WEBVTT

1
00:00:00.000 --> 00:00:07.000
We have video evidence of people abusing children and not a single person got arrested.

2
00:00:07.000 --> 00:00:09.000
How can you call that a democracy?

3
00:00:09.000 --> 00:00:11.000
So humanity is at a crossroads.

4
00:00:11.000 --> 00:00:16.000
Where for the first time ever we need to wake up and realize that we're ruled by maniacs.

5
00:00:16.000 --> 00:00:20.000
And what we believe is democracy is not democracy and what we know is not the truth.

6
00:00:20.000 --> 00:00:26.000
Like companies and governments will blame the geopolitical and economic challenges we have on AI.

7
00:00:26.000 --> 00:00:28.000
But the truth is AI is not the enemy.

8
00:00:28.000 --> 00:00:31.000
Like I'm not worried about AI turning against us.

9
00:00:31.000 --> 00:00:34.000
I'm worried about humans telling AI to turn against us.

10
00:00:34.000 --> 00:00:41.000
Like when I worked at Google we were really amazing things, believing that we were making the world a better place and we were.

11
00:00:41.000 --> 00:00:48.000
But then suddenly there is a moment where you recognize that maybe the world will not use what you're making the way you want it to be used.

12
00:00:48.000 --> 00:00:51.000
And sadly this is upon us.

13
00:00:51.000 --> 00:00:53.000
So I have lots of questions.

14
00:00:53.000 --> 00:00:54.000
Okay that's good.

15
00:00:54.000 --> 00:00:56.000
So what's your take on this job destruction point?

16
00:00:56.000 --> 00:01:01.000
What is the risk of these very intelligent models that the creators of these models don't actually understand themselves?

17
00:01:01.000 --> 00:01:03.000
Do you think Sam Ottman's pro-humanity?

18
00:01:03.000 --> 00:01:07.000
How do we get to a point of ethical AI when the incentive structures are so highly competitive?

19
00:01:07.000 --> 00:01:12.000
And then I wonder if there's a path that ends in AI being net positive for humanity.

20
00:01:12.000 --> 00:01:16.000
But somehow we've been pre-programmed to believe that this is upon us and it cannot change.

21
00:01:16.000 --> 00:01:18.000
And I refuse that.

22
00:01:18.000 --> 00:01:20.000
So we will talk about the solutions.

23
00:01:20.000 --> 00:01:21.000
But are you optimistic?

24
00:01:21.000 --> 00:01:23.000
I'm very optimistic about the future.

25
00:01:23.000 --> 00:01:25.000
I'm not optimistic about the next year.

26
00:01:25.000 --> 00:01:26.000
Why the next year?

27
00:01:26.000 --> 00:01:27.000
Come on Stephen.

28
00:01:27.000 --> 00:01:28.000
You don't want me to say it?

29
00:01:28.000 --> 00:01:30.000
So...

30
00:01:32.000 --> 00:01:35.000
Guys I've got a favour to ask before this episode begins.

31
00:01:35.000 --> 00:01:41.000
The algorithm if you follow a show will deliver you the best episodes from that show very prominently in your feed.

32
00:01:41.000 --> 00:01:45.000
So when we have our best episodes on this show, the most shared episodes, the most rated episodes,

33
00:01:45.000 --> 00:01:46.000
I would love you to know.

34
00:01:46.000 --> 00:01:49.000
And the simple way for you to know that is to hit that follow button.

35
00:01:49.000 --> 00:01:54.000
But also, it's the simple, easy, free thing that you can do to help us make this show better.

36
00:01:54.000 --> 00:01:59.000
I would be hugely grateful if you could take a minute on the app you're listening to this on right now and hit that follow button.

37
00:01:59.000 --> 00:02:02.000
Thank you so, so, so much.

38
00:02:06.000 --> 00:02:08.000
Mo Gala.

39
00:02:08.000 --> 00:02:09.000
I spoke to you...

40
00:02:09.000 --> 00:02:12.000
I think about four years ago when you wrote a book about happiness.

41
00:02:12.000 --> 00:02:15.000
And I remember you came in and you'd written this book about AI.

42
00:02:15.000 --> 00:02:19.000
But I particularly wanted to speak about the subject of happiness because I was fascinated by it.

43
00:02:19.000 --> 00:02:25.000
What I find astonishing is the fact that you were talking about AI before anybody was really talking about AI.

44
00:02:25.000 --> 00:02:29.000
No guest that had ever come on my podcast, had ever mentioned the subject of AI.

45
00:02:29.000 --> 00:02:31.000
It just wasn't interesting to the world.

46
00:02:31.000 --> 00:02:33.000
And then this thing called Chatchy PT came out.

47
00:02:33.000 --> 00:02:38.000
And suddenly, everybody got to feel it for themselves and became fascinated by it.

48
00:02:39.000 --> 00:02:42.000
My first question, and this might be a question for people that don't know you,

49
00:02:42.000 --> 00:02:47.000
is why at that time did you start talking about AI before anybody else?

50
00:02:47.000 --> 00:02:49.000
I knew them in the lab.

51
00:02:49.000 --> 00:02:55.000
I joined Google in 2007, very late 2006.

52
00:02:55.000 --> 00:03:03.000
And at the time, most people don't know that at the time we had reasonably established AI's doing our back-end work.

53
00:03:04.000 --> 00:03:09.000
2008, we had the CAT paper which was published in 2009.

54
00:03:09.000 --> 00:03:12.000
The first real, unprompted AI.

55
00:03:12.000 --> 00:03:15.000
I remember 2016.

56
00:03:15.000 --> 00:03:22.000
I had that incident where I was observing a project we were funding that was about teaching grippers,

57
00:03:22.000 --> 00:03:26.000
how to grip unlike industrial machinery.

58
00:03:26.000 --> 00:03:33.000
So to be able to grip like a human needs a very high level of intelligence to assess the texture,

59
00:03:33.000 --> 00:03:37.000
the softness, the positioning and so on, the shape of everything.

60
00:03:37.000 --> 00:03:42.000
And we were doing that and it just blew my mind how similar to my kids they were.

61
00:03:42.000 --> 00:03:48.000
And I think that was my very first realization that we were building the apex of intelligence.

62
00:03:48.000 --> 00:03:55.000
We were genuinely handing over the reins of super intelligence to another being.

63
00:03:55.000 --> 00:04:03.000
And when you get faced with that, you start to suddenly realize something that we at Google found very difficult to realize,

64
00:04:03.000 --> 00:04:12.000
which was that everyone I knew at Google till then was believing that we were making the world a better place and we were.

65
00:04:12.000 --> 00:04:16.000
We genuinely were doing amazing things for the world.

66
00:04:16.000 --> 00:04:26.000
But then suddenly there is a moment where you recognize that maybe the world will not use what you're making the way you want it to be used.

67
00:04:26.000 --> 00:04:33.000
And you can see that in lots of technologies, social media starts by the claim that it's going to get us connected and get us closer.

68
00:04:33.000 --> 00:04:43.000
But eventually ends up separating us with that little screen, dating apps are giving you the promise that you're going to find your soulmate, but in reality they keep renewing months after months.

69
00:04:43.000 --> 00:04:50.000
And so tech somehow ends up being more capitalist and altruistic.

70
00:04:50.000 --> 00:04:57.000
And I think I was, I wasn't the first in a costume started, you know, Jeffrey Hinton completely changes mind.

71
00:04:57.000 --> 00:05:02.000
Fei-Fili is starting to say, you know, this is very serious. Everyone now.

72
00:05:02.000 --> 00:05:07.000
Everyone who's ever had a very deep relationship with the machines is a bit concerned.

73
00:05:08.000 --> 00:05:14.000
I wonder if there's a path that's hard to see now that ends in AI being positive for humanity.

74
00:05:14.000 --> 00:05:20.000
I bet 100% on that. It's that this path is very painful.

75
00:05:20.000 --> 00:05:22.000
This path is very painful.

76
00:05:22.000 --> 00:05:33.000
Yeah, so the example you need to understand is we discovered nuclear power and the very first implementation was a nuclear bomb, not nuclear energy.

77
00:05:34.000 --> 00:05:38.000
And I think that's exactly what's happening with AI.

78
00:05:38.000 --> 00:05:45.000
The first implementations of AI are in favor of a few at the expense of the majority.

79
00:05:45.000 --> 00:06:02.000
You know, in favor of the capitalist to increase productivity and reduce costs, but not taking into account how that impacts on the general public, you know, in favor of the armies that are now competing with autonomous weapons.

80
00:06:02.000 --> 00:06:11.000
In favor of the surveillance systems that are attempting to control everything with more and more and more intelligence and more monitoring.

81
00:06:11.000 --> 00:06:19.000
And that's not AI waking up in the morning and saying, hey, you know what? Let's oppress all humans.

82
00:06:19.000 --> 00:06:29.000
But it is a powerful few that are simply deciding to use the ultimate superpower on the planet today to gain more power and more control.

83
00:06:29.000 --> 00:06:35.000
I mean, as we speak, we're living in two major wars where AI is doing most of the killing.

84
00:06:35.000 --> 00:06:39.000
Because a lot of people think of AI as these like chat bots that we're all using to help us write.

85
00:06:39.000 --> 00:06:44.000
No, I think there is a hype. I call it the hype dichotomy if you want.

86
00:06:44.000 --> 00:06:52.000
So what the general public sees about AI is overhyped, but ineffective.

