll pay, you'll pay much more for the thing that looks like it took a lot of effort.

434
00:34:45.700 --> 00:35:00.700
People will pay more for a real diamond than a synthetic lab-grown diamond, which is exactly the same thing. It's just carbon and the matrix. But they feel like, oh, mother nature took hundreds of millions of years of effort on this one, but not over here. It's just a few days in the lab.

435
00:35:00.700 --> 00:35:13.700
So there's a million ways where we care about that a lot. When it comes to this AI thing, yes, anybody who's just popping back some thing to you, it just feels like, all right, they took the path of these resistance. And I'm not so interested.

436
00:35:13.700 --> 00:35:28.700
I want to make from a neuroscience perspective, whether they benefit presumably they don't benefit too much either. I mean, it's hard to know exactly how many times they went back and forth with it. They could have said, hey, Chatchy, thank you for this. But I'm kind of this more of this person. When I really think about it.

437
00:35:28.700 --> 00:35:36.700
This is the thing that inspires me, not what you suggested. So somebody could put effort into it. It's just that we can't know that when we get the AI response.

438
00:35:36.700 --> 00:35:46.700
It seems to be a pretty consistent principle of life, generally, that like when you do something hard or when you put in effort, as you say, you tend to get back like an equal and opposite return relatively.

439
00:35:46.700 --> 00:35:58.700
So I would think that if I fought through, you know, maybe even using AI as a companion, but I fought then to write it out myself instead of just copying and pasting.

440
00:35:58.700 --> 00:36:04.700
One of the things I've learned from doing this podcast and all these episodes is everything is a trade off.

441
00:36:04.700 --> 00:36:18.700
And if you don't know what the trade you're making, then you're often at great risk. And so like some of my friends will say, oh, I take this pill and it's amazing. It does all these things for me. It's the most amazing thing ever. I can just focus for 24 hours a day and I'm so productive now.

442
00:36:18.700 --> 00:36:33.700
And I go, what's the downside? And they go, there's no downside. And I go, hmm. So that's what I mean. It's even worse when you don't, you don't know the trade you're making. And so with AI, I go, okay, if it's making me wildly more efficient or productive, what trade am I making?

443
00:36:33.700 --> 00:36:49.700
I think understanding this, it's probably not two categories, but a spectrum from vicious friction to virtuous friction, but really paying attention to what is virtuous friction, what would make me a better person if I actually put the effort into this, that matters a lot.

444
00:36:49.700 --> 00:37:06.700
And I will say, for us as professors, for you looking for job candidates, we need to change how we're asking the questions. If we just say, hey, right answer these five questions. Of course, everyone's going to use it. For example, in my classes at Stanford, I don't have people turn into final paper anymore.

445
00:37:06.700 --> 00:37:23.700
That was from previous life before AI. Now I have them do projects as their final thing where they're running an experiment on something. And of course they use AI to help them generate some of the issues, but they have to deal with other people and look at the data and figure out what's wrong and that kind of stuff.

446
00:37:23.700 --> 00:37:34.700
I worry that it's getting into the age of the whole calculate thing you said, well, maybe actually it is now you need to assess them on their ability to use the AI not to succeed without it.

447
00:37:34.700 --> 00:37:45.700
Yeah, great. This is the whole game for all of us, I think, is figuring out how to surf this wave of AI where it can make us superhumanic. We can just be better so much better than anything we ever were doing before.

448
00:37:45.700 --> 00:37:53.700
Because we have immediate access to knowledge and facts that either we had forgotten or we never knew existed. And so we should be surfing that wave.

449
00:37:53.700 --> 00:38:05.700
So I totally grew to you on that point. If you can figure out how to change your interview questions so that you're seeing, hey, can this person really get the speed with everything you know about learning and neuroplasticity and expanding one's brain.

450
00:38:05.700 --> 00:38:12.700
Is there anything else you can say to the audience about how they should use AI so that they become a superhuman?

451
00:38:12.700 --> 00:38:21.700
Interesting. You know, look, I've been talking to my friends about this issue a lot lately and I mentioned how I've become so much better at home improvement stuff. I just know so much more.

452
00:38:21.700 --> 00:38:31.700
Each one of my friends has something like that. We're like, hey, you know what? I've actually gotten so much better at this super random thing that I never even thought I, you know, I never thought about it explicitly.

453
00:38:31.700 --> 00:38:40.700
But because I'm always asking questions about that and it's giving me the answers, it's not simply that it gives me the answers and I forget it.

454
00:38:40.700 --> 00:38:50.700
It gives me the answers and I remember it. I'd become better and better because it's like the way that Alexander the Great had Aristotle as his tutor and could ask him anything and learn great stuff from him.

455
00:38:50.700 --> 00:38:58.700
We've all got Aristotle on our pocket now and we can become better at the things that we want to do the things that resonate with us for whatever reason.

456
00:38:58.700 --> 00:39:02.700
If everyone's got Aristotle in that pocket, how does one create an edge?

457
00:39:02.700 --> 00:39:10.700
I think it has to do with we're all just going to be running faster in the same way that when Steve Jobs introduced Apple computers, he said this is like a bicycle for the mind.

458
00:39:10.700 --> 00:39:22.700
What he meant by that was that for millions of years, we've been walking bipedally and then just in the last nanosecond of evolution, we invented the bicycle and suddenly humans can move faster because of the bicycle.

459
00:39:23.700 --> 00:39:31.700
And he said having a personal computer is like a bicycle for the mind and I think of AI now is like a motorcycle for the mind.

460
00:39:31.700 --> 00:39:42.700
It allows us to move so much faster. So now it's a motorcycle race and there will be people who are much faster than other people because they're really using that optimally.

461
00:39:42.700 --> 00:39:47.700
And that's what I mean is like, how do I create an edge of us as my, whoever I'm competing with and whatever industry I'm in?

462
00:39:47.700 --> 00:39:53.700
Well, for sure the people who are just copying and pasting the AI slot, that'll be easy to beat that crowd.

463
00:39:53.700 --> 00:39:57.700
But otherwise, I think it's just a matter of, hey, these are the newest things.

464
00:39:57.700 --> 00:40:07.700
It's like in history when the new sword gets invented or the new gunner, the new cannon, you know, you have to keep improving and using that and that's what's going on now with AI.

465
00:40:08.700 --> 00:40:20.700
And from a neuroscience perspective, if I wanted to use AI to based on all these things you've told me about novelty and all these other points that expand the connections across my brain and give me a big cognitive reserve.

466
00:40:20.700 --> 00:40:25.700
What might I, I install as a practice every week when I'm speaking to my AI?

467
00:40:25.700 --> 00:40:30.700
Oh, ask it questions that you're curious about about anything, just asking questions.

468
00:40:30.700 --> 00:40:32.700
Here's one thing I do all the time.

469
00:40:32.700 --> 00:40:37.700
I'll say, hey, I've been thinking about this, you know, on my podcast, I do a lot of monologues.

470
00:40:37.700 --> 00:40:41.700
And so I'll start talking to it and I'll say, hey, I've got this idea that I'm thinking about.

471
00:40:41.700 --> 00:40:44.700
What if, and then I'll say, here's my idea.

472
00:40:44.700 --> 00:40:46.700
Give me pros and cons.

473
00:40:46.700 --> 00:40:48.700
You know, tell me why this is wrong.

474
00:40:48.700 --> 00:40:55.700
And I do that pretty much with everything that I ask it if I'm proposing some, you know, stupid seed of an idea.

475
00:40:55.700 --> 00:40:59.700
And it really gives me the care arguments and I really engage with it.

476
00:40:59.700 --> 00:41:01.700
That is the important part, I think.

477
00:41:01.700 --> 00:41:06.700
And by the way, I just want to say, I think for the next generation that we're teaching, this,

478
00:41:06.700 --> 00:41:12.700
there are really only two things we can teach because all the details of, you know, hey, let's teach computer programming or something.

479
00:41:12.700 --> 00:41:14.700
It's probably already gone as a useful thing.

480
00:41:14.700 --> 00:41:19.700
So what we can teach is critical thinking and creativity.

481
00:41:19.700 --> 00:41:20.700
That's it.

482
00:41:20.700 --> 00:41:25.700
I think that's such an important point at this point about asking your AI why you might be wrong.

483
00:41:25.700 --> 00:41:26.700
Yeah.

484
00:41:26.700 --> 00:41:33.700
I think I've had most of my paradigm shifting moments when I've come to an AI model that I was using with a very, with very high conviction.

485
00:41:33.700 --> 00:41:40.700
And what the prompt that always, I think, is most sort of expansive in terms of my intellectual knowledge is when I say to it,

486
00:41:40.700 --> 00:41:43.700
be brutally honest about your opinion.

487
00:41:43.700 --> 00:41:48.700
Think for yourself and be objective and tell me where my blind spots are.

488
00:41:48.700 --> 00:41:53.700
There's something you need within us all where we don't actually want to be wrong.

489
00:41:53.700 --> 00:41:55.700
We often, I think, as an actual reflex.

490
00:41:55.700 --> 00:41:58.700
And this is why people get really sort of trapped in echo chambers of political opinion.

491
00:41:58.700 --> 00:42:05.700
And, you know, Leon Fessinger talked about the side of cognitive dissonance when something you believe contrasts with new information

492
00:42:05.700 --> 00:42:07.700
and how it makes you feel uncomfortable.

493
00:42:07.700 --> 00:42:12.700
There's something when I type that out, when I love the idea or the thing I've written or the memo I've written,

494
00:42:12.700 --> 00:42:15.700
this new idea, and I go, go on, tell me why I'm completely, completely wrong.

495
00:42:15.700 --> 00:42:17.700
And it evicerates me.

496
00:42:17.700 --> 00:42:21.700
It is both uncomfortable, but feels incredibly important.

497
00:42:21.700 --> 00:42:24.700
Because then it's like I've grown.

498
00:42:24.700 --> 00:42:28.700
But these AI is that that programmed almost to like kiss my ass.

499
00:42:28.700 --> 00:42:34.700
Yes, although, you know, Chachupati released a very psychophantic version, I don't know, maybe a year ago.

500
00:42:35.700 --> 00:42:38.700
Meaning it compliments you, you give some idea and it says,

501
00:42:38.700 --> 00:42:40.700
Oh, Steven, that's the best idea I've ever heard.

502
00:42:40.700 --> 00:42:41.700
You're ingenious and blah, blah.

503
00:42:41.700 --> 00:42:43.700
And that didn't last very long.

504
00:42:43.700 --> 00:42:45.700
That model because nobody actually liked it.

505
00:42:45.700 --> 00:42:47.700
So you're exactly right.

506
00:42:47.700 --> 00:42:53.700
And I'm sure most listeners know this, but you can tell your AI to be brutally honest with you all the time.

507
00:42:53.700 --> 00:42:55.700
You can tell them to do that all the time.

508
00:42:55.700 --> 00:42:56.700
And it'll do that.

509
00:42:56.700 --> 00:42:59.700
So you can, you can establish the kind of person that you're talking to.

510
00:42:59.700 --> 00:43:00.700
Here's the thing.

511
00:43:00.700 --> 00:43:01.700
You're right, of course.

512
00:43:01.700 --> 00:43:03.700
People don't like to be wrong.

513
00:43:03.700 --> 00:43:04.700
It can be socially embarrassing.

514
00:43:04.700 --> 00:43:05.700
It can be uncomfortable.

515
00:43:05.700 --> 00:43:08.700
And yet there's something very different when you're talking to your AI.

516
00:43:08.700 --> 00:43:09.700
It's a very private thing.

517
00:43:09.700 --> 00:43:11.700
And you say, hey, tell me why I'm brutally wrong.

518
00:43:11.700 --> 00:43:16.700
And when it tells you you think, oh, thank God, it's telling me that instead of like a real human.

519
00:43:16.700 --> 00:43:20.700
So I think a lot of that is alleviated with AI.

520
00:43:20.700 --> 00:43:23.700
We don't feel as bad about being wrong there.

521
00:43:23.700 --> 00:43:25.700
As you were saying, I just want to enchant Chachupati.

522
00:43:25.700 --> 00:43:26.700
And I type this in.

523
00:43:26.700 --> 00:43:28.700
It's my joke funny.

524
00:43:28.700 --> 00:43:30.700
And the joke I typed in is knock knock.

525
00:43:30.700 --> 00:43:31.700
Who's there?

526
00:43:31.700 --> 00:43:32.700
A lettuce.

527
00:43:32.700 --> 00:43:33.700
Letters who?

528
00:43:33.700 --> 00:43:35.700
Letters in and I'll tell you.

529
00:43:35.700 --> 00:43:36.700
Okay.

530
00:43:36.700 --> 00:43:37.700
You didn't laugh.

531
00:43:37.700 --> 00:43:38.700
I didn't laugh.

532
00:43:38.700 --> 00:43:39.700
Okay.

533
00:43:39.700 --> 00:43:40.700
Chachupati said, yes, it works as a joke.

534
00:43:40.700 --> 00:43:45.700
A solid structure uses the classic pun payoff, which is exactly how most knock knock jokes land.

535
00:43:45.700 --> 00:43:47.700
And then it's done a laughing emoji.

536
00:43:47.700 --> 00:43:49.700
I then said, be brutally honest and completely objective.

537
00:43:49.700 --> 00:43:50.700
Was that funny?

538
00:43:50.700 --> 00:43:54.700
It said, it's not very funny.

539
00:43:54.700 --> 00:43:55.700
Interesting.

540
00:43:55.700 --> 00:43:59.700
I mean, that's interesting because it depends, right?

541
00:43:59.700 --> 00:44:01.700
A little child actually finds that joke funny.

542
00:44:01.700 --> 00:44:08.700
And for a little child, they then get to repeat that to their classmate and they're learning how to do a joke and so on.

543
00:44:08.700 --> 00:44:13.700
So I'm not, I'm not sure I think there's a single answer to whether that can be funny or not.

544
00:44:13.700 --> 00:44:17.700
But the interesting thing is it just reinforcing what I already believed.

545
00:44:17.700 --> 00:44:24.700
And therefore, when we think about growth or having a growth mindset, if someone's just always reinforcing what you already believe in, no.

546
00:44:24.700 --> 00:44:27.700
I don't know if it's ever going to be a growth mindset.

547
00:44:27.700 --> 00:44:28.700
I mean, I just asked it again.

548
00:44:28.700 --> 00:44:29.700
I said, be really honest.

549
00:44:29.700 --> 00:44:31.700
And it said, it's absolutely not funny.

550
00:44:31.700 --> 00:44:32.700
Yeah.

551
00:44:32.700 --> 00:44:36.700
But remember, all it's doing is it's just it's a statistical parrot.