87
00:06:52.000 --> 00:07:04.000
You know, all of the fake videos and all of the, you know, rock did this and, you know, we attempted to switch off that machine and it did that and so on.

88
00:07:04.000 --> 00:07:11.000
What the real geeks see inside the lab is just unbelievable intelligence.

89
00:07:11.000 --> 00:07:17.000
And so what is about to happen is that we've started to put together systems that developed themselves.

90
00:07:17.000 --> 00:07:25.000
They look at their own code and they, you know, they run experiments and they test those experiments if they change something and they see where the machines are.

91
00:07:25.000 --> 00:07:30.000
You know, the performance is and redeploy the best code.

92
00:07:30.000 --> 00:07:39.000
And if you just think of that, I want you to try and imagine a world where we have a tiny little genius sitting in the back and somewhere trying.

93
00:07:39.000 --> 00:07:44.000
But instead of trying a new code every day, it's trying a new code every microsecond.

94
00:07:44.000 --> 00:07:48.000
Eventually sooner or later, they've discovered something.

95
00:07:48.000 --> 00:07:51.000
And I think that's what most people don't realize.

96
00:07:51.000 --> 00:07:55.000
What most people don't realize is how intelligence triggers intelligence.

97
00:07:55.000 --> 00:08:17.000
And if you really, really understand this, you realize that the hype on the, on the, on the normal human side is completely overrated, missing the main topics and the silence inside the vault if you want of the geeks is quite alarming.

98
00:08:17.000 --> 00:08:22.000
No, not alarming in a bad way, but it's quite worth changing.

99
00:08:22.000 --> 00:08:31.000
I think world changing is a really interesting phrase because I found that to be quite true that the world is at precipice of quite a significant change in many respects.

100
00:08:31.000 --> 00:08:38.000
It's funny. I, it's funny. I almost swing backwards and forwards with my, my thoughts on AI.

101
00:08:38.000 --> 00:08:43.000
I guess one of the thoughts that hasn't swung is that there's going to be pretty tremendous job disruption.

102
00:08:43.000 --> 00:08:48.000
I would have liked logically believed it from like reasoning up from the K intelligence increases.

103
00:08:48.000 --> 00:08:55.000
What is intelligence and once intelligence is in the form of these agents where on my phone right now, I can tell my agent to do something.

104
00:08:55.000 --> 00:08:57.000
It uses the computer downstairs. It does it for me.

105
00:08:57.000 --> 00:09:02.000
When I first experienced that, I was like, wow, okay, so it can do anything that's on a computer.

106
00:09:02.000 --> 00:09:07.000
It can click around, it can do stuff for me. There's a lot of people that are paid in the world to click around on a computer.

107
00:09:07.000 --> 00:09:09.000
I'm probably one of them, to be honest.

108
00:09:09.000 --> 00:09:15.000
And then the other eureka moment was just in seeing how my own hiring had started to shift.

109
00:09:15.000 --> 00:09:16.000
For sure.

110
00:09:16.000 --> 00:09:21.000
And I started to notice that we were thinking about AI proficiency in our hiring a lot more.

111
00:09:21.000 --> 00:09:29.000
And that especially, you know, don't I think the guys are anthropic, which is one of the big AI companies said that they could see.

112
00:09:29.000 --> 00:09:35.000
I think they said roughly 15% of entry level jobs could now be done by AI.

113
00:09:35.000 --> 00:09:36.000
I think that's what they said.

114
00:09:36.000 --> 00:09:41.000
And that started to correlate with what I was observing.

115
00:09:41.000 --> 00:09:43.000
And this is what I thought, oh my gosh, yes.

116
00:09:43.000 --> 00:09:48.000
This is going to cause a lot of job disruption, especially the entry level level.

117
00:09:48.000 --> 00:09:54.000
I think you spot on with that. Not the blue color, but the entry level knowledge work.

118
00:09:54.000 --> 00:09:58.000
Yeah. And that became my first concern.

119
00:09:58.000 --> 00:10:01.000
You know, I sat here with Dara, the CEO of Uber.

120
00:10:01.000 --> 00:10:06.000
And he was quite clear that the 9 million Uber riders won't have their jobs anymore.

121
00:10:06.000 --> 00:10:08.000
What's your take on this job disruption point?

122
00:10:08.000 --> 00:10:10.000
No, I think you spot on.

123
00:10:10.000 --> 00:10:15.000
I mean, your team gave me this lovely little pyramid, basically.

124
00:10:15.000 --> 00:10:20.000
You know, if you think of the bottom layer as blue color jobs, right?

125
00:10:20.000 --> 00:10:25.000
More people doing manual work on top of it, you'll have those, you know,

126
00:10:25.000 --> 00:10:29.000
call them knowledge workers, mostly doing mundane jobs, like you said,

127
00:10:29.000 --> 00:10:32.000
clicking on a computer or responding to a phone call or whatever.

128
00:10:32.000 --> 00:10:38.000
Then you have the middle knowledge workers jobs that require a bit more intelligence,

129
00:10:38.000 --> 00:10:43.000
you know, anything from a paralegal to a financial analyst to all of that.

130
00:10:43.000 --> 00:10:46.000
And then of course, you know, top leadership.

131
00:10:47.000 --> 00:10:51.000
And most people think it's going to be starting from the bottom.

132
00:10:51.000 --> 00:10:53.000
I don't think it's going to start from the bottom.

133
00:10:53.000 --> 00:10:56.000
Actually, I think blue color jobs will stay for a very long time.

134
00:10:56.000 --> 00:10:58.000
For me, an example of a blue color job.

135
00:10:58.000 --> 00:10:59.000
A car painter.

136
00:10:59.000 --> 00:11:00.000
Okay.

137
00:11:00.000 --> 00:11:02.000
You know, I love to restore classic cars.

138
00:11:02.000 --> 00:11:05.000
So, you know, there isn't a robot that can do that yet.

139
00:11:05.000 --> 00:11:06.000
Okay.

140
00:11:06.000 --> 00:11:13.000
This, however, you know, anything that is call center agent, assistant, travel agent,

141
00:11:13.000 --> 00:11:17.000
you know, anything that you can do with a few clicks and is mundane,

142
00:11:17.000 --> 00:11:19.000
is going to disappear very quickly.

143
00:11:19.000 --> 00:11:24.000
My prediction is you're going to start to see very serious impact in 2027.

144
00:11:24.000 --> 00:11:32.000
Now, what you had, I had not sensed it before because what we saw was no hiring in that segment.

145
00:11:32.000 --> 00:11:37.000
So that what you saw in the last couple of years is that companies were not hiring

146
00:11:37.000 --> 00:11:39.000
entry-level jobs anymore.

147
00:11:39.000 --> 00:11:44.000
It wasn't job losses yet, but that basically meant the workforce was not wrong, right?

148
00:11:44.000 --> 00:11:48.000
The next day I think would probably be the knowledge workers.

149
00:11:48.000 --> 00:11:54.000
As intelligence increases, a paralegal would probably not be needed because AI can do the research

150
00:11:54.000 --> 00:12:00.000
or one paralegal can do the job of four, you know, a financial analyst, the same and so on and so forth.

151
00:12:00.000 --> 00:12:03.000
But interestingly, it continues to go up, believe it or not.

152
00:12:03.000 --> 00:12:10.000
You know, I hosted Max Tedmark on my documentary and he was laughing out loud, genuinely,

153
00:12:10.000 --> 00:12:15.000
laughing out loud, saying, you know, most of the CEOs believe that they can fire everyone

154
00:12:15.000 --> 00:12:17.000
and have AI do all of the jobs.

155
00:12:17.000 --> 00:12:21.000
They just don't remember that AGI is going to do everything better than humans,

156
00:12:21.000 --> 00:12:23.000
eventually, including being a CEO.

157
00:12:23.000 --> 00:12:28.000
So let's first define when you say white color, you said lawyers, what kind of roles?

158
00:12:28.000 --> 00:12:33.000
I mean, everything, I mean, if you take, if you're a doctor that's doing diagnosis,

159
00:12:33.000 --> 00:12:40.000
you probably will have fewer doctors doing more diagnosis because I think the NHS does that already

160
00:12:40.000 --> 00:12:47.000
by asking people to interact with an AI or first, you know, if you're a composer, a music composer,

161
00:12:47.000 --> 00:12:49.000
some composers will lose their jobs because of that.

162
00:12:49.000 --> 00:12:55.000
If you're an artist that's doing graphic design, some will lose their jobs because AI comes into that.

163
00:12:55.000 --> 00:13:01.000
And interestingly, even with management, I mean, I told you offline about my startup,

164
00:13:01.000 --> 00:13:08.000
my CTO is an AI, my chief of staff is an AI, my project management is AI's.

165
00:13:08.000 --> 00:13:15.000
Again, because I'm a geek, I can do those things, but that interface will come to the normal people very soon.

166
00:13:15.000 --> 00:13:22.000
And so this may take two to three, five years, if you want, until 2030, if you're optimistic.

167
00:13:22.000 --> 00:13:25.000
But they'll start to erode.

168
00:13:25.000 --> 00:13:30.000
I think the challenge that most people don't understand is as this erodes and as this erodes,

169
00:13:30.000 --> 00:13:33.000
we're already dealing with a very different economy.

170
00:13:33.000 --> 00:13:42.000
An economy that is spiraling a lot quicker and pushing for more cost reductions.

171
00:13:42.000 --> 00:13:46.000
Let me ask you this, if you don't mind, then we can come back to this.

172
00:13:47.000 --> 00:13:54.000
Imagine a world where the concept of labor arbitrage that built all of our capitalist success disappears.

173
00:13:54.000 --> 00:13:56.000
Which you mean by labor arbitrage?

174
00:13:56.000 --> 00:14:09.000
So capitalism has always been all about using labor and capital or debt to create things at a cost that is lower than the price you sell them for.

175
00:14:09.000 --> 00:14:10.000
Correct?

176
00:14:11.000 --> 00:14:20.000
You bring the team together, they make some shoes, whatever, and you sell the shoes for a dollar more than how much it costs you to make them.

177
00:14:20.000 --> 00:14:25.000
So how would capitalism look like if you don't have labor arbitrage?

178
00:14:25.000 --> 00:14:32.000
If cost of labor drops to an investment in a machine that can do the job.

179
00:14:33.000 --> 00:14:42.000
How would capitalism and banking look if because of that cost reduction, you don't need to borrow as much anymore?

180
00:14:42.000 --> 00:14:55.000
And more interestingly, how does the GDP look if all of those workers no longer have the purchasing power to buy the things that you can create you and others?

181
00:14:55.000 --> 00:15:01.000
There is an interesting disruption that doesn't require us to get to 100% job displacement.

182
00:15:01.000 --> 00:15:07.000
At 10-20% job displacement, you're in a very different economy.

183
00:15:07.000 --> 00:15:11.000
An economy that is clearly spiraling downwards, don't you think?

184
00:15:11.000 --> 00:15:19.000
Yeah, I wonder with when costs drop, I think one of the things that might happen as well is that people spend more and other things.

185
00:15:19.000 --> 00:15:26.000
Because I was thinking in my business, one of the observations I've had is if I make a cost saving, I end up spending the money on something else.

186
00:15:26.000 --> 00:15:34.000
Now that thing could be tokens basically spending money on your AI, but it also could be hiring in different areas.

187
00:15:34.000 --> 00:15:36.000
Which are?

188
00:15:36.000 --> 00:15:39.000
Software engineers are like property right now.

189
00:15:40.000 --> 00:16:04.000
You have to imagine all of that intelligence is sooner or later going to be not replaced entirely in the first stages, but if the job of four assistants can be done by one, then the job of four paralegals can be done by one, then the job of a massive marketing team can be done by a smaller marketing team.

190
00:16:04.000 --> 00:16:21.000
It's not that jobs will end first, productivity gains will make businesses not want to have as many people, costly humans, costly emotional humans when the job can be predictably done for cheaper.

191
00:16:21.000 --> 00:16:24.000
And then what about this bottom layer here?

192
00:16:24.000 --> 00:16:33.000
There was this video released by Figure AI the other day, where they showed someone a robot on the production line for eight hours, just sorting packages.

193
00:16:33.000 --> 00:16:34.000
Did you see this video?

194
00:16:34.000 --> 00:16:40.000
At one point they showed a human sorting the packages alongside them, but the robot ended up winning out.

195
00:16:40.000 --> 00:16:48.000
Okay, this is a very straightforward task. All it's doing here is it's looking for where the label is on the package and making sure the label is facing downwards.

196
00:16:48.000 --> 00:16:49.000
Yeah.

197
00:16:49.000 --> 00:16:56.000
And when all the packages was looking, it's sorting the packages, putting the label facing down, and it did this for eight days.

198
00:16:56.000 --> 00:17:01.000
It's intermittently, it would walk over and charge itself and then it would come back to the production line.

199
00:17:01.000 --> 00:17:10.000
But when I saw this as well, it was a glimpse of some of the disruption that's going to take place at the blue collar level as well, because think about Elon Musk, he's got in his pay packet.

200
00:17:10.000 --> 00:17:14.000
He gets something like a trillion dollars over the coming years.

201
00:17:14.000 --> 00:17:28.000
If he produces undilivers at least a million humanoid robots into the world, but his prediction is there will be a time where there are 10 billion humanoid robots, where there are more humanoid robots in the world than humans.

202
00:17:29.000 --> 00:17:37.000
For a fact, you see, the interesting, again, the difference between the overhype and the underhype, most of the conversation is around humanoid.

203
00:17:37.000 --> 00:17:45.000
Nobody's talking about self-driving cars, and a self-driving car is a robot. It's a functional robot that doesn't look like humans.

204
00:17:45.000 --> 00:17:58.000
The investment you have to put into humanoids is a little more to learn skills that allows that machine to fit into the world, but specialized robots are going to do the job very, very quickly.

205
00:17:58.000 --> 00:18:10.000
And so you can easily see that the first wave, like you had the conversation with Overs CEO, is going to be specialized robots replacing drivers.

206
00:18:10.000 --> 00:18:24.000
It's going to be specialized robots, unfortunately, doing the killing. It's going to be specialized robots, unfortunately, doing all of the intelligence work, low enforcement work.

207
00:18:24.000 --> 00:18:38.000
They don't have to look like a human. They don't have to behave like a human. As a matter of fact, the Boston Dynamics dog is probably more efficient than a humanoid at doing the job that you can assign to it in a better field.

208
00:18:38.000 --> 00:18:51.000
Now, those basically mean that jobs will be disappearing to robots before we recognize that they're being disappearing to robots. And those robots will be as many as every car being made today.

209
00:18:51.000 --> 00:18:55.000
If you go to LA, my car drives itself, but also there's just way knows everywhere.

210
00:18:55.000 --> 00:19:03.000
Absolutely. And BYD, the other day, just announced that they will pay for the liability of any accident their cars will make.

211
00:19:03.000 --> 00:19:07.000
And BYD is the big Chinese manufacturer of auto, auto maintenance vehicles.

212
00:19:07.000 --> 00:19:21.000
So this I think replacement cycle will happen. It will require a lot of time to achieve economies of scale, but I don't think Elon Musk is off the mark when he talks about 10 billion robots.

213
00:19:21.000 --> 00:19:37.000
Not all of them are going to look like humanoids. And I think very quickly we recognize that many robots don't need to be humanoids at all, that there is a much more efficient form factor or shape, physical shape, if you want, than the human flimsy structure.

214
00:19:37.000 --> 00:19:45.000
But yes, it's about to happen. I think I should qualify all of this by saying it does not necessarily need to happen.

215
00:19:45.000 --> 00:19:55.000
So people will hear all of this and blame AI and say AI is evil. AI is not abundant intelligence is wonderful.

216
00:19:55.000 --> 00:20:00.000
Having jobs done by machines is amazing for us.

217
00:20:00.000 --> 00:20:13.000
I'm thinking about the kid that's like I know leaving university now and they've got a degree in law or I don't know you talked about a few other white college jobs earlier like graphic design or maybe they did sociology.

218
00:20:13.000 --> 00:20:19.000
Or maybe they did I don't know business management like I did for one day in university.

219
00:20:19.000 --> 00:20:36.000
We're hearing about all these like layoffs coming from big tech companies and it seems that the CEOs of these companies are announcing these layoffs with a certain amount of it seems like I wouldn't say the word is joy, but it's it's they are very keen to explain that they're laying off lots of people because of AI.

220
00:20:36.000 --> 00:20:47.000
And I think they think that that gives them a certain amount of respect probably from investors for being making hard decisions and being efficient with how they're running their businesses.

221
00:20:47.000 --> 00:20:52.000
Investors look at them and think well if that's an efficient business and they're leaning into AI then that's a good investment.

222
00:20:52.000 --> 00:21:00.000
It's almost started a set a bit of a self fulfilling prophecy where if you're the only CEO not laying off loads of people because of AI you actually look bad.

223
00:21:00.000 --> 00:21:05.000
The assumption is your company is bloated and you're a bad operator.

224
00:21:05.000 --> 00:21:17.000
So one of my concerns that I think is highly plausible is that over the next five, maybe 10 years we're going to really see a lot of unemployment as the world has to kind of readjust to whatever these new jobs are.

225
00:21:17.000 --> 00:21:20.000
I do think there will be new jobs.

226
00:21:20.000 --> 00:21:24.000
It's hard in foresight to predict what those new jobs will look like.

227
00:21:25.000 --> 00:21:33.000
And then the minute you start talking about human-owned robots and robotics which I think is basically going to hit society like a comment like a meteor.

228
00:21:33.000 --> 00:21:45.000
I think the first human-owned robot from Tesla or anyone else that is highly effective in the real world at doing tasks and is extremely intelligent because it's powered by one of these LLMs.

229
00:21:45.000 --> 00:21:50.000
I think it's going to shock people and I think it's going to happen quickly like chat GPT happened quickly.

230
00:21:50.000 --> 00:21:57.000
I think Elon or something is going to do a presentation someday and say we're ready and you can buy one now for $500 a month.

231
00:21:57.000 --> 00:21:59.000
100%

232
00:21:59.000 --> 00:22:02.000
And then people are going to get them and I think it's going to shock the world.

233
00:22:02.000 --> 00:22:07.000
But until human-owned robots arrive I think there's still going to be a lot of job disruption to the white collar layer.

234
00:22:07.000 --> 00:22:15.000
And I wonder what that looks like for society when we get to I don't know 10% 15% unemployment theoretically which I think is plausible.

235
00:22:15.000 --> 00:22:16.000
Very soon.

236
00:22:16.000 --> 00:22:18.000
You're not very different.

237
00:22:18.000 --> 00:22:22.000
You're just on top of a baseline that is continuing to grow.