552
00:44:36.700 --> 00:44:40.700
And so when you say be really honest, it thinks that's what it should answer.

553
00:44:40.700 --> 00:44:42.700
I said, also be even more honest.

554
00:44:42.700 --> 00:44:44.700
It says it's basically not funny at all.

555
00:44:44.700 --> 00:44:45.700
And you shouldn't say that to people.

556
00:44:45.700 --> 00:44:46.700
Okay.

557
00:44:46.700 --> 00:44:50.700
And it says, comedic originality, one out of ten, likely hit of real laughter, one out of ten.

558
00:44:50.700 --> 00:44:51.700
Well, that's quite good.

559
00:44:51.700 --> 00:44:52.700
That's quite accurate.

560
00:44:53.700 --> 00:44:54.700
Here's the thing.

561
00:44:54.700 --> 00:44:58.700
I've been thinking about this issue a lot about whether AI can be funny.

562
00:44:58.700 --> 00:45:00.700
And at the moment, it can't be.

563
00:45:00.700 --> 00:45:02.700
It's created repeating jokes.

564
00:45:02.700 --> 00:45:05.700
But it doesn't understand humor on its own.

565
00:45:05.700 --> 00:45:06.700
What it knows.

566
00:45:06.700 --> 00:45:11.700
If you ask it to make up a new joke, what it'll do is it'll have the first guy walks in the bar.

567
00:45:11.700 --> 00:45:15.700
Then the second guy walks in the bar and does X and that establishes the pattern.

568
00:45:15.700 --> 00:45:19.700
But then the third guy, it'll have break that pattern, which is the structure of a joke.

569
00:45:20.700 --> 00:45:23.700
But it doesn't know how to break the pattern in a way that's funny.

570
00:45:23.700 --> 00:45:25.700
It's just the third guy does some random thing.

571
00:45:25.700 --> 00:45:30.700
So AI, as it stands now, the way it's structured with what's called a transformer model,

572
00:45:30.700 --> 00:45:36.700
doesn't know how to think of the punchline and then go back and make the joke lead to that punchline.

573
00:45:36.700 --> 00:45:38.700
A lot of people don't either.

574
00:45:38.700 --> 00:45:39.700
Do you know what I mean?

575
00:45:39.700 --> 00:45:42.700
I say that in a fence wave, just to say that.

576
00:45:42.700 --> 00:45:43.700
I don't know.

577
00:45:43.700 --> 00:45:45.700
I often hear the claim that AI could never be creative.

578
00:45:45.700 --> 00:45:47.700
What's massively creative?

579
00:45:47.700 --> 00:45:54.700
Here's why creativity in the brain, all creativity is, is you absorb your world, the whole world around you,

580
00:45:54.700 --> 00:45:55.700
every experience you've ever had.

581
00:45:55.700 --> 00:46:01.700
And then you're bending and breaking and blending those cognitive concepts into new remixes.

582
00:46:01.700 --> 00:46:02.700
That's all creativity.

583
00:46:02.700 --> 00:46:06.700
And you're doing that all the time, whether you're just trying to think of what to say next,

584
00:46:06.700 --> 00:46:10.700
or what recipe to make next, or what patent to do, or what company to start,

585
00:46:10.700 --> 00:46:13.700
you're just remixing the stuff that you already know.

586
00:46:13.700 --> 00:46:16.700
And that's why, you know, I don't know, take Beethoven.

587
00:46:16.700 --> 00:46:21.700
He could have written any kind of music that was being done anywhere in the world.

588
00:46:21.700 --> 00:46:23.700
But of course, he didn't like that's what he grew up with.

589
00:46:23.700 --> 00:46:25.700
There's the music in his local culture and so on.

590
00:46:25.700 --> 00:46:28.700
What we have now is a much broader diet.

591
00:46:28.700 --> 00:46:31.700
As I mentioned before, where we can get everything going in.

592
00:46:31.700 --> 00:46:35.700
But the point I want to make here is that AI, that's what it does.

593
00:46:35.700 --> 00:46:37.700
It remixes stuff that's come in.

594
00:46:37.700 --> 00:46:39.700
So AI is massively creative.

595
00:46:39.700 --> 00:46:46.700
The part of creativity that AI can't do right now is selection, meaning it can generate 100 pictures.

596
00:46:46.700 --> 00:46:47.700
But it doesn't know which one to pick.

597
00:46:47.700 --> 00:46:50.700
It doesn't know which one is going to be the most appealing to you.

598
00:46:50.700 --> 00:46:52.700
But it can remix beautifully.

599
00:46:52.700 --> 00:46:53.700
But neither do humans, right?

600
00:46:53.700 --> 00:46:58.700
So if I asked an intern to make me 100 pictures, I mean, I could get my hair to pick one.

601
00:46:58.700 --> 00:47:02.700
But I wouldn't know what the intern or the AI wouldn't know which one I loved.

602
00:47:02.700 --> 00:47:04.700
The intern would have a much better shot at it.

603
00:47:05.700 --> 00:47:09.700
And as the intern is there for a while, here she becomes quite good at getting,

604
00:47:09.700 --> 00:47:10.700
oh, okay, I get Steven's taste.

605
00:47:10.700 --> 00:47:11.700
It would be this one.

606
00:47:11.700 --> 00:47:12.700
An AI, can't learn that.

607
00:47:12.700 --> 00:47:13.700
Well, my taste is.

608
00:47:13.700 --> 00:47:15.700
I don't think the AI could learn that about visual images.

609
00:47:15.700 --> 00:47:18.700
Because when it generates the pixels that's doing this, you know,

610
00:47:18.700 --> 00:47:22.700
this magical stuff under the hood where it's deciding which pixels and how they diffuse together

611
00:47:22.700 --> 00:47:24.700
and, you know, makes the image.

612
00:47:24.700 --> 00:47:29.700
But it doesn't know how to read that image like, oh, yeah, the way this is.

613
00:47:29.700 --> 00:47:31.700
And that'll really appeal to Steven.

614
00:47:31.700 --> 00:47:34.700
It does, it does, it's that seeing the image, except as a bunch of pixels.

615
00:47:34.700 --> 00:47:35.700
Hmm.

616
00:47:35.700 --> 00:47:36.700
Hmm.

617
00:47:36.700 --> 00:47:37.700
Can you be human for that?

618
00:47:37.700 --> 00:47:42.700
Because I feed, I was doing an experiment recently where I took my behind the scenes channel,

619
00:47:42.700 --> 00:47:44.700
which is a 30 minute long video.

620
00:47:44.700 --> 00:47:45.700
I dropped it into Gemini.

621
00:47:45.700 --> 00:47:49.700
And I'd say things to it like predict where people would drop off on the video.

622
00:47:49.700 --> 00:47:51.700
And then we upload the video to YouTube.

623
00:47:51.700 --> 00:47:52.700
We get the retention data back.

624
00:47:52.700 --> 00:47:57.700
And Gemini, in the last two times that I've done it, has a hundred percent record

625
00:47:57.700 --> 00:48:02.700
of noting that a minute seven where insert person talked for too long

626
00:48:02.700 --> 00:48:07.700
and might have been a bit more selly, might have tried to sell a hoodie, for example,

627
00:48:07.700 --> 00:48:10.700
in that part, it would say, you're going to lose people here.

628
00:48:10.700 --> 00:48:12.700
And it would very accurately say why?

629
00:48:12.700 --> 00:48:17.700
It would say because you talked for 74 seconds and it was jarring

630
00:48:17.700 --> 00:48:19.700
versus the moment that came before it.

631
00:48:19.700 --> 00:48:22.700
And when I feed the AI, I don't know, let's say thumbnails and say,

632
00:48:22.700 --> 00:48:24.700
which thumbnails are going to perform the best?

633
00:48:24.700 --> 00:48:29.700
Recently where we put four thumbnail test results that we knew the answer to into Gemini

634
00:48:29.700 --> 00:48:32.700
and said which one's going to win on YouTube baby testing?

635
00:48:32.700 --> 00:48:38.700
And it got a hundred percent accuracy of predicting on data we already had which one would win.

636
00:48:38.700 --> 00:48:40.700
And so now I don't know.

637
00:48:40.700 --> 00:48:44.700
I keep having these paradigm shifting moments where I'm going to only humans could do that.

638
00:48:44.700 --> 00:48:50.700
But increasingly, the AI's that we're experimenting with are making better creative decisions

639
00:48:50.700 --> 00:48:55.700
than now I can make myself as if the outcome of that creative decision is which one is people going to prefer.

640
00:48:55.700 --> 00:48:57.700
I'd say a year ago that wasn't the case.

641
00:48:57.700 --> 00:49:01.700
Okay, so I totally agree with you, but let me just mention one thing which is fascinating,

642
00:49:01.700 --> 00:49:05.700
which is that often the way it's doing it is not at all the way that a human would do,

643
00:49:05.700 --> 00:49:07.700
which might be fine for our purposes.

644
00:49:07.700 --> 00:49:14.700
But the data and the way that it's picking up on it, it might be something about how much I'm making this up.

645
00:49:14.700 --> 00:49:19.700
How much green was in the YouTube thumbnail image or how much red or whatever,

646
00:49:19.700 --> 00:49:24.700
whatever the thing is or just noticing that there's big font versus smaller font or whatever.

647
00:49:24.700 --> 00:49:28.700
The next time you try it, it says, oh yeah, this thumbnail is going to be great.

648
00:49:28.700 --> 00:49:31.700
And it's some ridiculous thumbnail that doesn't make any sense to you as a human,

649
00:49:31.700 --> 00:49:33.700
nor do your fellow humans.

650
00:49:33.700 --> 00:49:39.700
But it might say, oh yeah, this would be great because it's judging things on very weird dimensions that we can't always see.

651
00:49:39.700 --> 00:49:43.700
You know, in the example you gave about maybe it's because the text is bigger or the color red.

652
00:49:43.700 --> 00:49:46.700
But those are the same factors we think about as a human we think.

653
00:49:46.700 --> 00:49:49.700
If we know that if the font is bigger, they're performance better.

654
00:49:49.700 --> 00:49:51.700
We know that red performance better than green.

655
00:49:51.700 --> 00:49:53.700
Quite possibly, but here's the interesting thing.

656
00:49:53.700 --> 00:49:59.700
Human art constantly evolves and all AI is trained on is what has been done before and what has worked.

657
00:49:59.700 --> 00:50:06.700
And so if I asked it, let's say we composed five different songs and said, hey, which song is going to be better?

658
00:50:06.700 --> 00:50:10.700
It's going to say something that's right in the middle of the distribution of popular songs.

659
00:50:10.700 --> 00:50:13.700
But that's not what actually makes it next year and the year after.

660
00:50:13.700 --> 00:50:17.700
It's new things, it's new twists that nobody has seen before.

661
00:50:17.700 --> 00:50:18.700
That's what we love.

662
00:50:18.700 --> 00:50:20.700
That's what we seek as consumers.

663
00:50:20.700 --> 00:50:24.700
And so because AI can only be trained up on what already exists,

664
00:50:24.700 --> 00:50:27.700
it's never going to get the new thing at the edge.

665
00:50:27.700 --> 00:50:32.700
But if the AI was asked to, because I think the reason why a new song would break out,

666
00:50:32.700 --> 00:50:36.700
let's say, you know, a new Drake song comes out and it's a smash hit.

667
00:50:36.700 --> 00:50:39.700
If we think about that distribution curve, so like if I do it on the ground,

668
00:50:39.700 --> 00:50:44.700
you're saying that this middle section here is what sort of AI will aim at

669
00:50:44.700 --> 00:50:46.700
because it's the popular and the known.

670
00:50:46.700 --> 00:50:53.700
Well, if I tell AI to make a million songs, which is kind of what I guess is what's going on every day around the world,

671
00:50:53.700 --> 00:50:58.700
if you scattered them on this graph at like, you know, absolutely.

672
00:50:58.700 --> 00:51:02.700
And then the AI's most unusual song ends up taking off.

673
00:51:02.700 --> 00:51:04.700
But it's just because there's so many of them.

674
00:51:04.700 --> 00:51:08.700
Quite right. But that's the human selection part that we're seeing over there.

675
00:51:08.700 --> 00:51:12.700
If you asked, okay, out of all these dots, which do you think AI is going to be best?

676
00:51:12.700 --> 00:51:14.700
It's going to have to tell you the middle of the curve.

677
00:51:14.700 --> 00:51:17.700
But the surprising part is the part that you circled there,

678
00:51:17.700 --> 00:51:20.700
which is the one on the edge is the one that humans like, why?

679
00:51:20.700 --> 00:51:22.700
Because we're constant novelty seekers.

680
00:51:22.700 --> 00:51:24.700
We care about the things that are new.

681
00:51:24.700 --> 00:51:29.700
I think the point I'm getting at is that the creation of it,

682
00:51:29.700 --> 00:51:33.700
the creative process is still the same, which is like totally AI or human.

683
00:51:33.700 --> 00:51:35.700
Just trying a bunch of shit.

684
00:51:35.700 --> 00:51:37.700
And then the world going, ooh, that one.

685
00:51:37.700 --> 00:51:39.700
Oh, oh, I totally agree.

686
00:51:39.700 --> 00:51:42.700
This is consistent with what I was saying, which is that AI can be massively creative

687
00:51:42.700 --> 00:51:44.700
in terms of the generation of something.

688
00:51:44.700 --> 00:51:46.700
But you need humans to do the selection.

689
00:51:46.700 --> 00:51:51.700
I'm only arguing the point that AI is not good at saying, okay, I've generated a hundred songs.

690
00:51:51.700 --> 00:51:53.700
This is the one humans will choose.

691
00:51:53.700 --> 00:51:58.700
We end up saying, hey, wait, this one is just weird and unique enough that I really like that.

692
00:51:58.700 --> 00:52:03.700
It's interesting, because when you speak to record labels about music,

693
00:52:03.700 --> 00:52:10.700
what they're often doing is getting a format of a song that they know will work.

694
00:52:10.700 --> 00:52:12.700
So they're like, right, so it's going to be eight bars here.

695
00:52:12.700 --> 00:52:13.700
It's going to be this year.

696
00:52:13.700 --> 00:52:14.700
You're going to have a chorus.

697
00:52:14.700 --> 00:52:15.700
That's like, hooky.

698
00:52:15.700 --> 00:52:16.700
It's going to come background.

699
00:52:16.700 --> 00:52:17.700
It's going to build up pace.

700
00:52:17.700 --> 00:52:19.700
And there's like a rough format to it.

701
00:52:19.700 --> 00:52:24.700
And it's no surprise that Ed Sheeran has written so many songs for so many fucking people.