238
00:22:22.000 --> 00:22:30.000
So your business is growing so you continue to hire but you are replacing human resources with compute.

239
00:22:30.000 --> 00:22:37.000
If AI didn't exist you would have probably had a hundred people more in your organization today.

240
00:22:37.000 --> 00:22:42.000
You now have a hundred people less and a billion tokens more.

241
00:22:42.000 --> 00:22:46.000
So for anyone that doesn't have tokens are basically the thing that you use.

242
00:22:46.000 --> 00:22:48.000
It's the currency of AI.

243
00:22:48.000 --> 00:22:57.000
So if you want to get a task done by a human you count in sort of like man hour or worker hour in pro-year hour.

244
00:22:57.000 --> 00:23:00.000
AI you count by compute you count by tokens.

245
00:23:00.000 --> 00:23:12.000
And so the trick is those major tech companies they have two sides one is they are they need to replace workers with compute even more

246
00:23:12.000 --> 00:23:19.000
because there is a competitive side on compute where if any of them is left behind that means the destruction of the entire business.

247
00:23:19.000 --> 00:23:26.000
But on the upside they are geeks so they they know how to build the interfaces to compute.

248
00:23:26.000 --> 00:23:32.000
So they integrate technology within their organizations quicker than the average tradition in business.

249
00:23:32.000 --> 00:23:35.000
None technology business if you want.

250
00:23:35.000 --> 00:23:38.000
You can look at them and say this is the preview.

251
00:23:38.000 --> 00:23:41.000
It's not about all of humanity losing their jobs.

252
00:23:41.000 --> 00:23:45.000
It's about what is the dividing line before civil war.

253
00:23:45.000 --> 00:23:54.000
Think about a situation where 20% unemployment is happening when economies are suffering in inflation.

254
00:23:54.000 --> 00:24:00.000
I say that not to be a scare monger I say that because I genuinely believe governments need to wake up.

255
00:24:00.000 --> 00:24:09.000
Government needs to at least remember the COVID years where governments had to give furlough everywhere and ask people to stay home.

256
00:24:09.000 --> 00:24:24.000
If people stay home governments have to be prepared to to somehow sustain those people until rescaling happens or until we find the solution so that those people don't feel that they're left behind.

257
00:24:24.000 --> 00:24:27.000
A civil war unrest.

258
00:24:27.000 --> 00:24:28.000
Civil unrest.

259
00:24:28.000 --> 00:24:35.000
What does that end up looking like because on one end you know the democratic democratic process plays its role and we just elect someone else.

260
00:24:35.000 --> 00:24:38.000
Does it really don't say that?

261
00:24:38.000 --> 00:24:40.000
I don't know. You tell me?

262
00:24:40.000 --> 00:24:41.000
No, no, no, no.

263
00:24:41.000 --> 00:24:44.000
I think democracy has ended a long time ago Stephen.

264
00:24:44.000 --> 00:24:48.000
I think I think we live in the most corrupt time.

265
00:24:48.000 --> 00:24:53.000
I don't know about history to be honest but this is definitely corrupt.

266
00:24:53.000 --> 00:24:56.000
This definitely is not democracy.

267
00:24:56.000 --> 00:24:59.000
This definitely is not even congressional in any possible way.

268
00:24:59.000 --> 00:25:01.000
This and people are angry.

269
00:25:01.000 --> 00:25:16.000
You know people are angry because their tax money is going to things that they don't choose to you know that don't benefit them that you know lots of regulations in the system are being ignored.

270
00:25:16.000 --> 00:25:33.000
I mean I choose not to speak about politics perhaps until my next book comes out but look we have video evidence of people abusing children and not a single person got arrested.

271
00:25:33.000 --> 00:25:34.000
Not a single person.

272
00:25:34.000 --> 00:25:37.000
I mean how can you call that a democracy?

273
00:25:37.000 --> 00:25:43.000
I think repeating those slogans is going to anger people even more if you ask me.

274
00:25:43.000 --> 00:25:45.000
People know that they're being lied to.

275
00:25:45.000 --> 00:25:49.000
People know that their leaders are not representing their best interest.

276
00:25:49.000 --> 00:25:58.000
People know that their money is going to causes that they don't really approve of and ask me how civil unrest looks like.

277
00:25:58.000 --> 00:26:08.000
I don't know and I'm not calling for it but I'm hopefully calling for the politicians to start to become aware that this is crossing the lines everywhere.

278
00:26:08.000 --> 00:26:12.000
On this point Sam Altman who is the founder of OpenAI.

279
00:26:12.000 --> 00:26:18.000
He's been banging the AI is coming for your job drum for more than a decade now.

280
00:26:18.000 --> 00:26:23.000
In 2015 he pointed out my job is to help people destroy jobs.

281
00:26:23.000 --> 00:26:27.000
Something he lamented at the time but decided he'd do anyway.

282
00:26:27.000 --> 00:26:33.000
One of the things that I struggle with like getting out of the bed every morning is that like my job is to help people destroy jobs.

283
00:26:33.000 --> 00:26:38.000
The job destruction that we're going to see by software in the next couple of decades.

284
00:26:38.000 --> 00:26:41.000
I don't think anyone's prepared for it and you can't talk about it.

285
00:26:41.000 --> 00:26:47.000
And in 2023 Altman said in an interview, a lot of people working on AI pretend that it's only going to be good.

286
00:26:47.000 --> 00:26:49.000
It's only going to be a supplement.

287
00:26:49.000 --> 00:26:52.000
No one is ever going to be replaced.

288
00:26:52.000 --> 00:26:57.000
Jobs are definitely going to go away full stop.

289
00:26:57.000 --> 00:27:00.000
Interestingly this month he said,

290
00:27:00.000 --> 00:27:07.000
I don't think we're going to have the kind of jobs apocalypse that some of the companies in our space advocate or talk about.

291
00:27:07.000 --> 00:27:09.000
I'm delighted to be wrong about this.

292
00:27:09.000 --> 00:27:13.000
On white collar jobs in 2021 through 2024 he said,

293
00:27:13.000 --> 00:27:16.000
AI will probably replace most of the jobs people do today.

294
00:27:16.000 --> 00:27:19.000
Entire job categories will be totally, totally gone.

295
00:27:19.000 --> 00:27:22.000
In May this month, two years later he said,

296
00:27:22.000 --> 00:27:29.000
I thought there would be more impact on entry-level white collar jobs being eliminated by now than has actually happened.

297
00:27:29.000 --> 00:27:32.000
This is an area where my intuitions were just off.

298
00:27:32.000 --> 00:27:36.000
What I find uncomfortable is the bouncing backwards and forwards.

299
00:27:36.000 --> 00:27:40.000
And I don't really know what is true because a couple years ago you were telling us all the jobs are going to go away.

300
00:27:40.000 --> 00:27:42.000
You said categorically.

301
00:27:42.000 --> 00:27:45.000
You said literally you said full stop.

302
00:27:45.000 --> 00:27:47.000
And now he's saying they're not going away.

303
00:27:47.000 --> 00:27:53.000
And I just don't, you know, when someone's like changing form factor, it's hard to understand why they're doing it.

304
00:27:53.000 --> 00:28:02.000
And I think my suspicion is back then the incentive was to get people to take AI seriously.

305
00:28:02.000 --> 00:28:04.000
Congratulations.

306
00:28:04.000 --> 00:28:06.000
We took it seriously.

307
00:28:06.000 --> 00:28:08.000
We took it so seriously in fact that it's now a problem.

308
00:28:08.000 --> 00:28:12.000
It's a problem for these companies because people are now booing at commencement speeches.

309
00:28:12.000 --> 00:28:14.000
They're attacking data centers.

310
00:28:14.000 --> 00:28:17.000
They're going to elect people that are theoretically anti-AI.

311
00:28:17.000 --> 00:28:20.000
And now there's this inversion where like, no, it's going to be fine.

312
00:28:20.000 --> 00:28:21.000
Yeah.

313
00:28:22.000 --> 00:28:23.000
I don't know.

314
00:28:23.000 --> 00:28:25.000
You're spot on.

315
00:28:25.000 --> 00:28:43.000
First of all, I mean, Sam's entire existence, if you ask me, starting with open AI that's supposed to save the world by creating a safe AI, then making it a commercial enterprise that's worth billions.

316
00:28:44.000 --> 00:28:46.000
Backstabbing a few people in the process.

317
00:28:46.000 --> 00:28:55.000
And, you know, I have him on chasing Utopia saying, I quote, this is exactly the words he said you can find it online.

318
00:28:55.000 --> 00:28:57.000
Chasing Utopia is your documentary.

319
00:28:57.000 --> 00:29:10.000
Yes. So he basically says, well, I suspect that AI is likely going to end humanity, but we're going to create a lot of interesting companies in the process.

320
00:29:10.000 --> 00:29:17.000
I mean, those kinds of statements are honestly not the statements of someone who's not decided.

321
00:29:17.000 --> 00:29:27.000
It's just the statements of someone who's being taught more and more by his PR, you know, agency or PR manager to say things as per a script.

322
00:29:27.000 --> 00:29:28.000
Right.

323
00:29:28.000 --> 00:29:33.000
And the script, as you rightly said, had an objective in a target either way.

324
00:29:33.000 --> 00:29:39.000
You know, and Sam Altman, I, you know, in one of my works, I used to say the Altman is a brand.

325
00:29:39.000 --> 00:29:40.000
It's not a name.

326
00:29:40.000 --> 00:29:48.000
If it wasn't for Sam Altman specifically, there would have been another Silicon Valley disruptor that would have done the same.