702
00:52:25.700 --> 00:52:27.700
When I spent some time working with Sony,

703
00:52:27.700 --> 00:52:29.700
they had a brand new boy band in the wake of one direction.

704
00:52:29.700 --> 00:52:33.700
And when I sat with the boy band and was introducing myself, they said to me,

705
00:52:33.700 --> 00:52:36.700
oh yes, so here are the boy bands first three songs.

706
00:52:36.700 --> 00:52:39.700
And Ed Sheeran has written all of them.

707
00:52:39.700 --> 00:52:40.700
And I was like, what?

708
00:52:40.700 --> 00:52:43.700
I thought, I thought like, no, Ed Sheeran has written all of them.

709
00:52:43.700 --> 00:52:45.700
And then what we do is we give them to the boy band.

710
00:52:45.700 --> 00:52:47.700
And then the boy band sing them.

711
00:52:47.700 --> 00:52:51.700
And they're pretty much guaranteed to be hits because Ed Sheeran has like a formula.

712
00:52:51.700 --> 00:52:56.700
The way he writes is really in like Vogue right now.

713
00:52:56.700 --> 00:53:00.700
And people tend to think a lot that the songs that are number one in the charts are there

714
00:53:00.700 --> 00:53:02.700
because just because someone had creative genius.

715
00:53:02.700 --> 00:53:04.700
And of course that is the case sometimes.

716
00:53:04.700 --> 00:53:08.700
But there is a lot of this writing going on and then handing the formula over

717
00:53:08.700 --> 00:53:10.700
because someone has cracked the code of a hit.

718
00:53:10.700 --> 00:53:11.700
Right.

719
00:53:11.700 --> 00:53:12.700
But here's the thing.

720
00:53:12.700 --> 00:53:15.700
And you know that we all know this, which is that the code never lasts.

721
00:53:15.700 --> 00:53:20.700
So humans have this pole where they're always seeking things between novels

722
00:53:21.700 --> 00:53:22.700
and familiarity.

723
00:53:22.700 --> 00:53:28.700
So we like things where we recognize the brand and we recognize what the singer has done before.

724
00:53:28.700 --> 00:53:31.700
But there has to be novelty or else we're not going to go for it.

725
00:53:31.700 --> 00:53:35.700
We're not going to listen to that boy band for the next 10 years doing the same song over and over.

726
00:53:35.700 --> 00:53:39.700
So you're of course right that we, you know, we want a bit of familiarity.

727
00:53:39.700 --> 00:53:40.700
We want to be anchored.

728
00:53:40.700 --> 00:53:43.700
But we definitely seek the new.

729
00:53:43.700 --> 00:53:44.700
This is what humans always do.

730
00:53:44.700 --> 00:53:49.700
This is why car companies always release the next model even though the current model is perfectly fine.

731
00:53:49.700 --> 00:53:50.700
This is why haircuts evolve.

732
00:53:50.700 --> 00:53:55.700
This is why fashion evolves through the years because we always care about novelty.

733
00:53:55.700 --> 00:54:00.700
And the other thing in the music industry that I think is also creating a hit

734
00:54:00.700 --> 00:54:03.700
is I was reading many years ago about some psychology which you'll probably know much more about.

735
00:54:03.700 --> 00:54:10.700
That says exactly what you just said, which is we love something when it is familiar but new.

736
00:54:10.700 --> 00:54:11.700
Exactly.

737
00:54:11.700 --> 00:54:15.700
So the way that the record industry and the radio industry makes something familiar

738
00:54:15.700 --> 00:54:22.700
is they blast the same song at you on every radio station for a long period of time until it breaks past

739
00:54:22.700 --> 00:54:26.700
being just novel, just new and it becomes familiar.

740
00:54:26.700 --> 00:54:33.700
And like I saw this graph which shows that the song that you'll love is right there in the middle of like it's new enough

741
00:54:33.700 --> 00:54:40.700
that you're still into it but it's familiar now because you've heard it so many times that you love it.

742
00:54:40.700 --> 00:54:46.700
And if anyone listening, the first time you hear a song you might not love it as much as once you've heard it like 20 times.

743
00:54:46.700 --> 00:54:48.700
And then at some point you've heard it too much.

744
00:54:48.700 --> 00:54:49.700
Yeah.

745
00:54:49.700 --> 00:54:51.700
And it comes back down the other side of the cover.

746
00:54:51.700 --> 00:54:52.700
It's now too familiar.

747
00:54:52.700 --> 00:54:53.700
Yeah, that's exactly right.

748
00:54:53.700 --> 00:54:56.700
And so we're always seeking that tension in the middle.

749
00:54:56.700 --> 00:54:59.700
And companies run into this all the time.

750
00:54:59.700 --> 00:55:04.700
Like sometimes they try things that are too novel that just completely fail.

751
00:55:04.700 --> 00:55:07.700
You know Coca-Cola tried this long time but was introducing new Coke and Owen liked it or whatever.

752
00:55:08.700 --> 00:55:14.700
And other companies like was that company Blackberry with the little thumb things that you can press the physical keyboard on the phone.

753
00:55:14.700 --> 00:55:17.700
They failed because they wouldn't change fast enough.

754
00:55:17.700 --> 00:55:21.700
But companies that make it are always staying in that sweet spot.

755
00:55:26.700 --> 00:55:28.700
Whoa, what's on your face?

756
00:55:28.700 --> 00:55:30.700
This is my bun charge face mask.

757
00:55:30.700 --> 00:55:31.700
I've been wearing this for some time now.

758
00:55:31.700 --> 00:55:32.700
They're a sponsor of the podcast.

759
00:55:32.700 --> 00:55:34.700
I put this on for 15, 20 minutes a day.

760
00:55:34.700 --> 00:55:36.700
I can sit here in the chair and wear it.

761
00:55:36.700 --> 00:55:39.700
Boost my college and production helps with fine line blemishes.

762
00:55:39.700 --> 00:55:40.700
My complexion gets better.

763
00:55:40.700 --> 00:55:43.700
And then people know people listen to podcasts because I look better.

764
00:55:43.700 --> 00:55:46.700
Professional grade equipment in such a small box.

765
00:55:46.700 --> 00:55:47.700
It's non-invasive.

766
00:55:47.700 --> 00:55:50.700
And having sat here with so many of the world's leading health professionals.

767
00:55:50.700 --> 00:55:54.700
There's various things that I repeatedly hear work and some things I'm a bit skeptical about.

768
00:55:54.700 --> 00:55:58.700
This is one of the things that almost all of my guests on this show have confirmed works.

769
00:55:58.700 --> 00:56:01.700
It is really, really, really effective.

770
00:56:01.700 --> 00:56:05.700
And they offer fast, free shipping worldwide with easy returns and exchanges.

771
00:56:05.700 --> 00:56:08.700
And you'll also get a one year warranty on all of their products.

772
00:56:08.700 --> 00:56:10.700
And their HSA and FSA eligible.

773
00:56:10.700 --> 00:56:13.700
Giving you tax resavings up to 40%.

774
00:56:13.700 --> 00:56:19.700
And you can get 20% off when you order through my link at bondcharge.com slash D O A C.

775
00:56:19.700 --> 00:56:23.700
That's bondcharge.com slash D O A C.

776
00:56:23.700 --> 00:56:24.700
The deal applies sight wide.

777
00:56:24.700 --> 00:56:27.700
I'm 100% more productive using this app.

778
00:56:27.700 --> 00:56:30.700
Despite spending 50% less time typing.

779
00:56:30.700 --> 00:56:32.700
And that might confuse you but let me explain.

780
00:56:32.700 --> 00:56:35.700
Which is exactly why I invested in Whisper Flow.

781
00:56:35.700 --> 00:56:37.700
There are also one of our sponsors on this podcast.

782
00:56:37.700 --> 00:56:40.700
Whisper Flow turns your speech into text.

783
00:56:40.700 --> 00:56:43.700
So you can send it in any app or device at any time.

784
00:56:43.700 --> 00:56:46.700
And I promise you, it doesn't seem to ever make mistakes.

785
00:56:46.700 --> 00:56:50.700
This is the most accurate voice dictation I have ever used.

786
00:56:50.700 --> 00:56:52.700
After a decade of trying to get one to work.

787
00:56:52.700 --> 00:56:54.700
That only doesn't save me a ton of time.

788
00:56:54.700 --> 00:56:59.700
It also corrects your speech if you change your mind mid-sentence before turning it into text on the device.

789
00:56:59.700 --> 00:57:00.700
I love it.

790
00:57:00.700 --> 00:57:02.700
And I know my team loves it too because when I posted it in our Slack channel.

791
00:57:02.700 --> 00:57:04.700
Asking if anybody wanted a pro version.

792
00:57:04.700 --> 00:57:07.700
Half the office said yes and they had it within an hour.

793
00:57:07.700 --> 00:57:08.700
Which tells me everything.

794
00:57:08.700 --> 00:57:13.700
This is the tool you and your team need to speed yourselves up and to capture those important ideas.

795
00:57:13.700 --> 00:57:15.700
So that they don't disappear.

796
00:57:15.700 --> 00:57:20.700
Head over to Whisper Flow.ai slash Stephen to download it now.

797
00:57:20.700 --> 00:57:25.700
That's W-I-S-P-R-F-L-O-W dot ai slash Stephen.

798
00:57:26.700 --> 00:57:28.700
When you think about the brain and how it's built.

799
00:57:28.700 --> 00:57:33.700
And then you think about the exact technology that they've used to create AI.

800
00:57:33.700 --> 00:57:34.700
Isn't it very, very similar?

801
00:57:34.700 --> 00:57:40.700
And if so, if it is similar, what does that say about human's role in the future?

802
00:57:40.700 --> 00:57:42.700
It's similar, but it's not the same.

803
00:57:42.700 --> 00:57:45.700
Which is why with AI you get what we call jagged intelligence.

804
00:57:45.700 --> 00:57:49.700
Meaning that it can do something so extraordinarily smart.

805
00:57:49.700 --> 00:57:52.700
And then in the next moment, given the answer that's weird and doesn't make any sense.

806
00:57:53.700 --> 00:57:55.700
AI still is doing this.

807
00:57:55.700 --> 00:57:57.700
It's not thinking like we think.

808
00:57:57.700 --> 00:57:58.700
Okay, why?

809
00:57:58.700 --> 00:58:05.700
It's because AI, as we think about it now, really started, of course, decades and decades ago, where people said look.

810
00:58:05.700 --> 00:58:10.700
You've got all these billions of cells, neurons in the brain that are connected to each other.

811
00:58:10.700 --> 00:58:12.700
What if we ignore all that complexity?

812
00:58:12.700 --> 00:58:15.700
We just say look, imagine that you have units that are connected to each other.

813
00:58:15.700 --> 00:58:19.700
We're going to forget about, you know, a single cell in the brain is as complicated as a city.

814
00:58:19.700 --> 00:58:22.700
It's got the entire human genome, it's trafficking millions of proteins.

815
00:58:22.700 --> 00:58:23.700
Let's put all that aside.

816
00:58:23.700 --> 00:58:28.700
Just imagine it's a circle and it's connected to other cells and each connection has a certain strength.

817
00:58:28.700 --> 00:58:31.700
And that's what we call an artificial neural network.

818
00:58:31.700 --> 00:58:33.700
Now that went off in its own direction.

819
00:58:33.700 --> 00:58:39.700
And the kind of amazing, surprising part is how successful it's been to just get rid of all the detail.

820
00:58:39.700 --> 00:58:43.700
But it's still super different than what human brains are like.

821
00:58:43.700 --> 00:58:49.700
So just an example, this thing I mentioned at the very beginning about how we're a team of rivals under the hood.

822
00:58:49.700 --> 00:58:54.700
You've got all these different competing neural networks that are trying to drive your behavior and so on.

823
00:58:54.700 --> 00:59:02.700
The fact that we're emotional, the fact that we are driven by different appetites, whether food or sexuality or whatever it is.

824
00:59:02.700 --> 00:59:05.700
But, you know, you're a chat GPT, you don't want that in the chat GPT.

825
00:59:05.700 --> 00:59:08.700
So it's just an artificial neural network, many layers deep.

826
00:59:08.700 --> 00:59:10.700
And it's extraordinary at what it does.

827
00:59:10.700 --> 00:59:12.700
But it's so different than human.

828
00:59:12.700 --> 00:59:16.700
For example, the fact that it's read everything on the planet and remembers it.

829
00:59:16.700 --> 00:59:20.700
And you have it, you would need to lead a thousand lifetimes to read that much.

830
00:59:20.700 --> 00:59:22.700
And of course you wouldn't remember much of it.

831
00:59:22.700 --> 00:59:24.700
It's very different is the point I'm making.

832
00:59:24.700 --> 00:59:28.700
They both have converged on something that we would call intelligence.

833
00:59:28.700 --> 00:59:30.700
But it's a pretty different structure.

834
00:59:30.700 --> 00:59:32.700
Even though AI was inspired by the brain.

835
00:59:32.700 --> 00:59:34.700
That's what Jeffrey Hinton was telling me.

836
00:59:34.700 --> 00:59:38.700
He was telling me that like much of the breakthroughs that have made AI what it is today

837
00:59:38.700 --> 00:59:42.700
came from understanding how the brain works.

838
00:59:42.700 --> 00:59:46.700
Yeah, but that's interesting because Hinton isn't incentivized to say that.

839
00:59:46.700 --> 00:59:50.700
But a neuroscientist, he's incentivized to say that.

840
00:59:50.700 --> 00:59:56.700
People doing AI, of course, are paying a lot of attention to how this is structured like the brain.

841
00:59:56.700 --> 01:00:01.700
Because before that people would do things like probability theory or rules.

842
01:00:01.700 --> 01:00:06.700
They would try to do AI by trying to say, okay, if this then do that.

843
01:00:06.700 --> 01:00:10.700
But when people start doing artificial neural networks that led to a lot of success.

844
01:00:10.700 --> 01:00:16.700
I'm only pointing out that the artificial neural network looks a lot like the brain on the surface.

845
01:00:16.700 --> 01:00:18.700
He said, hey, you got units and you got connections.

846
01:00:18.700 --> 01:00:20.700
But beyond that, there's a lot of differences.

847
01:00:20.700 --> 01:00:24.700
And why are those differences significant as it relates to what's possible?

848
01:00:24.700 --> 01:00:28.700
Because what we've developed is this a new species essentially.

849
01:00:28.700 --> 01:00:32.700
That is incredibly impressive, but it ain't a human brain.

850
01:00:32.700 --> 01:00:34.700
It's different than a human brain.

851
01:00:34.700 --> 01:00:38.700
There may be all kinds of similarities, things that we even come to understand are similar.

852
01:00:38.700 --> 01:00:40.700
But there are so many differences.

853
01:00:40.700 --> 01:00:42.700
Here's an example.

854
01:00:42.700 --> 01:00:46.700
We humans do one trial learning all the time, meaning if I say,

855
01:00:46.700 --> 01:00:50.700
or when you were a kid and your mom said, hey, Steven, this is a pomegranate.

856
01:00:50.700 --> 01:00:52.700
You say, okay, pomegranate, got it.

857
01:00:52.700 --> 01:00:56.700
But you can't, when you're training up an artificial neural network,

858
01:00:56.700 --> 01:01:02.700
like at OpenAI or Gemini or Anthropic, you have to give thousands or millions of examples of everything

859
01:01:02.700 --> 01:01:06.700
to learn anything. There's no one trial learning on those systems.

860
01:01:06.700 --> 01:01:10.700
And they have to be trained at the cost of billions of dollars.

861
01:01:10.700 --> 01:01:14.700
Then they can do a run where you ask a question and it answers the question.

862
01:01:14.700 --> 01:01:18.700
But brains in the real world don't have that luxury of having a training phase

863
01:01:18.700 --> 01:01:20.700
and then an action phase.

864
01:01:20.700 --> 01:01:22.700
We have to learn on the fly. It's very different.

865
01:01:22.700 --> 01:01:28.700
So I guess the pattern question is, does it change what's possible

866
01:01:28.700 --> 01:01:34.700
for the brain versus the artificial neural networks we see in AI?

867
01:01:34.700 --> 01:01:36.700
Is there some limitation based on what you've just said?

868
01:01:36.700 --> 01:01:40.700
That means this brain in front of me, this human brain in front of me,

869
01:01:40.700 --> 01:01:42.700
will always be better than the AI at something.

870
01:01:42.700 --> 01:01:46.700
Because I'm trying to track forward about what this means for the future of humans.

871
01:01:46.700 --> 01:01:50.700
I think it's an interesting question that we'll have to see.

872
01:01:50.700 --> 01:01:56.700
But it's clearly the case that we know what it is to be a human from the inside.

873
01:01:56.700 --> 01:02:00.700
And when I'm making a model of you and who you are and you're making a model of me,

874
01:02:00.700 --> 01:02:04.700
we have assumptions about what it is like to be a human.

875
01:02:04.700 --> 01:02:06.700
AI only watches human behavior from the outside.

876
01:02:06.700 --> 01:02:08.700
And so it can tell a lot of great stuff.

877
01:02:08.700 --> 01:02:12.700
But it doesn't really know what it is to be a human.

878
01:02:12.700 --> 01:02:16.700
So if I ask you some question about what would it be like if this or that happened,

879
01:02:16.700 --> 01:02:20.700
it can answer based on observing lots of things.

880
01:02:20.700 --> 01:02:22.700
But it can only ever know from the outside.

881
01:02:22.700 --> 01:02:24.700
In terms of why that matters.

882
01:02:24.700 --> 01:02:28.700
Because if I ask my AI, my fiancee's been like this today

883
01:02:28.700 --> 01:02:30.700
or if I ask my best friend, my fiancee's been like this today,

884
01:02:30.700 --> 01:02:34.700
if both of them give me the same useful answer, it doesn't really matter what's going on.

885
01:02:34.700 --> 01:02:36.700
I agree with you.

886
01:02:36.700 --> 01:02:40.700
I'm actually writing a new podcast on this about what you can tell from the outside

887
01:02:40.700 --> 01:02:44.700
and what you can tell from the inside and whether that difference matters.

888
01:02:44.700 --> 01:02:48.700
Look, an example is, you know, I last got a Tesla with full-soft driving

889
01:02:48.700 --> 01:02:52.700
and I was watching as it was full-soft driving us coming up on a very complicated traffic situation.

890
01:02:52.700 --> 01:02:54.700
I thought, well, what's my car going to do?

891
01:02:54.700 --> 01:02:56.700
How is it possibly going to understand?

892
01:02:56.700 --> 01:02:58.700
But what it did is it slowed down and came to a stop,

893
01:02:58.700 --> 01:03:00.700
which was exactly the right thing.

894
01:03:00.700 --> 01:03:02.700
And I thought, oh, that's interesting.

895
01:03:02.700 --> 01:03:06.700
Algorithmically, it might think of it very differently than I am thinking about the situation.

896
01:03:06.700 --> 01:03:08.700
It doesn't matter. It comes to the same conclusion.

897
01:03:08.700 --> 01:03:10.700
It's in the same place.

898
01:03:10.700 --> 01:03:14.700
We have yet to see where these differences matter

899
01:03:14.700 --> 01:03:16.700
and what it is to be a human.

900
01:03:16.700 --> 01:03:18.700
But I can tell you one thing.

901
01:03:18.700 --> 01:03:20.700
We care about other humans.

902
01:03:20.700 --> 01:03:24.700
The only prediction is that there's going to be actually a renaissance in things like live theater

903
01:03:24.700 --> 01:03:26.700
and live performances.

904
01:03:26.700 --> 01:03:28.700
When things first came out like Napster,

905
01:03:28.700 --> 01:03:32.700
everyone thought, okay, that's the death of concerts.

906
01:03:32.700 --> 01:03:34.700
That's the death of musicians.

907
01:03:34.700 --> 01:03:36.700
But in fact, you look at a Taylor Swift concert,

908
01:03:36.700 --> 01:03:40.700
Gajillions of people there paying lots of money.

909
01:03:40.700 --> 01:03:42.700
Everyone loves the thing.

910
01:03:42.700 --> 01:03:44.700
Why? Because they're going to see the real Taylor Swift in person.

911
01:03:44.700 --> 01:03:46.700
And I have noticed, I give a lot of talks on the road.

912
01:03:46.700 --> 01:03:48.700
I have noticed an increase in the number of talks.

913
01:03:48.700 --> 01:03:50.700
Since I came out a few years ago,

914
01:03:50.700 --> 01:03:52.700
the first thing that my friend said to me is,

915
01:03:52.700 --> 01:03:58.700
hey, did you know David that you can use 11 labs and hey Jen

916
01:03:58.700 --> 01:04:00.700
and you can make an avatar of yourself

917
01:04:00.700 --> 01:04:02.700
and you can use your voice

918
01:04:02.700 --> 01:04:04.700
and use Chatsy Petita to generate what you're going to say

919
01:04:04.700 --> 01:04:06.700
and have a fully virtual version of you.

920
01:04:06.700 --> 01:04:08.700
He said, my friend he gives talks to.

921
01:04:08.700 --> 01:04:10.700
He said, maybe we can start doing this and do virtual talks.

922
01:04:10.700 --> 01:04:12.700
I said, nobody's going to want that.

923
01:04:12.700 --> 01:04:16.700
In fact, what's happened is more people want to fly us across the country

924
01:04:16.700 --> 01:04:18.700
to have a stand there in person

925
01:04:18.700 --> 01:04:20.700
because it really matters to see fellow humans.

926
01:04:20.700 --> 01:04:22.700
And I think that's only going to increase.

927
01:04:22.700 --> 01:04:24.700
I completely agree with you.

928
01:04:24.700 --> 01:04:26.700
I think it's so funny.

929
01:04:26.700 --> 01:04:30.700
I did a post on my link to the other day saying that

930
01:04:30.700 --> 01:04:34.700
maybe the like interesting paradox or interesting outcome of AI

931
01:04:34.700 --> 01:04:38.700
is that every other iteration of technology

932
01:04:38.700 --> 01:04:40.700
made us less human.

933
01:04:40.700 --> 01:04:44.700
And maybe the intelligence now has gotten to a point

934
01:04:44.700 --> 01:04:48.700
where it's now forcing us to be more human

935
01:04:48.700 --> 01:04:50.700
because that is all that kind of remains in a way

936
01:04:50.700 --> 01:04:54.700
that maybe the technology has gotten so good

937
01:04:54.700 --> 01:04:56.700
like social media didn't make us more human in any capacity

938
01:04:56.700 --> 01:04:58.700
but maybe this is the moment where it goes,

939
01:04:58.700 --> 01:05:00.700
we've got this now.

940
01:05:00.700 --> 01:05:02.700
Go do what only you as a human can do

941
01:05:02.700 --> 01:05:04.700
which is like go out there, Taylor Swift

942
01:05:04.700 --> 01:05:06.700
and sing in front of people IRL

943
01:05:06.700 --> 01:05:08.700
go and do something in the real world.

944
01:05:08.700 --> 01:05:10.700
Even for nurses and doctors

945
01:05:10.700 --> 01:05:12.700
maybe they shouldn't be filling out admin and paperwork anymore

946
01:05:12.700 --> 01:05:14.700
maybe they should be holding your hand

947
01:05:14.700 --> 01:05:16.700
and giving you in real life care

948
01:05:16.700 --> 01:05:18.700
that only a human could do.

949
01:05:18.700 --> 01:05:20.700
I totally agree.

950
01:05:20.700 --> 01:05:22.700
And so maybe that's the positive upside to all of this

951
01:05:22.700 --> 01:05:24.700
is finally, we've been on this journey with technology

952
01:05:24.700 --> 01:05:26.700
and finally it's delivered upon its promise.

953
01:05:26.700 --> 01:05:28.700
I totally agree.

954
01:05:28.700 --> 01:05:30.700
And by the way, AI relationships

955
01:05:30.700 --> 01:05:32.700
by one estimate there's a billion people

956
01:05:32.700 --> 01:05:34.700
having relationships with AI

957
01:05:34.700 --> 01:05:36.700
like a girlfriend or boyfriend kind of thing.

958
01:05:36.700 --> 01:05:40.700
Okay, and so for people like us who grew up before that existed,

959
01:05:40.700 --> 01:05:42.700
we think, oh my gosh, that's weird.

960
01:05:42.700 --> 01:05:44.700
But in fact, I think it might become helpful

961
01:05:44.700 --> 01:05:46.700
because it can be a sandbox

962
01:05:46.700 --> 01:05:48.700
as long as we have the proper feedback.

963
01:05:48.700 --> 01:05:51.700
In the end, we have millions of years of evolution

964
01:05:51.700 --> 01:05:54.700
driving us towards being with the person you love

965
01:05:54.700 --> 01:05:56.700
touching another human being, watching the stars

966
01:05:56.700 --> 01:05:58.700
taking her out to dinner with your parents.

967
01:05:58.700 --> 01:06:00.700
Like, we care about that.

968
01:06:00.700 --> 01:06:02.700
And so this worry that people sometimes talk about

969
01:06:02.700 --> 01:06:04.700
about people are just going to be on their phone

970
01:06:04.700 --> 01:06:06.700
with their AI relationship.

971
01:06:06.700 --> 01:06:08.700
I don't think it's realistic for almost everybody

972
01:06:08.700 --> 01:06:12.700
because it gives us the chance to hopefully sandbox

973
01:06:12.700 --> 01:06:14.700
some things about relationships

974
01:06:14.700 --> 01:06:16.700
and get over some dumb things with relationships.

975
01:06:16.700 --> 01:06:18.700
And then we can actually be with our fellow human.

976
01:06:18.700 --> 01:06:20.700
Counter-argument would be that

977
01:06:20.700 --> 01:06:22.700
maybe there's going to be a biification

978
01:06:22.700 --> 01:06:24.700
of splitting of society where some people

979
01:06:24.700 --> 01:06:26.700
are going to become even more addicted to the technology

980
01:06:26.700 --> 01:06:30.700
because the AI is now much smarter at retention.

981
01:06:30.700 --> 01:06:34.700
Like, I know exactly what I need to say to you

982
01:06:34.700 --> 01:06:38.700
based on your brain, Dr. David,

983
01:06:38.700 --> 01:06:42.700
to make you not put this device down.

984
01:06:42.700 --> 01:06:44.700
Yes, but fundamentally,

985
01:06:44.700 --> 01:06:46.700
I want to be in contact with my wife.

986
01:06:46.700 --> 01:06:48.700
I mean, that's the evolution

987
01:06:48.700 --> 01:06:52.700
of hundreds of millions of years

988
01:06:52.700 --> 01:06:54.700
is that I want to make babies.

989
01:06:54.700 --> 01:06:56.700
I want to go and eat dinner with somebody.

990
01:06:56.700 --> 01:06:58.700
And as much as I might find my phone appealing,

991
01:06:58.700 --> 01:07:00.700
I'm not going to sit it across from me

992
01:07:00.700 --> 01:07:02.700
at a nice Italian restaurant and sit there like that.

993
01:07:02.700 --> 01:07:04.700
I mean, a lot of people, Dave.

994
01:07:04.700 --> 01:07:06.700
I mean, we might have found some right restaurants

995
01:07:06.700 --> 01:07:08.700
because we have a rule where we don't touch off ends

996
01:07:08.700 --> 01:07:10.700
when we're at date night.

997
01:07:10.700 --> 01:07:12.700
And I have to look around and I'm like,

998
01:07:12.700 --> 01:07:14.700
oh, my God.

999
01:07:14.700 --> 01:07:16.700
How are all these guys getting away with this?

1000
01:07:16.700 --> 01:07:18.700
Like, but you see what I'm saying?

1001
01:07:18.700 --> 01:07:20.700
Some people, they just have a different sort of

1002
01:07:20.700 --> 01:07:22.700
proclivity.

1003
01:07:22.700 --> 01:07:24.700
They have a different wiring, which means that

1004
01:07:24.700 --> 01:07:26.700
instead of doing the hard thing of going out there

1005
01:07:26.700 --> 01:07:28.700
and going on a first date and being rejected,

1006
01:07:28.700 --> 01:07:32.700
pornography or a virtual wife might be a substitute for that.

1007
01:07:32.700 --> 01:07:34.700
Yeah.

1008
01:07:34.700 --> 01:07:36.700
Now, I agree with you. There will be bifurcations.

1009
01:07:36.700 --> 01:07:38.700
One question, I don't know the answer to you,

1010
01:07:38.700 --> 01:07:40.700
but one question is, what would that person

1011
01:07:40.700 --> 01:07:42.700
have done in previous generations?

1012
01:07:42.700 --> 01:07:46.700
You know, is it really the case that person would have gone out

1013
01:07:46.700 --> 01:07:48.700
and had a great successful relationship

1014
01:07:48.700 --> 01:07:50.700
or would they always have had troubles relating to people?

1015
01:07:50.700 --> 01:07:54.700
Yeah, I sat with a few

1016
01:07:54.700 --> 01:07:56.700
experts and experts that are studied dopamine.

1017
01:07:56.700 --> 01:07:58.700
Dr. Anna Lemke was one.

1018
01:07:58.700 --> 01:08:00.700
Yeah.