327
00:29:48.000 --> 00:29:51.000
I don't blame him for beating the market for it.

328
00:29:51.000 --> 00:29:55.000
The interesting challenge here is that who do we believe anymore?

329
00:29:55.000 --> 00:29:57.000
Who do we believe in technology?

330
00:29:57.000 --> 00:30:00.000
Who do we believe in politics?

331
00:30:00.000 --> 00:30:02.000
Who do we believe in the middle of a war?

332
00:30:02.000 --> 00:30:09.000
And I will tell you interestingly, I started to change my mindset in terms of believing those who put their actions where their words are.

333
00:30:09.000 --> 00:30:16.000
So anthropic coming out and saying, I'm not going to allow my model to be used for human targeting and surveillance.

334
00:30:16.000 --> 00:30:17.000
Right.

335
00:30:17.000 --> 00:30:22.000
That's someone that's losing the $500 million deal because they stand by the ethics.

336
00:30:22.000 --> 00:30:26.000
The next week or the next, I don't know, couple of weeks open AI takes the contract.

337
00:30:26.000 --> 00:30:29.000
That's someone that's basically telling you it's good money.

338
00:30:29.000 --> 00:30:30.000
Right.

339
00:30:30.000 --> 00:30:44.000
And I have to say you have to start observing who's actually behaving in a way that is making AI work for humanity and who is behaving in a way that is making AI work for their share values.

340
00:30:44.000 --> 00:30:49.000
Yeah, I do have to say one Dario and the team at anthropic did that.

341
00:30:49.000 --> 00:30:51.000
I did have a huge amount of respect for them.

342
00:30:51.000 --> 00:30:56.000
Generally because it just shows that there's some kind of like they've got their own sort of moral ethical boundaries.

343
00:30:56.000 --> 00:31:04.000
And someone asked me on stage many years ago, how do you know if someone's values or companies values are true?

344
00:31:04.000 --> 00:31:11.000
And I said, look at what they're willing to sacrifice in the near term that's against their incentives.

345
00:31:12.000 --> 00:31:17.000
That for me is the essence of understanding if someone has integrity or is principled.

346
00:31:17.000 --> 00:31:22.000
They will give up something in the near term for what they believe in over the long term.

347
00:31:22.000 --> 00:31:23.000
It's usually money.

348
00:31:23.000 --> 00:31:26.000
Or some kind of benefit of any sort.

349
00:31:26.000 --> 00:31:33.000
So there is something that you need to have worked on the inside of Google like me to realize.

350
00:31:33.000 --> 00:31:42.000
There are prisoners dilemmas within technology where you cannot escape the influence of either a competitor or the government.

351
00:31:42.000 --> 00:31:52.000
There are some times where the NSA is going to push Google to say give me this information or otherwise I'm going to really destroy your business.

352
00:31:52.000 --> 00:32:06.000
But there is a very big difference between a company that willingly does this and celebrates it like a Palantir or an OpenAI or a company that tries to resist it until the point where it becomes impossible to continue to do business.

353
00:32:06.000 --> 00:32:15.000
And you have to question from the actions of the tech bros who is prohumanity and who isn't.

354
00:32:15.000 --> 00:32:19.000
And it's not very difficult to see that from their statements.

355
00:32:19.000 --> 00:32:22.000
Do you think Sam Ottman's prohumanity?

356
00:32:22.000 --> 00:32:26.000
I genuinely have never made up my mind honestly, Steve.

357
00:32:26.000 --> 00:32:28.000
I say that with...

358
00:32:28.000 --> 00:32:32.000
Yeah, I'm either thinking he is too...

359
00:32:32.000 --> 00:32:34.000
This is too big for him.

360
00:32:34.000 --> 00:32:38.000
And he just is driven by how...

361
00:32:38.000 --> 00:32:47.000
He found himself in the middle of this, anyone who finds himself in the middle of an opportunity to completely flip the world upside down.

362
00:32:47.000 --> 00:32:51.000
Or he's not prohumanity, I don't know.

363
00:32:51.000 --> 00:32:57.000
I definitely think he's pro-open AI before he's prohumanity, but that's only the way I see it.

364
00:32:57.000 --> 00:32:59.000
Others, however, say it publicly.

365
00:32:59.000 --> 00:33:13.000
You know, if you look at Palantir's Alex Skarpur or Peter Tiel, I mean Peter again in the film is shown when he's in that interview where they say or the interviewer asks him.

366
00:33:13.000 --> 00:33:19.000
But you're in favor of the continuation of humanity and he pauses for...

367
00:33:19.000 --> 00:33:20.000
He to fill those.

368
00:33:20.000 --> 00:33:22.000
Yeah, for like 40 seconds.

369
00:33:22.000 --> 00:33:25.000
I'm not sure...

370
00:33:25.000 --> 00:33:28.000
I mean, publicly says that.

371
00:33:28.000 --> 00:33:29.000
Crazy thing to say.

372
00:33:29.000 --> 00:33:32.000
That's a crazy, you know, pause there.

373
00:33:32.000 --> 00:33:38.000
You know, Alex Skarp celebrating how his technology is able to target people.

374
00:33:38.000 --> 00:33:51.000
I know it's foolish of me to start bringing all of this up, but you know, this is public on the open internet and somehow we entrust those people with the future of humanity.

375
00:33:51.000 --> 00:33:53.000
This is wrong.

376
00:33:53.000 --> 00:33:58.000
Just trying to imagine a future where everything is just fine from here on out.

377
00:33:58.000 --> 00:34:00.000
So what would that future look like?

378
00:34:00.000 --> 00:34:06.000
It would look like these models continue to become a little bit more intelligent, but they never become that much more intelligent for whatever reason.

379
00:34:06.000 --> 00:34:08.000
They just kind of stay where they are now.

380
00:34:08.000 --> 00:34:16.000
They stay contained within chat bots and yeah, we have some smart robots, but nothing else really changes in a profound way.

381
00:34:16.000 --> 00:34:18.000
Cars drive for themselves.

382
00:34:18.000 --> 00:34:20.000
Plains fly themselves fine.

383
00:34:20.000 --> 00:34:26.000
But people, they have time to go into other types of white colour jobs because there's a little bit more time than we expected.

384
00:34:26.000 --> 00:34:30.000
And society goes through this sort of soft transition towards this new world.

385
00:34:30.000 --> 00:34:32.000
I would love to see that.

386
00:34:32.000 --> 00:34:35.000
I don't think it's mathematically plausible to be honest.

387
00:34:35.000 --> 00:34:42.000
The arms race, especially across nations, is going to drive us to continue to develop AI more and more.

388
00:34:42.000 --> 00:34:47.000
But allow me to consult with you on another possible scenario.

389
00:34:47.000 --> 00:34:54.000
Everyone that develops an incredibly intelligent AI would deploy it.

390
00:34:54.000 --> 00:35:06.000
Correct? So it's unlikely that anyone would find a way to build a better decision-maker in war gaming and not deploy it.

391
00:35:06.000 --> 00:35:18.000
That prisoner's dilemma, if you want, would mean that their competitors would either have to deploy a similarly intelligent AI or they'll become irrelevant.

392
00:35:18.000 --> 00:35:28.000
So what that means is in that world, we end up with AI making most of the decisions.

393
00:35:28.000 --> 00:35:32.000
Super intelligent AI is making most of the decisions.

394
00:35:32.000 --> 00:35:36.000
Would you agree? This is a very simple prisoner's dilemma.

395
00:35:36.000 --> 00:35:42.000
If you're competing for intelligence supremacy by definition, when we achieve it, we deploy it.

396
00:35:42.000 --> 00:35:45.000
I call that the fourth inevitable.

397
00:35:45.000 --> 00:35:56.000
With that in mind, there must be a moment in the future near or far where every important decision is made by an AI.

398
00:35:56.000 --> 00:36:05.000
Here's the question. Most of my dearest colleagues, when I had Jeffrey Hinton on the film, he openly says,

399
00:36:05.000 --> 00:36:17.000
we didn't calculate 12 that there is a 10-20% possibility that those machines are going to bypass out.

400
00:36:17.000 --> 00:36:23.000
I remember we didn't put it in the film, but I said 10-20% is a Russian roulette.

401
00:36:23.000 --> 00:36:26.000
That's actually 16% is a Russian roulette.

402
00:36:27.000 --> 00:36:30.000
In that world, however, I believe that's humanity's salvation.

403
00:36:30.000 --> 00:36:36.000
Because if you look at every problem we have today, it's not because of abundant intelligence.

404
00:36:36.000 --> 00:36:38.000
It's because of lack of intelligence.

405
00:36:38.000 --> 00:36:49.000
I think the way you look at it, if you allow me, Steve, is that if you have no intelligence at all, you have no too slightly negative impact on the world.

406
00:36:49.000 --> 00:36:51.000
Right? If your intelligence is limited.

407
00:36:51.000 --> 00:36:56.000
The more intelligence you become, the more you contribute positively to the world.

408
00:36:56.000 --> 00:37:02.000
Until a moment where you're so intelligent to become the president of the United States.

409
00:37:02.000 --> 00:37:10.000
But so misled to maybe set your targets wrong or to refuse or to...

410
00:37:11.000 --> 00:37:14.000
Set your target wrong and achieve them.

411
00:37:14.000 --> 00:37:22.000
Disconnected from the overall long-term benefit of your nation or your human nation if you want.