1019
01:08:00.700 --> 01:08:02.700
She's my colleague. She's your colleague.

1020
01:08:02.700 --> 01:08:04.700
She talks a lot about how we all have different types

1021
01:08:04.700 --> 01:08:06.700
of addictive substances.

1022
01:08:06.700 --> 01:08:10.700
And like, you know, we will think like heroin is addictive

1023
01:08:10.700 --> 01:08:12.700
for everybody and alcohol is addictive.

1024
01:08:12.700 --> 01:08:14.700
And I used to think of it on a spectrum,

1025
01:08:14.700 --> 01:08:16.700
but actually she said like for her,

1026
01:08:16.700 --> 01:08:18.700
her addiction was romantic erotic novels.

1027
01:08:18.700 --> 01:08:20.700
Yeah.

1028
01:08:20.700 --> 01:08:22.700
And she almost ruined her relationship because of erotic novels.

1029
01:08:22.700 --> 01:08:24.700
There's something that I would read in just around the bit.

1030
01:08:24.700 --> 01:08:28.700
But so maybe this new technology is particularly addictive

1031
01:08:28.700 --> 01:08:30.700
to a certain type of person.

1032
01:08:30.700 --> 01:08:32.700
Yeah, I think that's exactly right.

1033
01:08:32.700 --> 01:08:34.700
And I think we're going to see that with everything.

1034
01:08:34.700 --> 01:08:36.700
I mean, the wild part about human society

1035
01:08:36.700 --> 01:08:40.700
is that there's so little that we have in common,

1036
01:08:40.700 --> 01:08:42.700
meaning everybody is really different.

1037
01:08:42.700 --> 01:08:44.700
This is something I've studied in my lab for decades.

1038
01:08:44.700 --> 01:08:48.700
Is this issue about what are the subtle differences

1039
01:08:48.700 --> 01:08:50.700
from person to person?

1040
01:08:50.700 --> 01:08:52.700
Like, oh, this person is a psychopath.

1041
01:08:52.700 --> 01:08:54.700
This person has schizophrenia.

1042
01:08:54.700 --> 01:08:56.700
But the more subtle things, I'll just give you an example.

1043
01:08:56.700 --> 01:09:02.700
Like, if I ask you to imagine to visualize, let's say, an ant

1044
01:09:02.700 --> 01:09:06.700
on a purple and white table cloth

1045
01:09:06.700 --> 01:09:10.700
crawling towards a jar of red jelly,

1046
01:09:10.700 --> 01:09:12.700
do you see that in your head like a movie

1047
01:09:12.700 --> 01:09:14.700
or do you have like no particular picture at all

1048
01:09:14.700 --> 01:09:16.700
or somewhere in between?

1049
01:09:16.700 --> 01:09:18.700
What do you experience?

1050
01:09:18.700 --> 01:09:20.700
Ant crawling towards a jar of jelly.

1051
01:09:20.700 --> 01:09:22.700
Yes.

1052
01:09:22.700 --> 01:09:24.700
Yeah, I see a big black ant.

1053
01:09:24.700 --> 01:09:26.700
And then this jar of jelly is like overflowing

1054
01:09:26.700 --> 01:09:28.700
down the sides with a wooden lid on top of it

1055
01:09:28.700 --> 01:09:30.700
and the ant is almost there.

1056
01:09:30.700 --> 01:09:32.700
Oh, wow. Okay. So you have a...

1057
01:09:32.700 --> 01:09:34.700
So what you have, I'm just guessing where you are,

1058
01:09:34.700 --> 01:09:36.700
but you are on the end of the spectrum

1059
01:09:36.700 --> 01:09:38.700
that we call hyperfantasia, which means

1060
01:09:38.700 --> 01:09:40.700
we have very rich visualization.

1061
01:09:40.700 --> 01:09:42.700
You're like seeing it like a picture or a movie.

1062
01:09:42.700 --> 01:09:44.700
Is that accurate?

1063
01:09:44.700 --> 01:09:46.700
Yes. Okay. I happen to meet the other end

1064
01:09:46.700 --> 01:09:48.700
of the spectrum called afantasia,

1065
01:09:48.700 --> 01:09:50.700
where I don't have any visual images at all.

1066
01:09:50.700 --> 01:09:54.700
There's no... I don't see things visually in any way.

1067
01:09:54.700 --> 01:09:56.700
And it turns out the whole population is spread evenly

1068
01:09:56.700 --> 01:09:58.700
along the spectrum.

1069
01:09:58.700 --> 01:10:00.700
I'll just give a quick side note, which is that

1070
01:10:00.700 --> 01:10:03.700
for many years I've been talking with Ed Katmull about this.

1071
01:10:03.700 --> 01:10:05.700
He's the guy who started Pixar films.

1072
01:10:05.700 --> 01:10:07.700
So he's got all the patents on how to do ray tracing

1073
01:10:07.700 --> 01:10:10.700
and how to make these beautiful animated characters, right?

1074
01:10:10.700 --> 01:10:12.700
Ed Katmull is afantasia, like I am.

1075
01:10:12.700 --> 01:10:14.700
And when he learned about this, he got really interested

1076
01:10:14.700 --> 01:10:16.700
and he gave the questionnaire to everybody at Pixar.

1077
01:10:16.700 --> 01:10:18.700
And it turns out many of his best animators and directors

1078
01:10:18.700 --> 01:10:20.700
are afantajic.

1079
01:10:20.700 --> 01:10:22.700
They don't picture anything inside their heads.

1080
01:10:22.700 --> 01:10:25.700
Now, this seems surprising and strange, right?

1081
01:10:25.700 --> 01:10:28.700
But it turns out that if you are an afantasia kid,

1082
01:10:28.700 --> 01:10:30.700
you're going to become better at drawing

1083
01:10:30.700 --> 01:10:32.700
because you have to really pay attention

1084
01:10:32.700 --> 01:10:34.700
to the subject out there and really have a dialogue

1085
01:10:34.700 --> 01:10:36.700
with the page with your pencil.

1086
01:10:36.700 --> 01:10:38.700
Whereas a kid whose hyperfantasia might say,

1087
01:10:38.700 --> 01:10:40.700
oh, I know what a horse looks like and just draws it.

1088
01:10:40.700 --> 01:10:42.700
Okay. So anyway, not tracks.

1089
01:10:42.700 --> 01:10:44.700
Yeah. Yeah.

1090
01:10:44.700 --> 01:10:46.700
So it turns out there's a real spectrum across the population,

1091
01:10:46.700 --> 01:10:48.700
meaning inside your head and my head,

1092
01:10:48.700 --> 01:10:50.700
we're having pretty different experiences.

1093
01:10:50.700 --> 01:10:54.700
But I've studied this along dozens of different axes.

1094
01:10:54.700 --> 01:10:56.700
And everyone's got different things going on.

1095
01:10:56.700 --> 01:10:58.700
Just as one example, do you know about synesthesia?

1096
01:10:58.700 --> 01:11:00.700
Have you ever heard of this?

1097
01:11:00.700 --> 01:11:02.700
Forget, is that forgetting something?

1098
01:11:02.700 --> 01:11:04.700
No, synesthesia is having a blending of the senses.

1099
01:11:04.700 --> 01:11:06.700
So someone with synesthesia might look at letters

1100
01:11:06.700 --> 01:11:08.700
and it triggers a color experience in the head.

1101
01:11:08.700 --> 01:11:10.700
So they look at Jay and that triggers green

1102
01:11:10.700 --> 01:11:12.700
and that triggers blue and whatever.

1103
01:11:12.700 --> 01:11:14.700
It's different for each person.

1104
01:11:14.700 --> 01:11:16.700
Or you might hear music and it triggers a visual experience

1105
01:11:16.700 --> 01:11:18.700
or you might taste something and it puts a feeling

1106
01:11:18.700 --> 01:11:20.700
on your fingertips or whatever.

1107
01:11:20.700 --> 01:11:22.700
It's a blending of the senses.

1108
01:11:22.700 --> 01:11:24.700
At least 3% of the population has this.

1109
01:11:24.700 --> 01:11:26.700
It's not a disease or disorder.

1110
01:11:26.700 --> 01:11:28.700
It's just an alternative perceptual reality.

1111
01:11:28.700 --> 01:11:30.700
So if you have a fantasia,

1112
01:11:30.700 --> 01:11:32.700
does that mean that you can't picture your kids?

1113
01:11:32.700 --> 01:11:36.700
It means that the way I picture them is not visually.

1114
01:11:36.700 --> 01:11:40.700
I mean, there's sort of a very general, but for me,

1115
01:11:40.700 --> 01:11:44.700
it's more motoric imagery and audio imagery.

1116
01:11:44.700 --> 01:11:48.700
I'm imagining talking to them and being with them and being close to them

1117
01:11:48.700 --> 01:11:50.700
and probably some olfactory imagery,

1118
01:11:50.700 --> 01:11:52.700
how they smell and the whole thing.

1119
01:11:52.700 --> 01:11:56.700
I have a very rich notion of what it is to use my kids.

1120
01:11:56.700 --> 01:11:58.700
But it's a pretty terrible visual picture, not much there.

1121
01:11:58.700 --> 01:12:02.700
So imagine people at home have done that same experiment

1122
01:12:02.700 --> 01:12:04.700
while they were listening.

1123
01:12:04.700 --> 01:12:08.700
You can see a picture, an ant walking towards a jar of jam.

1124
01:12:08.700 --> 01:12:10.700
And if they find themselves on the A-fantasia,

1125
01:12:10.700 --> 01:12:12.700
I can't remember the two.

1126
01:12:12.700 --> 01:12:14.700
A-fantasia, yeah, or hyper-fantasia.

1127
01:12:14.700 --> 01:12:16.700
So hyper-fantasia because you can picture it.

1128
01:12:16.700 --> 01:12:18.700
A-fantasia because you can't.

1129
01:12:18.700 --> 01:12:20.700
Yes.

1130
01:12:20.700 --> 01:12:22.700
What does that potentially suggest about nothing?

1131
01:12:22.700 --> 01:12:24.700
Now, here's the interesting parts.

1132
01:12:24.700 --> 01:12:26.700
We've had lots of studies about what this translates to

1133
01:12:26.700 --> 01:12:28.700
in terms of your capacities in the world.

1134
01:12:28.700 --> 01:12:30.700
Nothing. Why does it translate to nothing?

1135
01:12:30.700 --> 01:12:34.700
You can accomplish tasks in a hundred different ways.

1136
01:12:34.700 --> 01:12:36.700
And so some people are doing these very visually.

1137
01:12:36.700 --> 01:12:38.700
Other people are doing it where they're like

1138
01:12:38.700 --> 01:12:40.700
picturing it with their motor systems.

1139
01:12:40.700 --> 01:12:42.700
Others are doing it, you know,

1140
01:12:42.700 --> 01:12:44.700
as I mentioned with sound or smell or whatever.

1141
01:12:44.700 --> 01:12:46.700
Or others are doing it just purely conceptually,

1142
01:12:46.700 --> 01:12:48.700
just thinking through how the steps would go.

1143
01:12:48.700 --> 01:12:50.700
But there's nothing obvious.

1144
01:12:50.700 --> 01:12:54.700
Other than this thing, I mentioned about visual artists

1145
01:12:54.700 --> 01:12:56.700
often being A-fantasia.

1146
01:12:56.700 --> 01:12:58.700
Otherwise, you can kind of accomplish anything.

1147
01:12:58.700 --> 01:13:00.700
I run multiple companies that have multiple sales teams.

1148
01:13:00.700 --> 01:13:02.700
And one of the things as a founder of a company

1149
01:13:02.700 --> 01:13:04.700
that's often confusing is you find it hard to figure out

1150
01:13:04.700 --> 01:13:06.700
where sales are.

1151
01:13:06.700 --> 01:13:08.700
So about 10 years ago, I started using pipe drive

1152
01:13:08.700 --> 01:13:10.700
in my former company.

1153
01:13:10.700 --> 01:13:12.700
And it's also the reason why I switched over

1154
01:13:12.700 --> 01:13:14.700
all of my commercial teams in my current media company

1155
01:13:14.700 --> 01:13:16.700
called Stephen.com to use pipe drive as well.

1156
01:13:16.700 --> 01:13:18.700
Not only did they sponsor this show,

1157
01:13:18.700 --> 01:13:20.700
but they've been in an incredibly effective way

1158
01:13:20.700 --> 01:13:22.700
of scaling our sales engine over the years.

1159
01:13:22.700 --> 01:13:24.700
Pipedrive is an easy to use intelligence CRM.

1160
01:13:24.700 --> 01:13:26.700
And at its very core,

1161
01:13:26.700 --> 01:13:30.700
it makes your sales process visible through one dashboard.

1162
01:13:30.700 --> 01:13:32.700
A visual pipeline showing every deal,

1163
01:13:32.700 --> 01:13:34.700
what stage it's in, what needs to happen next.

1164
01:13:34.700 --> 01:13:36.700
And it's all in real time with no delay.

1165
01:13:36.700 --> 01:13:38.700
It doesn't magically close the deal for you, of course.

1166
01:13:38.700 --> 01:13:42.700
But it does replace complexity with clarity.

1167
01:13:42.700 --> 01:13:44.700
If you want to join over 100,000 companies

1168
01:13:44.700 --> 01:13:46.700
already using pipe drive,

1169
01:13:46.700 --> 01:13:48.700
you can use my link for a 30 day free trial

1170
01:13:48.700 --> 01:13:50.700
with no credit card payment needed.

1171
01:13:50.700 --> 01:13:54.700
Head to pipedrive.com slash CEO to get started.

1172
01:13:54.700 --> 01:13:58.700
That's pipedrive.com slash CEO.

1173
01:13:58.700 --> 01:14:00.700
I'll see you over there.

1174
01:14:00.700 --> 01:14:02.700
This is something that I've made for you.

1175
01:14:02.700 --> 01:14:04.700
I've realised that the Diavisio audience are strivers

1176
01:14:04.700 --> 01:14:06.700
whether it's in business or health.

1177
01:14:06.700 --> 01:14:08.700
We all have big goals that we want to accomplish.

1178
01:14:08.700 --> 01:14:10.700
And one of the things I've learnt

1179
01:14:10.700 --> 01:14:12.700
is that when you aim at the big, big, big goal,

1180
01:14:12.700 --> 01:14:16.700
it can feel incredibly psychologically uncomfortable

1181
01:14:16.700 --> 01:14:18.700
because it's kind of like being stood at the foot

1182
01:14:18.700 --> 01:14:20.700
of Mount Everest and looking upwards.

1183
01:14:20.700 --> 01:14:22.700
The way to accomplish your goals

1184
01:14:22.700 --> 01:14:26.700
is by breaking them down into tiny small steps.