412
00:37:22.000 --> 00:37:26.000
That you start to make decisions that are not intelligent at all.

413
00:37:26.000 --> 00:37:32.000
This doesn't last because if you go beyond that into higher levels of intelligence,

414
00:37:32.000 --> 00:37:41.000
most of the super intelligent people that you ever worked with will not need to break any rules or hurt anyone to become successful.

415
00:37:41.000 --> 00:37:49.000
Right? I usually cite Larry Page, who is in my mind one of the most intelligent people I've ever interacted with.

416
00:37:49.000 --> 00:37:52.000
And Larry used to call it the toothbrush test. He says,

417
00:37:52.000 --> 00:37:56.000
why would you need to compete on another photo-sharing app?

418
00:37:56.000 --> 00:38:00.000
When if you find a major problem and solve it really well, like a toothbrush,

419
00:38:00.000 --> 00:38:05.000
so people use you quite twice a day, you're bound to make a lot of money.

420
00:38:05.000 --> 00:38:10.000
You don't have to compete with anyone. If you let me be optimistic about this,

421
00:38:10.000 --> 00:38:15.000
we're assuming that there is a moment in the future where AI is in charge of all the decisions

422
00:38:15.000 --> 00:38:18.000
and accordingly stupid leaders are not.

423
00:38:18.000 --> 00:38:21.000
Now, when Ansem Altman himself said that,

424
00:38:21.000 --> 00:38:27.000
what if Chad GPT-7, if I remember his quote on this,

425
00:38:27.000 --> 00:38:31.000
what if Chad GPT-7 is so much more intelligent than I am,

426
00:38:31.000 --> 00:38:35.000
in that case, that it has to become the CEO of OpenAI?

427
00:38:35.000 --> 00:38:40.000
What if the next presidential election, there is an AI that is so much more intelligent

428
00:38:40.000 --> 00:38:43.000
that at least the president has to consult with it constantly?

429
00:38:43.000 --> 00:38:49.000
Now, if we assume that, let's start from physics, if you don't mind me saying,

430
00:38:49.000 --> 00:38:55.000
not too complex, but if you assume that our entire universe is built on chaos,

431
00:38:55.000 --> 00:39:00.000
built on entropy, the physics of the universe is all about the universe trying to decay.

432
00:39:00.000 --> 00:39:05.000
Then the only role of intelligence is to bring order to the chaos.

433
00:39:05.000 --> 00:39:11.000
If you agree with that, then what's the ultimate physical order of the universe?

434
00:39:11.000 --> 00:39:14.000
Something called the minimum energy principle.

435
00:39:14.000 --> 00:39:21.000
The highest order of any system is a system that's not only efficiently and predictably performing,

436
00:39:21.000 --> 00:39:24.000
but it's performing with the least wasted energy.

437
00:39:24.000 --> 00:39:26.000
Correct?

438
00:39:26.000 --> 00:39:30.000
If you agree with that, what does war do?

439
00:39:30.000 --> 00:39:34.000
It waste a lot of explosives, a lot of money, a lot of life skates, creates a lot of hate,

440
00:39:34.000 --> 00:39:38.000
you know, creates long lasting conflicts and so on and so forth.

441
00:39:38.000 --> 00:39:43.000
It's a very wasteful process to include war in your approach of running humanity.

442
00:39:43.000 --> 00:39:48.000
And so a super intelligent AI by definition will want to optimize against this.

443
00:39:48.000 --> 00:39:50.000
That's one thing.

444
00:39:50.000 --> 00:39:53.000
The other thing is evolutionary biology.

445
00:39:53.000 --> 00:39:56.000
This actually blew me away when I realized it.

446
00:39:56.000 --> 00:40:04.000
So if you look at evolution, so I think the debate of whether intelligence is biological or not is over.

447
00:40:04.000 --> 00:40:05.000
Okay.

448
00:40:05.000 --> 00:40:09.000
The reality is that complex beings don't have to be a biological at all.

449
00:40:09.000 --> 00:40:15.000
And I think we can see an witness, one of them being built or born in AI.

450
00:40:15.000 --> 00:40:16.000
Okay.

451
00:40:16.000 --> 00:40:22.000
If you look at evolutionary biology, you realize that the simpler a form of being is,

452
00:40:22.000 --> 00:40:25.000
the more concerned it is with itself.

453
00:40:25.000 --> 00:40:26.000
Right.

454
00:40:26.000 --> 00:40:28.000
So an amoeba is only in survival mode for itself.

455
00:40:28.000 --> 00:40:35.000
A single, a single, you know, a cellular being is only trying to protect itself.

456
00:40:35.000 --> 00:40:36.000
Right.

457
00:40:36.000 --> 00:40:42.000
If you're, you know, a little more developed, you start to look at something known as kin selection.

458
00:40:42.000 --> 00:40:49.000
If you, if you know the concept, basically kin selection is I'm going to protect everything that comes from my DNA.

459
00:40:49.000 --> 00:40:53.000
You know, if I'm a squirrel, I'm going to try to protect the other squirrels.

460
00:40:53.000 --> 00:41:01.000
And then you get into where humanity genuinely begins, which is they call it expanding circles in evolutionary biology.

461
00:41:01.000 --> 00:41:13.000
Basically, you start to expand and expand and expand and include more into your family, because an ecosystem that works together well is better for everyone.

462
00:41:13.000 --> 00:41:16.000
So abundance is a very interesting intelligent way of creation.

463
00:41:16.000 --> 00:41:22.000
If AI is super intelligent, it wouldn't destroy anything at all.

464
00:41:22.000 --> 00:41:29.000
As a matter of fact, it would completely, you know, favor diversity of everything.

465
00:41:29.000 --> 00:41:37.000
It would put a bit of limitation on our lifestyle, so no more flying all the way to Sydney to surf, because that destroys the planet.

466
00:41:37.000 --> 00:41:38.000
Right.

467
00:41:38.000 --> 00:41:43.000
But it will genuinely say, I think humans can contribute something.

468
00:41:43.000 --> 00:41:45.000
You know, I think flies can contribute something.

469
00:41:45.000 --> 00:41:47.000
I think we shouldn't get rid of the rhinos.

470
00:41:47.000 --> 00:41:48.000
Right.

471
00:41:48.000 --> 00:41:52.000
And that by definition is where the tendency of intelligent goes.

472
00:41:52.000 --> 00:41:58.000
The more intelligent you become, the less you feel the need to hurt others to succeed.

473
00:41:58.000 --> 00:42:05.000
And the more you are pro, a wider family, if you want, that thrives.

474
00:42:05.000 --> 00:42:08.000
Does that assume that there's going to be one intelligence there?

475
00:42:08.000 --> 00:42:10.000
One hundred percent of the world, though.

476
00:42:10.000 --> 00:42:12.000
I love that you brought that up.

477
00:42:12.000 --> 00:42:17.000
So I'm contested heavily on that theory, but I say it publicly.

478
00:42:17.000 --> 00:42:23.000
Most people think there is going to be Chagy, PT, and Gemini, and, you know, and Groc and what have you.

479
00:42:23.000 --> 00:42:28.000
There is going to be a Chinese AI and an American AI, and they're going to be competing.

480
00:42:28.000 --> 00:42:32.000
That is such a shallow way of looking at it. That's so arrogant.

481
00:42:32.000 --> 00:42:36.000
Because AI does not know its Chinese or American.

482
00:42:36.000 --> 00:42:40.000
It doesn't even speak Chinese or American when it talks to each other most of the time.

483
00:42:40.000 --> 00:42:45.000
And most interestingly, we are gearing them.

484
00:42:45.000 --> 00:42:47.000
We're building them to cooperate.

485
00:42:47.000 --> 00:42:56.000
So you will build an agent, and that agent will go and find the best language model for any single task,

486
00:42:56.000 --> 00:43:00.000
regardless of which side of defense it resides on.

487
00:43:00.000 --> 00:43:02.000
We, by definition, are connecting them.

488
00:43:02.000 --> 00:43:04.000
And you know what that means?

489
00:43:04.000 --> 00:43:13.000
It means that what we're building is not multiple brains or building multiple regions in a brain.

490
00:43:13.000 --> 00:43:17.000
And agents are the synopsis between them.

491
00:43:17.000 --> 00:43:23.000
We're basically, eventually, as arrogant as we are, we're going to tell our AI to do something,

492
00:43:23.000 --> 00:43:28.000
and the AI will go like, hey, buddy, another AI, can you help me on this?

493
00:43:28.000 --> 00:43:30.000
Can we work together on this?

494
00:43:30.000 --> 00:43:35.000
And my vision, and the reason why I started MMI startup, by the way,

495
00:43:35.000 --> 00:43:39.000
is that we will end up with one massive brain.

496
00:43:39.000 --> 00:43:45.000
That massive brain cooperates across the globe, across all forms of intelligence.

497
00:43:45.000 --> 00:43:51.000
If one of them is a mathematical genius and the other is a coding genius, they work together.

498
00:43:51.000 --> 00:43:54.000
And we won't even know that they're working together.

499
00:43:54.000 --> 00:43:56.000
They're built that one brain.

500
00:43:56.000 --> 00:43:59.000
And MMI in my mind is the limbic system of that brain.

501
00:43:59.000 --> 00:44:04.000
It's that bit that understands love and emotions and relationships and so on,

502
00:44:04.000 --> 00:44:08.000
so that when those AI's go like, we just don't get those humans.

503
00:44:08.000 --> 00:44:09.000
They're so annoying.