1185
01:14:26.700 --> 01:14:28.700
And we call this an R team the 1%.

1186
01:14:28.700 --> 01:14:30.700
And actually, this philosophy is highly responsible

1187
01:14:30.700 --> 01:14:32.700
for much of our success here.

1188
01:14:32.700 --> 01:14:34.700
So what we've done so that you at home

1189
01:14:34.700 --> 01:14:36.700
can accomplish any big goal that you have

1190
01:14:36.700 --> 01:14:38.700
is we've made these 1% diaries.

1191
01:14:38.700 --> 01:14:40.700
And we've released these last year

1192
01:14:40.700 --> 01:14:42.700
and they all sold out.

1193
01:14:42.700 --> 01:14:44.700
So I asked my team over and over again

1194
01:14:44.700 --> 01:14:46.700
to bring the diaries back,

1195
01:14:46.700 --> 01:14:48.700
but also to introduce some new colours

1196
01:14:48.700 --> 01:14:50.700
and to make some minor tweaks to the diaries.

1197
01:14:50.700 --> 01:14:54.700
So now we have a better range for you.

1198
01:14:54.700 --> 01:14:56.700
So if you have a big goal in mind

1199
01:14:56.700 --> 01:14:58.700
and you need a framework and a process

1200
01:14:58.700 --> 01:15:00.700
and some motivation,

1201
01:15:00.700 --> 01:15:02.700
then I highly recommend you get one of these diaries

1202
01:15:02.700 --> 01:15:04.700
before they all sell out once again.

1203
01:15:04.700 --> 01:15:06.700
And you can get yours at the diary.com.

1204
01:15:06.700 --> 01:15:08.700
And if you want the link,

1205
01:15:08.700 --> 01:15:10.700
the link is in the description below.

1206
01:15:10.700 --> 01:15:12.700
I heard that you might have,

1207
01:15:12.700 --> 01:15:16.700
after many, many decades of people debating this,

1208
01:15:16.700 --> 01:15:18.700
you might have figured out the reason why we dream.

1209
01:15:18.700 --> 01:15:20.700
Yeah, yeah.

1210
01:15:20.700 --> 01:15:22.700
It's actually after millennia of people debating this.

1211
01:15:22.700 --> 01:15:24.700
This is the cool part.

1212
01:15:24.700 --> 01:15:26.700
So remember I mentioned earlier

1213
01:15:26.700 --> 01:15:28.700
that if you go blind,

1214
01:15:28.700 --> 01:15:30.700
the visual cortex of the back of the brain

1215
01:15:30.700 --> 01:15:32.700
gets taken over by hearing and by touch

1216
01:15:32.700 --> 01:15:34.700
and by other things.

1217
01:15:34.700 --> 01:15:36.700
And it's no longer visual cortex.

1218
01:15:36.700 --> 01:15:38.700
Well, what we realized is that

1219
01:15:38.700 --> 01:15:40.700
because we live on a planet

1220
01:15:40.700 --> 01:15:42.700
that rotates into darkness for half the time,

1221
01:15:42.700 --> 01:15:44.700
the visual cortex,

1222
01:15:44.700 --> 01:15:46.700
the visual part of your brain,

1223
01:15:46.700 --> 01:15:48.700
is that a disadvantage.

1224
01:15:48.700 --> 01:15:50.700
So what I realized is that

1225
01:15:50.700 --> 01:15:52.700
the purpose of dreaming is to defend

1226
01:15:52.700 --> 01:15:54.700
the visual territory from takeover,

1227
01:15:54.700 --> 01:15:56.700
from the other senses.

1228
01:15:56.700 --> 01:15:58.700
So every 90 minutes,

1229
01:15:58.700 --> 01:16:00.700
you've got these,

1230
01:16:00.700 --> 01:16:02.700
you've got this very ancient thing in your midbrain

1231
01:16:02.700 --> 01:16:04.700
that shoots random activity

1232
01:16:04.700 --> 01:16:06.700
into the visual system.

1233
01:16:06.700 --> 01:16:08.700
And only the visual system,

1234
01:16:08.700 --> 01:16:10.700
only this very tiny part of the visual system.

1235
01:16:10.700 --> 01:16:12.700
Every 90 minutes, you just blast random activity in here.

1236
01:16:12.700 --> 01:16:14.700
And the reason is,

1237
01:16:14.700 --> 01:16:16.700
we're just defending that territory against takeover.

1238
01:16:16.700 --> 01:16:18.700
Now, the reason that all this came together

1239
01:16:18.700 --> 01:16:20.700
is because our colleagues at Harvard did an experiment

1240
01:16:20.700 --> 01:16:22.700
where they took normally sighted people

1241
01:16:22.700 --> 01:16:24.700
and they blindfolded them tightly for 60 minutes.

1242
01:16:24.700 --> 01:16:28.700
And it turns out that 60 minutes was sufficient

1243
01:16:28.700 --> 01:16:32.700
for the visual cortex to start responding to sound

1244
01:16:32.700 --> 01:16:34.700
and to touch.

1245
01:16:34.700 --> 01:16:36.700
You could start seeing that takeover happening after 60 minutes.

1246
01:16:36.700 --> 01:16:38.700
And that's when we realized,

1247
01:16:38.700 --> 01:16:42.700
wow, this part of the brain really needs a way of defending itself.

1248
01:16:42.700 --> 01:16:44.700
Now, because the brain is a natural storyteller,

1249
01:16:44.700 --> 01:16:46.700
if you blast random activity in there,

1250
01:16:46.700 --> 01:16:48.700
it'll put that together

1251
01:16:48.700 --> 01:16:50.700
and some sort of visual story about what's happening,

1252
01:16:50.700 --> 01:16:52.700
mostly based on what connections are hot from the day.

1253
01:16:52.700 --> 01:16:54.700
But that's why we dream.

1254
01:16:54.700 --> 01:17:00.700
So we dream to stop the other parts of our brain

1255
01:17:00.700 --> 01:17:02.700
overtaking the visual part of our brain,

1256
01:17:02.700 --> 01:17:04.700
overpowering it.

1257
01:17:04.700 --> 01:17:06.700
And I guess ultimately making us go blind.

1258
01:17:06.700 --> 01:17:08.700
Yeah, that's exactly right.

1259
01:17:08.700 --> 01:17:10.700
If we lived on a different kind of planet

1260
01:17:10.700 --> 01:17:12.700
that did not rotate into darkness,

1261
01:17:12.700 --> 01:17:14.700
then we presumably wouldn't dream.

1262
01:17:14.700 --> 01:17:16.700
Would we even need to close our eyes?

1263
01:17:16.700 --> 01:17:18.700
Not necessarily.

1264
01:17:18.700 --> 01:17:20.700
Yeah.

1265
01:17:20.700 --> 01:17:22.700
It may be that in the sleeping state,

1266
01:17:22.700 --> 01:17:24.700
in the state of deep sleep,

1267
01:17:24.700 --> 01:17:26.700
the brain is doing particular things,

1268
01:17:26.700 --> 01:17:28.700
like taking out the trash and cleaning some things up.

1269
01:17:28.700 --> 01:17:30.700
That might be necessary.

1270
01:17:30.700 --> 01:17:32.700
Who knows? But yeah, I don't think we would need to dream.

1271
01:17:32.700 --> 01:17:34.700
We wouldn't need a blast random activity in there.

1272
01:17:34.700 --> 01:17:36.700
If our eyes were always open, for example,

1273
01:17:36.700 --> 01:17:38.700
and it was always light out.

1274
01:17:38.700 --> 01:17:42.700
All that other examples in the animal kingdom.

1275
01:17:42.700 --> 01:17:44.700
Yes, support this.

1276
01:17:44.700 --> 01:17:46.700
Yes, thank you for asking that.

1277
01:17:46.700 --> 01:17:48.700
This is why this new theory about why we dream is taking off

1278
01:17:48.700 --> 01:17:50.700
because we can make quantitative predictions

1279
01:17:50.700 --> 01:17:52.700
across animal species.

1280
01:17:52.700 --> 01:17:54.700
So for example, in our last paper,

1281
01:17:54.700 --> 01:17:56.700
we looked at 25 different species of primates,

1282
01:17:56.700 --> 01:17:58.700
apes and monkeys.

1283
01:17:58.700 --> 01:18:00.700
And we looked at how plastic their brains are.

1284
01:18:00.700 --> 01:18:02.700
In other words, how flexible the whole circuitry was.

1285
01:18:02.700 --> 01:18:04.700
And how much they dream at night,

1286
01:18:04.700 --> 01:18:06.700
which you can tell by looking at rapid eye movements.

1287
01:18:06.700 --> 01:18:08.700
You know, when you dream at night,

1288
01:18:08.700 --> 01:18:10.700
your eyes are shooting back and forth like that.

1289
01:18:10.700 --> 01:18:12.700
It's called our EM rapid eye movement sleep.

1290
01:18:12.700 --> 01:18:14.700
So you can measure that in other animals.

1291
01:18:14.700 --> 01:18:16.700
Their eyes are moving back and forth.

1292
01:18:16.700 --> 01:18:18.700
So we correlated how plastic the brain is,

1293
01:18:18.700 --> 01:18:20.700
and how much dreams sleep you have.

1294
01:18:20.700 --> 01:18:22.700
And it correlates perfectly,

1295
01:18:22.700 --> 01:18:24.700
which is to say humans,

1296
01:18:24.700 --> 01:18:26.700
which are the most plastic,

1297
01:18:26.700 --> 01:18:28.700
have dream sleep all the time.

1298
01:18:28.700 --> 01:18:30.700
And by the way, when you're an infant,

1299
01:18:30.700 --> 01:18:32.700
you have dream sleep for half of your sleep time.

1300
01:18:32.700 --> 01:18:34.700
50% of the time.

1301
01:18:34.700 --> 01:18:36.700
You get less and less dream sleep,

1302
01:18:36.700 --> 01:18:38.700
because you just don't need it as much anymore.

1303
01:18:38.700 --> 01:18:40.700
But anyway, when we look across species,

1304
01:18:40.700 --> 01:18:42.700
it correlates perfectly.

1305
01:18:42.700 --> 01:18:44.700
If you're a monkey that drops into the world,

1306
01:18:44.700 --> 01:18:46.700
sort of already fully baked,

1307
01:18:46.700 --> 01:18:48.700
and you don't need to have much plasticity,

1308
01:18:48.700 --> 01:18:50.700
you don't have much dream sleep either.

1309
01:18:50.700 --> 01:18:52.700
Interesting.

1310
01:18:52.700 --> 01:18:54.700
It seems like a very strange thing.

1311
01:18:54.700 --> 01:18:56.700
It sounds like it's a very strange thing

1312
01:18:56.700 --> 01:18:58.700
for the brain to do.

1313
01:18:58.700 --> 01:19:00.700
But it also is perfectly plausible,

1314
01:19:00.700 --> 01:19:02.700
based on everything you've said.

1315
01:19:02.700 --> 01:19:04.700
The animals dream at night.

1316
01:19:04.700 --> 01:19:06.700
Even like animals at the bottom of the ocean.

1317
01:19:06.700 --> 01:19:08.700
Yes, it's harder to measure stuff all the way

1318
01:19:08.700 --> 01:19:10.700
at the bottom of the ocean,

1319
01:19:10.700 --> 01:19:12.700
but fish do have what is equivalent to dream sleep,

1320
01:19:12.700 --> 01:19:14.700
where you're just zapping activity in there.

1321
01:19:14.700 --> 01:19:16.700
And by the way, even animals that have gone blind,

1322
01:19:16.700 --> 01:19:18.700
like there's a,

1323
01:19:18.700 --> 01:19:20.700
there's a mammal called the blind mole rat,

1324
01:19:20.700 --> 01:19:22.700
which lives in darkness,

1325
01:19:22.700 --> 01:19:24.700
and has eyes, but they're blind,

1326
01:19:24.700 --> 01:19:26.700
because over evolutionary time,

1327
01:19:26.700 --> 01:19:28.700
they've lost vision.

1328
01:19:28.700 --> 01:19:30.700
But they still dream,

1329
01:19:30.700 --> 01:19:32.700
this is so ancient,

1330
01:19:32.700 --> 01:19:34.700
that all animals have to defend themselves

1331
01:19:34.700 --> 01:19:36.700
against the darkness by keeping

1332
01:19:36.700 --> 01:19:38.700
their visual systems going.

1333
01:19:38.700 --> 01:19:40.700
And so even though the animal went blind,

1334
01:19:40.700 --> 01:19:42.700
the rest of the brain didn't catch up.

1335
01:19:42.700 --> 01:19:44.700
I mean, this evolution goes.

1336
01:19:44.700 --> 01:19:46.700
It's funny, because it's kind of like

1337
01:19:46.700 --> 01:19:48.700
that evolution gave us this TV

1338
01:19:48.700 --> 01:19:50.700
that comes on at night time,

1339
01:19:50.700 --> 01:19:52.700
when the real TV,

1340
01:19:52.700 --> 01:19:54.700
our real life, turns off.

1341
01:19:54.700 --> 01:19:56.700
And it just puts on this fake TV set

1342
01:19:56.700 --> 01:19:58.700
to keep that part of the brain

1343
01:19:58.700 --> 01:20:00.700
and it doesn't deteriorate,

1344
01:20:00.700 --> 01:20:02.700
and atrophy.

1345
01:20:02.700 --> 01:20:04.700
It's exactly right.

1346
01:20:04.700 --> 01:20:06.700
Yeah, it's exactly right.

1347
01:20:06.700 --> 01:20:08.700
Which means dreams are quite pointless.

1348
01:20:08.700 --> 01:20:10.700
Outside of just protecting

1349
01:20:10.700 --> 01:20:12.700
on neurological matter.

1350
01:20:12.700 --> 01:20:14.700
I suspect so.

1351
01:20:14.700 --> 01:20:16.700
It might be that the particular pathways

1352
01:20:16.700 --> 01:20:18.700
that get traveled down,

1353
01:20:18.700 --> 01:20:20.700
you know, maybe there's some meaning there.

1354
01:20:20.700 --> 01:20:22.700
I, my own suspicion is that

1355
01:20:22.700 --> 01:20:24.700
it's like if I went to your bookshelf

1356
01:20:24.700 --> 01:20:26.700
and I picked a random book up

1357
01:20:26.700 --> 01:20:28.700
I might find some meaning in that.

1358
01:20:28.700 --> 01:20:30.700
I might say, well, that was just the

1359
01:20:30.700 --> 01:20:32.700
sentence that I needed to hear.

1360
01:20:32.700 --> 01:20:34.700
But it's not really. It's just that it has some meaning to me.