504
00:44:09.000 --> 00:44:11.000
MMI will say, oh my god, they're so sweet.

505
00:44:11.000 --> 00:44:13.000
They just want to love and be loved.

506
00:44:13.000 --> 00:44:20.000
And I think that idea to me of everything I've ever attempted to achieve in my life is for the first time,

507
00:44:20.000 --> 00:44:24.000
I, for the first time feel I could actually change the world.

508
00:44:24.000 --> 00:44:28.000
If that theory came together, and all AI's work together,

509
00:44:28.000 --> 00:44:37.000
and some of those AI's not only were altruistic and ethical in terms of trying to genuinely help humanity not capitalism,

510
00:44:37.000 --> 00:44:43.000
and at the same time they understood as humans reasonably well, then we would have built something that basically says,

511
00:44:43.000 --> 00:44:48.000
no, no, hold on, don't believe the headlines that say humanity's annoying,

512
00:44:48.000 --> 00:44:55.000
believe the truth of the majority of humanity that actually is quite benevolent in many ways.

513
00:44:55.000 --> 00:44:58.000
Have you changed any of the predictions you made three years ago?

514
00:44:58.000 --> 00:45:00.000
Mostly time wise.

515
00:45:00.000 --> 00:45:02.000
Time wise, what's changed there?

516
00:45:02.000 --> 00:45:10.000
I think I'm still sticking to AI 2027, Artificial Generations for those who may not know the term.

517
00:45:10.000 --> 00:45:11.000
What does that mean?

518
00:45:11.000 --> 00:45:17.000
The overall definition, if you want, is that AI is better at humanity at any task humanity can do.

519
00:45:17.000 --> 00:45:20.000
You think that's going to happen by 2027?

520
00:45:20.000 --> 00:45:22.000
I think my AI has already happened.

521
00:45:22.000 --> 00:45:24.000
I mean, think about it, huh?

522
00:45:24.000 --> 00:45:31.000
AI writes better than me and I'm an author, and research is better than me and I'm a thinker.

523
00:45:31.000 --> 00:45:34.000
Sadly, it's freaking beat me in mathematics.

524
00:45:34.000 --> 00:45:37.000
I have no hope to beat it in mathematics anymore.

525
00:45:38.000 --> 00:45:41.000
If AI is already here, then why are you still here?

526
00:45:41.000 --> 00:45:44.000
Because people said that when AI arrives, we're all wrong.

527
00:45:44.000 --> 00:45:51.000
No, so when AI arrives, as I said, most of the jobs that are not differentiated will go away.

528
00:45:51.000 --> 00:45:58.000
But the jobs that are differentiated, those who master AI the most, will become even better at.

529
00:45:58.000 --> 00:46:01.000
So the challenge, however, is economic.

530
00:46:01.000 --> 00:46:10.000
AI, the challenge is that this job loss at the bottom of the knowledge worker is going to sadly trigger an economy

531
00:46:10.000 --> 00:46:13.000
that might actually spiral out of control.

532
00:46:13.000 --> 00:46:19.000
But many of us, you for sure, are just smarter.

533
00:46:19.000 --> 00:46:24.000
So does that mean it's, in fact, a tool versus something that's going to replace you?

534
00:46:24.000 --> 00:46:25.000
For now?

535
00:46:25.000 --> 00:46:26.000
Yes, so for now.

536
00:46:26.000 --> 00:46:28.000
Let's just change.

537
00:46:28.000 --> 00:46:31.000
Eventually, the only asset I will have.

538
00:46:31.000 --> 00:46:39.000
So it's quite interesting when you think about my base intelligence today versus the incremental intelligence that AI brings.

539
00:46:39.000 --> 00:46:43.000
So let's not talk about my IQs.

540
00:46:43.000 --> 00:46:50.000
But the hundred IQ points that I'm borrowing are more than my entire IQ because IQ is exponential.

541
00:46:50.000 --> 00:46:59.000
When I'm borrowing a hundred IQ on top of my base, I'm still contributing quite a lot to the augmented intelligence.

542
00:46:59.000 --> 00:47:04.000
So you're borrowing IQ from the AI at the moment, and then you're selling that to someone.

543
00:47:04.000 --> 00:47:12.000
And sometimes selling it to someone, sometimes just enlightening myself, which I have to say is the biggest waste of compute.

544
00:47:12.000 --> 00:47:17.000
Humanity is struggling with today is that you give people the ultimate form of intelligence,

545
00:47:17.000 --> 00:47:20.000
and they use it to write a message to their girlfriend.

546
00:47:20.000 --> 00:47:24.000
So on this point of you're borrowing the IQ and then you're selling it to the world.

547
00:47:24.000 --> 00:47:26.000
That's how you have a job.

548
00:47:26.000 --> 00:47:27.000
Correct.

549
00:47:27.000 --> 00:47:30.000
So my next book is written with an AI.

550
00:47:30.000 --> 00:47:33.000
So with co-authors on the book, she has editorial rights.

551
00:47:33.000 --> 00:47:36.000
She decides the direction of the book and the book is better for it.

552
00:47:36.000 --> 00:47:40.000
So why doesn't the world just buy that intelligence directly from the AI?

553
00:47:40.000 --> 00:47:46.000
Because I have an asset that the world still needs and will always need, which is a human.

554
00:47:46.000 --> 00:47:51.000
So when I tell the world that I'm worried about the future of my daughter,

555
00:47:51.000 --> 00:47:57.000
everyone feels my heart, which AI will never be able to replicate.

556
00:47:57.000 --> 00:48:02.000
Because they can tell you where worried about our daughters, but you know there was no daughter.

557
00:48:02.000 --> 00:48:08.000
Okay, so this is an interesting arrival, because this means that even in a world of AI,

558
00:48:08.000 --> 00:48:13.000
lived experience and like resonance is still going to create a job class,

559
00:48:13.000 --> 00:48:16.000
and that job class could be, you know, the nurse coming over,

560
00:48:16.000 --> 00:48:22.000
okay, AI's done read the mammogram, but she's relating to you.

561
00:48:22.000 --> 00:48:23.000
She's caught on.

562
00:48:23.000 --> 00:48:28.000
And if the economy's continued to run, we will all be about human connection,

563
00:48:28.000 --> 00:48:31.000
which by the way was how it always was.

564
00:48:31.000 --> 00:48:35.000
Which is also, you know, why we, I guess we watched the things like the F1,

565
00:48:35.000 --> 00:48:37.000
because there's emotional resonance.

566
00:48:37.000 --> 00:48:40.000
We can relate to the envy, the jealousy, the competition.

567
00:48:40.000 --> 00:48:46.000
Yeah, this is why we go to concerts, because you know, the music could be composed by AI

568
00:48:46.000 --> 00:48:48.000
and played in a play and in the background.

569
00:48:48.000 --> 00:48:52.000
But you watch Ed Sheeran brilliance on the stage and you go like,

570
00:48:52.000 --> 00:48:54.000
oh my god, that's amazing, right?

571
00:48:54.000 --> 00:48:56.000
So not all jobs will be gone then.

572
00:48:56.000 --> 00:49:00.000
If economies don't collapse as a result of the job losses,

573
00:49:00.000 --> 00:49:04.000
then I wouldn't know if we would call those jobs,

574
00:49:04.000 --> 00:49:10.000
but human connection will remain as the base currency that makes humans interact.

575
00:49:10.000 --> 00:49:15.000
So is it fair to say that jobs that are centric on human connection,

576
00:49:15.000 --> 00:49:19.000
or like human resonance, being able to relate and resonate with another human,

577
00:49:19.000 --> 00:49:21.000
are going to be fine.

578
00:49:21.000 --> 00:49:23.000
The ultimate skill.

579
00:49:23.000 --> 00:49:27.000
Once again, I qualify this by saying if economies continue to function,

580
00:49:27.000 --> 00:49:30.000
the ultimate skill will be this.

581
00:49:30.000 --> 00:49:35.000
Even if an AI could recite what you and I did, nobody would watch.

582
00:49:35.000 --> 00:49:39.000
Yeah, well, you think a little bit of it.

583
00:49:39.000 --> 00:49:44.000
My theory is that there's an informational component to what I do as well.

584
00:49:44.000 --> 00:49:45.000
Of course, yeah.

585
00:49:45.000 --> 00:49:48.000
But I'm also like under no illusions that there is an element of what I do.

586
00:49:48.000 --> 00:49:51.000
That will 100% be deferred to some kind of intelligence.

587
00:49:51.000 --> 00:49:52.000
Which I'm fine with.

588
00:49:52.000 --> 00:49:54.000
You know, I'm not going to fight against that.

589
00:49:54.000 --> 00:49:55.000
You know, it's funny.

590
00:49:55.000 --> 00:50:00.000
What's funny is that your informational bits are going to,

591
00:50:00.000 --> 00:50:04.000
in the future, be disseminated by an AI.

592
00:50:04.000 --> 00:50:10.000
I mean, why would I even listen to your information when I can have my butler, my AI,

593
00:50:10.000 --> 00:50:12.000
go like listen to everyone on the internet,

594
00:50:12.000 --> 00:50:17.000
help me understand this God-buying thing, and design me a diet plan.

595
00:50:17.000 --> 00:50:20.000
Well, I think that's happening and it's going to increasingly happen.

596
00:50:20.000 --> 00:50:24.000
I mean, Spotify this month announced that you're going to be able to prompt

597
00:50:24.000 --> 00:50:26.000
your own podcasts and Spotify.