1361
01:20:34.700 --> 01:20:36.700
Anyway, the point is if you blast random activity in there,

1362
01:20:36.700 --> 01:20:38.700
I might dream about something where I wake up

1363
01:20:38.700 --> 01:20:40.700
and say, oh, that was pretty useful.

1364
01:20:40.700 --> 01:20:42.700
But the thing that I think it's overlooked

1365
01:20:42.700 --> 01:20:44.700
is that most dreams are totally useless and bizarre.

1366
01:20:44.700 --> 01:20:46.700
Dr. David, what is the most important thing

1367
01:20:46.700 --> 01:20:48.700
we haven't talked about that we should have talked about?

1368
01:20:48.700 --> 01:20:50.700
As it specifically relates to

1369
01:20:50.700 --> 01:20:52.700
people that are trying to

1370
01:20:52.700 --> 01:20:54.700
improve their lives,

1371
01:20:54.700 --> 01:20:56.700
get better, whatever that subjective mission is.

1372
01:20:56.700 --> 01:20:58.700
I'm the brain.

1373
01:20:58.700 --> 01:21:00.700
There are probably a lot of things.

1374
01:21:00.700 --> 01:21:02.700
But I got to say,

1375
01:21:02.700 --> 01:21:04.700
the thing that I've been thinking about so much lately

1376
01:21:04.700 --> 01:21:06.700
is just about our political

1377
01:21:06.700 --> 01:21:08.700
interfacing with one another.

1378
01:21:08.700 --> 01:21:10.700
And so I do feel that

1379
01:21:10.700 --> 01:21:14.700
really learning the skills of dialogue

1380
01:21:14.700 --> 01:21:16.700
with our fellow humans,

1381
01:21:16.700 --> 01:21:18.700
where we listen to what they're saying

1382
01:21:18.700 --> 01:21:20.700
and try to better understand what their

1383
01:21:20.700 --> 01:21:22.700
internal model is.

1384
01:21:22.700 --> 01:21:24.700
And I think it's really useful to

1385
01:21:24.700 --> 01:21:26.700
agree with them.

1386
01:21:26.700 --> 01:21:28.700
But it is saying, hey,

1387
01:21:28.700 --> 01:21:30.700
somebody is coming from this perspective.

1388
01:21:30.700 --> 01:21:32.700
Let me see if I can understand that.

1389
01:21:32.700 --> 01:21:34.700
I think that matters a lot.

1390
01:21:34.700 --> 01:21:36.700
And I also think that

1391
01:21:36.700 --> 01:21:38.700
because we're so highly predisposed

1392
01:21:38.700 --> 01:21:40.700
for in groups and out groups,

1393
01:21:40.700 --> 01:21:42.700
it's really useful to figure out

1394
01:21:42.700 --> 01:21:44.700
how to complexify those relationships,

1395
01:21:44.700 --> 01:21:46.700
meaning how do you figure out

1396
01:21:46.700 --> 01:21:48.700
all the things that cross-cut

1397
01:21:48.700 --> 01:21:50.700
in the relationship so that you say,

1398
01:21:50.700 --> 01:21:52.700
anyway, because actually

1399
01:21:52.700 --> 01:21:54.700
they belong to the same group I do.

1400
01:21:54.700 --> 01:21:56.700
And they love surfing as much as I do

1401
01:21:56.700 --> 01:21:58.700
and they love golden retriever dogs

1402
01:21:58.700 --> 01:22:00.700
and they grew up in my hometown

1403
01:22:00.700 --> 01:22:02.700
and whatever.

1404
01:22:02.700 --> 01:22:04.700
Finding those things explicitly

1405
01:22:04.700 --> 01:22:06.700
helps the brain

1406
01:22:06.700 --> 01:22:08.700
to keep these circuits on

1407
01:22:08.700 --> 01:22:10.700
that are involved in seeing another person

1408
01:22:10.700 --> 01:22:12.700
as a person.

1409
01:22:12.700 --> 01:22:14.700
We have all the social circuitry

1410
01:22:14.700 --> 01:22:16.700
that is all about

1411
01:22:16.700 --> 01:22:18.700
understanding other people

1412
01:22:18.700 --> 01:22:20.700
and when things get dehumanized,

1413
01:22:20.700 --> 01:22:22.700
that actually gets dialed away down.

1414
01:22:22.700 --> 01:22:24.700
When we look at, you know,

1415
01:22:24.700 --> 01:22:26.700
let's say a homeless person

1416
01:22:26.700 --> 01:22:28.700
or a drug addict

1417
01:22:28.700 --> 01:22:30.700
or someone who we think of as our enemy

1418
01:22:30.700 --> 01:22:32.700
or an out group,

1419
01:22:32.700 --> 01:22:34.700
that gets dialed down.

1420
01:22:34.700 --> 01:22:36.700
So we don't think of them as a person anymore.

1421
01:22:36.700 --> 01:22:38.700
We think of them as an object

1422
01:22:38.700 --> 01:22:40.700
to get around.

1423
01:22:40.700 --> 01:22:42.700
So this is what I think is really important

1424
01:22:42.700 --> 01:22:44.700
is figuring out what we can do

1425
01:22:44.700 --> 01:22:46.700
to keep that social circuitry still going

1426
01:22:46.700 --> 01:22:48.700
in my context and conversation.

1427
01:22:48.700 --> 01:22:50.700
And this is one of the most important things

1428
01:22:50.700 --> 01:22:52.700
we can do as citizens

1429
01:22:52.700 --> 01:22:54.700
in a rapidly changing world.

1430
01:22:54.700 --> 01:22:56.700
As it relates to

1431
01:22:56.700 --> 01:22:58.700
things like dementia,

1432
01:22:58.700 --> 01:23:00.700
which I know is a fear

1433
01:23:00.700 --> 01:23:02.700
that a lot of people have a lot of people

1434
01:23:02.700 --> 01:23:04.700
suffering with dementia.

1435
01:23:04.700 --> 01:23:06.700
I think increasingly, in fact,

1436
01:23:06.700 --> 01:23:08.700
if I was trying to stave off dementia,

1437
01:23:08.700 --> 01:23:10.700
what advice would you give me, David?

1438
01:23:10.700 --> 01:23:12.700
Yeah, keep your brain active.

1439
01:23:12.700 --> 01:23:14.700
Keep it active till the day you die

1440
01:23:14.700 --> 01:23:16.700
and it's something like, you know,

1441
01:23:16.700 --> 01:23:18.700
Sudoku, drop it

1442
01:23:18.700 --> 01:23:20.700
and pick up something that you're not good at.

1443
01:23:20.700 --> 01:23:22.700
And in simple times, why?

1444
01:23:22.700 --> 01:23:24.700
It's because you're forcing your brain to make changes.

1445
01:23:24.700 --> 01:23:26.700
Otherwise, your brain says,

1446
01:23:26.700 --> 01:23:28.700
okay, I got this. I got the world.

1447
01:23:28.700 --> 01:23:30.700
I understand what's going on.

1448
01:23:30.700 --> 01:23:32.700
There's no real particular need for me to change.

1449
01:23:32.700 --> 01:23:34.700
And the fact is that the structure of the brain

1450
01:23:34.700 --> 01:23:36.700
is always degenerating.

1451
01:23:36.700 --> 01:23:38.700
And when you get something like a disease

1452
01:23:38.700 --> 01:23:40.700
like Alzheimer's disease,

1453
01:23:40.700 --> 01:23:42.700
it degenerates much faster.

1454
01:23:42.700 --> 01:23:44.700
And it's fashioning new

1455
01:23:44.700 --> 01:23:46.700
pads that had not been walked before.

1456
01:23:46.700 --> 01:23:48.700
So that there's more

1457
01:23:48.700 --> 01:23:50.700
to degenerate,

1458
01:23:50.700 --> 01:23:52.700
which gives me more leftover

1459
01:23:52.700 --> 01:23:54.700
once that degeneration begins.

1460
01:23:54.700 --> 01:23:56.700
Yeah, I think that's a good way to look at it.

1461
01:23:56.700 --> 01:23:58.700
Your pathways are falling apart.

1462
01:23:58.700 --> 01:24:00.700
And if you can build new pathways,

1463
01:24:00.700 --> 01:24:02.700
which requires effort,

1464
01:24:02.700 --> 01:24:04.700
you have to actually care and pursue

1465
01:24:04.700 --> 01:24:06.700
and do the thing,

1466
01:24:06.700 --> 01:24:08.700
even as parts of the thing you've fallen apart,

1467
01:24:08.700 --> 01:24:10.700
you still have ways of getting from A to B.

1468
01:24:10.700 --> 01:24:12.700
In terms of chemicals,

1469
01:24:12.700 --> 01:24:14.700
or supplement,

1470
01:24:14.700 --> 01:24:16.700
or food, I don't know.

1471
01:24:16.700 --> 01:24:18.700
Yeah, obviously there's just been a lot more emphasis

1472
01:24:18.700 --> 01:24:20.700
on getting good sleep and good diet.

1473
01:24:20.700 --> 01:24:22.700
And this stuff really matters.

1474
01:24:22.700 --> 01:24:24.700
I think that's really useful for the brain.

1475
01:24:24.700 --> 01:24:26.700
I mean, it's fascinating to watch what's happened

1476
01:24:26.700 --> 01:24:28.700
in the latest generation in terms of alcohol consumption.

1477
01:24:28.700 --> 01:24:30.700
I live up in Silicon Valley,

1478
01:24:30.700 --> 01:24:32.700
and there's a lot of people

1479
01:24:32.700 --> 01:24:34.700
who have wineries just north of me.

1480
01:24:34.700 --> 01:24:36.700
And they're like selling half their acreage.

1481
01:24:36.700 --> 01:24:38.700
It's absolutely fascinating to see what's happening there.

1482
01:24:38.700 --> 01:24:40.700
Who's in her 20s,

1483
01:24:40.700 --> 01:24:42.700
who said that she's in favor

1484
01:24:42.700 --> 01:24:44.700
of bringing drinking back.

1485
01:24:44.700 --> 01:24:46.700
Why? Because she said,

1486
01:24:46.700 --> 01:24:48.700
we go to parties, and everything's so awkward,

1487
01:24:48.700 --> 01:24:50.700
and no one knows how to talk to one another.

1488
01:24:50.700 --> 01:24:52.700
And so they're missing something else.

1489
01:24:52.700 --> 01:24:54.700
They're missing the dumb mistakes category

1490
01:24:54.700 --> 01:24:56.700
that we all got to enjoy growing up.

1491
01:24:56.700 --> 01:24:58.700
So it is a really interesting balance

1492
01:24:58.700 --> 01:25:02.700
of how abstinuous one wants to become.

1493
01:25:02.700 --> 01:25:04.700
David, we have a closing tradition where the last guest

1494
01:25:04.700 --> 01:25:06.700
leaves a question to the next guest.

1495
01:25:06.700 --> 01:25:08.700
Well, question left for you is,

1496
01:25:08.700 --> 01:25:10.700
what do you wish most for our planet

1497
01:25:10.700 --> 01:25:12.700
over the next 10 years?

1498
01:25:18.700 --> 01:25:20.700
The whole list of the top 10.

1499
01:25:20.700 --> 01:25:22.700
Yeah.

1500
01:25:22.700 --> 01:25:24.700
It can't be well-paced.

1501
01:25:24.700 --> 01:25:26.700
You know, I think I would come back to this piece

1502
01:25:26.700 --> 01:25:28.700
about the complexification of relationships,

1503
01:25:28.700 --> 01:25:30.700
which is to say,

1504
01:25:30.700 --> 01:25:32.700
if we could just get a little bit smarter

1505
01:25:32.700 --> 01:25:36.700
about understanding people are out groups

1506
01:25:36.700 --> 01:25:38.700
as being humans with lives,

1507
01:25:38.700 --> 01:25:40.700
with their own thing going on,

1508
01:25:40.700 --> 01:25:44.700
doesn't mean we have to love them or agree with them,

1509
01:25:44.700 --> 01:25:46.700
but if we can just get to that point,

1510
01:25:46.700 --> 01:25:48.700
I don't think we'll ever hit world peace,

1511
01:25:48.700 --> 01:25:50.700
but at least we'd have slightly less polarization.

1512
01:25:50.700 --> 01:25:52.700
So I'm definitely in favor of that,

1513
01:25:52.700 --> 01:25:54.700
and I do think it's possible,

1514
01:25:54.700 --> 01:25:56.700
and I do think AI can help us get there

1515
01:25:56.700 --> 01:25:58.700
by challenging us on these points,

1516
01:25:58.700 --> 01:26:00.700
and saying, hey,

1517
01:26:00.700 --> 01:26:03.700
that group that you've already dismissed as an outgroup,

1518
01:26:03.700 --> 01:26:06.700
what if I told you this story about this person?

1519
01:26:06.700 --> 01:26:08.700
What if I introduced you to this person?

1520
01:26:08.700 --> 01:26:10.700
That kind of stuff.

1521
01:26:10.700 --> 01:26:12.700
And having, there's all kinds of social movements

1522
01:26:12.700 --> 01:26:14.700
that have sprung up that allow people

1523
01:26:14.700 --> 01:26:16.700
of different political opinions to come together in a room

1524
01:26:16.700 --> 01:26:18.700
and talk with one another.

1525
01:26:18.700 --> 01:26:20.700
Again, it's not that anyone has to change their mind,

1526
01:26:20.700 --> 01:26:22.700
but they can say, hey, you know what?

1527
01:26:22.700 --> 01:26:24.700
I really liked that person.

1528
01:26:24.700 --> 01:26:26.700
I thought that was a cool person, a sweet person,

1529
01:26:26.700 --> 01:26:28.700
a nice person.

1530
01:26:28.700 --> 01:26:30.700
It's my own eyes, has a different opinion on this idea.

1531
01:26:30.700 --> 01:26:32.700
Is that wishful thinking to some degree?

1532
01:26:32.700 --> 01:26:34.700
I don't think so, because these things are happening

1533
01:26:34.700 --> 01:26:36.700
all over the place.

1534
01:26:36.700 --> 01:26:38.700
But the macro is division, isn't it?

1535
01:26:38.700 --> 01:26:40.700
It's polarization echo chambers.

1536
01:26:40.700 --> 01:26:42.700
I think there's now 20 social networks,

1537
01:26:42.700 --> 01:26:45.700
some crazy number that have more than 20 million people on them,

1538
01:26:45.700 --> 01:26:47.700
which means that social networks are splintering off

1539
01:26:47.700 --> 01:26:49.700
into niches and interests,

1540
01:26:49.700 --> 01:26:51.700
and there's like rumble and bumble,

1541
01:26:51.700 --> 01:26:53.700
and there's like threads and eggs

1542
01:26:53.700 --> 01:26:55.700
and Facebook, snap, Instagram,

1543
01:26:55.700 --> 01:26:58.700
and what we're seeing is more and more interest group.

1544
01:26:58.700 --> 01:27:00.700
And also the other thing with algorithms

1545
01:27:00.700 --> 01:27:02.700
is we went from having a social graph

1546
01:27:02.700 --> 01:27:04.700
where if I had 1,000 people follow me,

1547
01:27:04.700 --> 01:27:06.700
those 1,000 people would see my stuff,

1548
01:27:06.700 --> 01:27:07.700
to now these interest graphs,

1549
01:27:07.700 --> 01:27:10.700
where it doesn't matter if I have one follower or one million followers,

1550
01:27:10.700 --> 01:27:13.700
the algorithm is going to decide who's interested in that thing

1551
01:27:13.700 --> 01:27:14.700
and it's going to serve it to them

1552
01:27:14.700 --> 01:27:16.700
because that's the most attentive thing

1553
01:27:16.700 --> 01:27:19.700
if you're a publicly listed company that's driven by our revenue.

1554
01:27:19.700 --> 01:27:21.700
So you've got this algorithm that's actually forcing you

1555
01:27:21.700 --> 01:27:23.700
into this entire, you know,

1556
01:27:23.700 --> 01:27:24.700
tighter, tighter echo chambers.

1557
01:27:24.700 --> 01:27:26.700
And even if someone that's been on social media

1558
01:27:26.700 --> 01:27:27.700
for 15 years, some ran social media companies,

1559
01:27:27.700 --> 01:27:29.700
this is one of the great things I've noticed is,

1560
01:27:29.700 --> 01:27:31.700
when I had a million followers back in the day,

1561
01:27:31.700 --> 01:27:32.700
I would reach those people,

1562
01:27:32.700 --> 01:27:34.700
because they'd hit follow or subscribe.

1563
01:27:34.700 --> 01:27:36.700
Now, even on our YouTube channel,

1564
01:27:36.700 --> 01:27:38.700
61% of you don't subscribe,

1565
01:27:38.700 --> 01:27:41.700
and please subscribe.

1566
01:27:41.700 --> 01:27:44.700
And that's in part because the algorithm

1567
01:27:44.700 --> 01:27:47.700
is now doing the work of deciding who to show it to,

1568
01:27:47.700 --> 01:27:50.700
who will be retained.

1569
01:27:50.700 --> 01:27:53.700
Here's what I would say.

1570
01:27:53.700 --> 01:27:55.700
There's absolutely nothing new about echo chambers

1571
01:27:55.700 --> 01:27:58.700
because it was always the case that your neighbors

1572
01:27:58.700 --> 01:28:00.700
and your community and whatever,

1573
01:28:00.700 --> 01:28:02.700
that's what you thought was reality.

1574
01:28:02.700 --> 01:28:04.700
I'm actually quite optimistic about the existence,

1575
01:28:04.700 --> 01:28:06.700
the mere existence of the internet,

1576
01:28:06.700 --> 01:28:08.700
because at least we are exposed to the fact

1577
01:28:08.700 --> 01:28:10.700
that there are lots of different points of view.

1578
01:28:10.700 --> 01:28:12.700
It used to be in places like the USSR.

1579
01:28:12.700 --> 01:28:14.700
They controlled the media tightly

1580
01:28:14.700 --> 01:28:18.700
so that everything you saw was a news-approved story.

1581
01:28:18.700 --> 01:28:20.700
But now you see all the points of view.

1582
01:28:20.700 --> 01:28:22.700
Now many of them might try to be crazy in whatever,

1583
01:28:22.700 --> 01:28:24.700
but at least you know that there are people out there

1584
01:28:24.700 --> 01:28:25.700
that believe in that,

1585
01:28:25.700 --> 01:28:27.700
and I think that's really useful.

1586
01:28:27.700 --> 01:28:29.700
If I had to decide between state control

1587
01:28:29.700 --> 01:28:30.700
where there's a single story

1588
01:28:30.700 --> 01:28:34.700
or seeing the whole messy spectrum of opinions,

1589
01:28:34.700 --> 01:28:36.700
I'd rather see the latter.

1590
01:28:36.700 --> 01:28:37.700
What about the middle?

1591
01:28:37.700 --> 01:28:39.700
You know, one of the phrases,

1592
01:28:39.700 --> 01:28:41.700
that's, again, a principle that's helped me think is

1593
01:28:41.700 --> 01:28:42.700
that the treat is in the middle.

1594
01:28:42.700 --> 01:28:44.700
And generally, I try to understand what the middle looks like.

1595
01:28:44.700 --> 01:28:46.700
So you've got state control over here.

1596
01:28:46.700 --> 01:28:48.700
You've got aggressive algorithm.

1597
01:28:48.700 --> 01:28:51.700
That's sort of reinforcing whatever you currently believe.

1598
01:28:51.700 --> 01:28:53.700
Is there not some kind of middle ground

1599
01:28:53.700 --> 01:28:57.700
where the algorithms have to let up a little bit?

1600
01:28:57.700 --> 01:28:59.700
And of course, we're not going to go for state controlled.

1601
01:28:59.700 --> 01:29:01.700
Here's my prediction in 2026.

1602
01:29:01.700 --> 01:29:03.700
Is that there is a market opportunity

1603
01:29:03.700 --> 01:29:06.700
for a new social media company to come along,

1604
01:29:06.700 --> 01:29:08.700
because everybody is aware of exactly this problem

1605
01:29:08.700 --> 01:29:09.700
that you're pointing out.

1606
01:29:09.700 --> 01:29:11.700
Everyone hates when they serve,

1607
01:29:11.700 --> 01:29:14.700
and they get served exactly what they're supposed to get served,

1608
01:29:14.700 --> 01:29:16.700
and they get off after an hour or two,

1609
01:29:16.700 --> 01:29:18.700
and they feel like they've wasted their lives.

1610
01:29:18.700 --> 01:29:20.700
And there's a real opportunity for a social media company

1611
01:29:20.700 --> 01:29:22.700
to come along and say, you know what?

1612
01:29:22.700 --> 01:29:24.700
We're not building our algorithm like the other guys.

1613
01:29:24.700 --> 01:29:27.700
It's not about just trying to get engagement at any cost

1614
01:29:27.700 --> 01:29:30.700
with, you know, incendiary posts.

1615
01:29:30.700 --> 01:29:34.700
But instead, we're looking for ways to connect people.

1616
01:29:34.700 --> 01:29:39.700
So if you and I both love this particular thing,

1617
01:29:39.700 --> 01:29:42.700
this particular cuisine or location

1618
01:29:42.700 --> 01:29:44.700
or whatever it is, we get connected,

1619
01:29:44.700 --> 01:29:46.700
we see each other's stuff,

1620
01:29:46.700 --> 01:29:48.700
and the algorithm carefully,

1621
01:29:48.700 --> 01:29:50.700
temporarily sequences things,

1622
01:29:50.700 --> 01:29:52.700
so that we come to have a certain connection threshold

1623
01:29:52.700 --> 01:29:54.700
before we find out, whoa,

1624
01:29:54.700 --> 01:29:56.700
you have a totally different political opinion

1625
01:29:56.700 --> 01:29:58.700
than I do on subject X.

1626
01:29:58.700 --> 01:30:00.700
Wow, I didn't know that.

1627
01:30:00.700 --> 01:30:01.700
But I really like Stephen,

1628
01:30:01.700 --> 01:30:03.700
so I'm going to lean in and listen a little bit more.

1629
01:30:03.700 --> 01:30:05.700
I think this is very easy to do,

1630
01:30:05.700 --> 01:30:07.700
and I think the connection will be part of the selling point

1631
01:30:07.700 --> 01:30:08.700
of the media company is saying,

1632
01:30:08.700 --> 01:30:10.700
hey, we are here,

1633
01:30:10.700 --> 01:30:12.700
not to enrage you,

1634
01:30:12.700 --> 01:30:14.700
but to actually build connection.

1635
01:30:14.700 --> 01:30:16.700
It sounds like how social media started.

1636
01:30:16.700 --> 01:30:18.700
Yeah, and I think there's a return.

1637
01:30:18.700 --> 01:30:21.700
I think there's probably a near-science basis

1638
01:30:21.700 --> 01:30:23.700
as to why we ended up.

1639
01:30:23.700 --> 01:30:24.700
Yeah.

1640
01:30:24.700 --> 01:30:26.700
No, it's an economic basis.

1641
01:30:26.700 --> 01:30:27.700
Yeah, yeah, yeah.

1642
01:30:27.700 --> 01:30:29.700
But the fact is there's now an economic opportunity

1643
01:30:29.700 --> 01:30:31.700
and everyone sees the landscape.

1644
01:30:31.700 --> 01:30:33.700
What I'm trying to say is that that social network

1645
01:30:33.700 --> 01:30:35.700
wouldn't be that attentive by design

1646
01:30:35.700 --> 01:30:37.700
because it wouldn't trigger my dopamine.

1647
01:30:37.700 --> 01:30:39.700
It wouldn't be a slot machine.

1648
01:30:39.700 --> 01:30:41.700
TikTok is a slot machine.

1649
01:30:41.700 --> 01:30:43.700
Ping, ping, randomize returns.

1650
01:30:43.700 --> 01:30:45.700
Ping, ping, ping, dopamine, ping, ping, ping.

1651
01:30:45.700 --> 01:30:47.700
So there's other social networks

1652
01:30:47.700 --> 01:30:49.700
that wasn't playing with my dopamine in such way.

1653
01:30:49.700 --> 01:30:51.700
I don't know whether I'd be addicted enough to return.

1654
01:30:51.700 --> 01:30:53.700
Therefore, they wouldn't sell their ads.

1655
01:30:53.700 --> 01:30:54.700
The economic return.

1656
01:30:54.700 --> 01:30:55.700
Therefore, they wouldn't do very well.

1657
01:30:55.700 --> 01:30:56.700
Here's the thing.

1658
01:30:56.700 --> 01:30:58.700
I don't know if the story is that simple

1659
01:30:58.700 --> 01:31:00.700
that we all want to do slot machines all the time.

1660
01:31:00.700 --> 01:31:01.700
I don't think we do.

1661
01:31:01.700 --> 01:31:02.700
Exactly.

1662
01:31:02.700 --> 01:31:04.700
Because the fact is that a lot of people go to Las Vegas

1663
01:31:04.700 --> 01:31:06.700
and do slot machines sometime.

1664
01:31:06.700 --> 01:31:08.700
But we don't do that all the time.

1665
01:31:08.700 --> 01:31:09.700
It's kind of rare, actually.

1666
01:31:09.700 --> 01:31:12.700
What we really desire are meaningful connections.

1667
01:31:12.700 --> 01:31:14.700
We really desire feeling like,

1668
01:31:14.700 --> 01:31:15.700
hey, you know what?

1669
01:31:15.700 --> 01:31:17.700
I met this person online that I'm following

1670
01:31:17.700 --> 01:31:18.700
and he's following me.

1671
01:31:18.700 --> 01:31:21.700
And we really connect on all these points.

1672
01:31:21.700 --> 01:31:24.700
And oh, by the way, I then found out, interestingly,

1673
01:31:24.700 --> 01:31:26.700
he's got a totally different opinion about Iran

1674
01:31:26.700 --> 01:31:28.700
or abortion or whatever that I do.

1675
01:31:28.700 --> 01:31:29.700
But that's cool.

1676
01:31:29.700 --> 01:31:31.700
Now we're listening to each other.

1677
01:31:31.700 --> 01:31:33.700
It kind of goes back to your point now.

1678
01:31:33.700 --> 01:31:35.700
The very start we were talking about, you know,

1679
01:31:35.700 --> 01:31:37.700
having an internal battle, like, do I want the cookie

1680
01:31:37.700 --> 01:31:38.700
or do I want the salad?

1681
01:31:38.700 --> 01:31:41.700
Yeah, unfortunately in the world, we live in, you know,

1682
01:31:41.700 --> 01:31:43.700
the cookie is going to give me a dopamine hit.

1683
01:31:43.700 --> 01:31:45.700
Yes, but we don't eat cookies all the time.

1684
01:31:45.700 --> 01:31:46.700
This is the point.

1685
01:31:46.700 --> 01:31:48.700
We do eat salads much of the time

1686
01:31:48.700 --> 01:31:50.700
because we're not just unconscious

1687
01:31:50.700 --> 01:31:52.700
our time and time that are doing the cookies.

1688
01:31:52.700 --> 01:31:54.700
I talk to David Eagleman.

1689
01:31:54.700 --> 01:31:55.700
Thank you so much for the work that you do.

1690
01:31:55.700 --> 01:31:58.700
I'm going to link your book below so everyone can read this book.

1691
01:31:58.700 --> 01:31:59.700
You've got a new book on the way,

1692
01:31:59.700 --> 01:32:00.700
which I'm very excited about as well.

1693
01:32:00.700 --> 01:32:01.700
What's that book going to be about?

1694
01:32:01.700 --> 01:32:02.700
And when is that out?

1695
01:32:02.700 --> 01:32:05.700
That's about the Ulysses contract and they'll come out in 2027.

1696
01:32:05.700 --> 01:32:06.700
Okay.

1697
01:32:06.700 --> 01:32:08.700
For anyone that wants to know how to change your life

1698
01:32:08.700 --> 01:32:10.700
by changing your brain, I think this is the perfect book to read.

1699
01:32:10.700 --> 01:32:14.700
It's a New York Times bestselling author.

1700
01:32:14.700 --> 01:32:16.700
And the book is absolutely fascinating.

1701
01:32:16.700 --> 01:32:19.700
It was actually learning about this subject matter in LiveWides

1702
01:32:19.700 --> 01:32:23.700
that helped me to pursue more of a growth mindset

1703
01:32:23.700 --> 01:32:25.700
and just a growth mentality across my life

1704
01:32:25.700 --> 01:32:27.700
and to realize that if I'm not something now,

1705
01:32:27.700 --> 01:32:29.700
it doesn't mean that I can't be tomorrow.

1706
01:32:29.700 --> 01:32:31.700
So thank you so much for the work that you do, David.

1707
01:32:31.700 --> 01:32:33.700
And it's been truly illuminating.

1708
01:32:33.700 --> 01:32:37.700
I'm sure my neural pathways have expanded in really important ways

1709
01:32:37.700 --> 01:32:38.700
because of this.

1710
01:32:38.700 --> 01:32:39.700
Great.

1711
01:32:39.700 --> 01:32:40.700
Thank you, Stephen.

1712
01:32:59.700 --> 01:33:02.700
You