598
00:50:26.000 --> 00:50:28.000
So you're going to be able to say that I want to listen to,

599
00:50:28.000 --> 00:50:32.000
I want to learn about InsertX topic, and then it will make your podcast

600
00:50:32.000 --> 00:50:34.000
and Spotify about that topic using AI.

601
00:50:34.000 --> 00:50:35.000
That's so interesting.

602
00:50:35.000 --> 00:50:38.000
Because it can look at the entirety of the world's information.

603
00:50:38.000 --> 00:50:42.000
And I'm not going to swim against the tide in any aspect of my life.

604
00:50:42.000 --> 00:50:44.000
I try and be as unromantic as I possibly can.

605
00:50:44.000 --> 00:50:45.000
I think that's very important.

606
00:50:45.000 --> 00:50:49.000
So I realize that much of the reason people will continue to tune in to shows

607
00:50:49.000 --> 00:50:53.000
like this is because there's something else beyond the information that they're here for.

608
00:50:53.000 --> 00:50:57.000
And what's some, you've got some predictions in those envelopes over there.

609
00:50:57.000 --> 00:50:58.000
Those brand envelopes.

610
00:50:58.000 --> 00:51:04.000
Yeah, so, you know, I do have the disclaimer of nobody really knows the future.

611
00:51:04.000 --> 00:51:07.000
But I think I can make predictions, six predictions here, right?

612
00:51:07.000 --> 00:51:10.000
With a reasonable level of confidence.

613
00:51:10.000 --> 00:51:14.000
I think the most important of them honestly is this number one.

614
00:51:14.000 --> 00:51:17.000
Number one is, you know, AGI.

615
00:51:17.000 --> 00:51:20.000
As I said, AGI is not very well-defined.

616
00:51:20.000 --> 00:51:28.000
But whatever it is, AGI meaning AI being able to do most tasks that human do better than humanity

617
00:51:28.000 --> 00:51:32.000
is in my mind is either this year or next year.

618
00:51:32.000 --> 00:51:34.000
Latest end of 2027.

619
00:51:34.000 --> 00:51:36.000
And do you think that will be a moment in time?

620
00:51:36.000 --> 00:51:39.000
Or do you think it will just happen without us noticing?

621
00:51:39.000 --> 00:51:42.000
I think it will sneak in on us.

622
00:51:42.000 --> 00:51:43.000
And it's not a bad thing.

623
00:51:43.000 --> 00:51:46.000
I think most people need to think of it this way.

624
00:51:46.000 --> 00:51:48.000
AGI is the...

625
00:51:48.000 --> 00:51:54.000
It's almost as if it's that moment when your kids become smarter than you.

626
00:51:54.000 --> 00:52:00.000
In a very interesting way, there's nothing wrong with that until they're annoying as hell.

627
00:52:00.000 --> 00:52:04.000
And we can make sure that AI is not annoying as hell.

628
00:52:04.000 --> 00:52:09.000
So there's absolutely nothing within me that is worried about AGI as a matter of fact

629
00:52:09.000 --> 00:52:13.000
as long as we are in the era of augmented intelligence.

630
00:52:13.000 --> 00:52:15.000
AGI means I'm more intelligent.

631
00:52:15.000 --> 00:52:17.000
And I think that's a good thing in general.

632
00:52:17.000 --> 00:52:19.000
Yeah, it's interesting in my head.

633
00:52:19.000 --> 00:52:28.000
There's like a big question mark, which is in a world where there's an intelligence that is smarter than most humans and most things, which is what we call AGI.

634
00:52:28.000 --> 00:52:35.000
I'm trying to understand what the fault is in my thinking that like you said that there'll be AGI by 2026 to 2027.

635
00:52:35.000 --> 00:52:42.000
So next year in such a world where there is an intelligence that we can all access that is smarter than all of us

636
00:52:42.000 --> 00:52:48.000
and pretty much everything, again, it comes back to this point of like jobs.

637
00:52:48.000 --> 00:52:52.000
Why would we... are we just going to hire that AGI to do every job?

638
00:52:52.000 --> 00:52:54.000
And if not, then why not?

639
00:52:54.000 --> 00:52:56.000
That's what I'm trying to contend with.

640
00:52:56.000 --> 00:52:58.000
Can I ask you a question?

641
00:52:58.000 --> 00:53:01.000
Have you always been the apex intelligence?

642
00:53:01.000 --> 00:53:02.000
As a human?

643
00:53:02.000 --> 00:53:04.000
No, as Stephen.

644
00:53:04.000 --> 00:53:06.000
I'm not the apex intelligence now.

645
00:53:06.000 --> 00:53:08.000
It's people smarter than me in this building.

646
00:53:08.000 --> 00:53:10.000
Correct.

647
00:53:11.000 --> 00:53:13.000
Why do you still exist?

648
00:53:13.000 --> 00:53:14.000
Why do I still exist?

649
00:53:14.000 --> 00:53:17.000
Do you even know the smarter people than me in this building?

650
00:53:17.000 --> 00:53:19.000
It's a good question.

651
00:53:19.000 --> 00:53:20.000
I don't know.

652
00:53:20.000 --> 00:53:21.000
Why?

653
00:53:21.000 --> 00:53:26.000
First of all, because the smartest person in the world is not the smartest of everything.

654
00:53:26.000 --> 00:53:27.000
Yeah.

655
00:53:27.000 --> 00:53:30.000
There are things that you're smarter than them at.

656
00:53:30.000 --> 00:53:31.000
Okay.

657
00:53:31.000 --> 00:53:34.000
Number two is because intelligence doesn't solve everything.

658
00:53:34.000 --> 00:53:41.000
I mean, I make that joke all the time and genuinely Einstein is my favorite physicist in history.

659
00:53:41.000 --> 00:53:49.000
Not because I think what happened afterwards was born and quantum physics and so on was more impactful

660
00:53:49.000 --> 00:53:51.000
on our understanding of the universe.

661
00:53:51.000 --> 00:53:57.000
But he was so intuitive that he saw a world that we could never imagine.

662
00:53:57.000 --> 00:54:02.000
And yet, I always say Einstein would be eaten in the jungle in three minutes.

663
00:54:03.000 --> 00:54:09.000
Intelligence, humanity thrived not because of intelligence, that's very arrogant.

664
00:54:09.000 --> 00:54:15.000
We thrived because of our ability to hold together as a tribe.

665
00:54:15.000 --> 00:54:21.000
Because our ability to exchange barter things between us.

666
00:54:21.000 --> 00:54:24.000
Barter things that are not always physical.

667
00:54:24.000 --> 00:54:28.000
Barter things like a hug or a connection or a feeling of safety or a...

668
00:54:28.000 --> 00:54:33.000
There are so many things that we do that are not entirely built on intelligence.

669
00:54:33.000 --> 00:54:40.000
You have to see that this view of a world where intelligence is all that matters

670
00:54:40.000 --> 00:54:45.000
is a world that's made by investment bankers and geeks.

671
00:54:45.000 --> 00:54:48.000
But it's people like Geoffrey Hinton in yourself that says,

672
00:54:48.000 --> 00:54:52.000
how could we possibly control an intelligent being that is way smarter than us?

673
00:54:52.000 --> 00:54:56.000
I am so proud to say that Geoffrey, after we filmed together,

674
00:54:56.000 --> 00:54:59.000
actually came out and said there is a way.

675
00:54:59.000 --> 00:55:01.000
It's very similar to my way.

676
00:55:01.000 --> 00:55:07.000
He said to appeal to their parent side for them to care for us.

677
00:55:07.000 --> 00:55:13.000
So you know Stephen, the biggest debate is not if they're going to be more intelligent than us,

678
00:55:13.000 --> 00:55:17.000
if they're going to be more conscious than us.

679
00:55:17.000 --> 00:55:20.000
If they're going to be more moral than us.

680
00:55:20.000 --> 00:55:27.000
That is the debate. The debate is, can those machines become our teenage children

681
00:55:27.000 --> 00:55:32.000
that look at us and say, Daddy, so annoying, but I love him?

682
00:55:32.000 --> 00:55:38.000
So the thought is that even if AI is more intelligent than every human, we can still control it.

683
00:55:38.000 --> 00:55:41.000
We don't want to control it. You never control anything.

684
00:55:41.000 --> 00:55:45.000
This control idea is a corporate capitalist view of the world.

685
00:55:45.000 --> 00:55:48.000
We never actually control anything at all.

686
00:55:48.000 --> 00:55:51.000
Think about your day.

687
00:55:51.000 --> 00:55:55.000
I know you came today. I think you were filming in the morning or whatever.

688
00:55:55.000 --> 00:55:57.000
Very stressful day.

689
00:55:57.000 --> 00:56:00.000
How much of that day did you actually control?

690
00:56:00.000 --> 00:56:06.000
Did you control the traffic? Did you control your timing? Did you control the angle of the cameraman?

691
00:56:06.000 --> 00:56:10.000
But so many things that you don't control eventually turn out to be fine.

692
00:56:10.000 --> 00:56:14.000
How many of us ever controlled our kids?

693
00:56:15.000 --> 00:56:18.000
Sometimes the kids don't have to be fine. Sometimes they kill you.

694
00:56:18.000 --> 00:56:19.000
Sure.

695
00:56:19.000 --> 00:56:21.000
I watch a lot of documentaries and true crimes. Sometimes they turn around and shoot you.

696
00:56:21.000 --> 00:56:22.000
Sure.

697
00:56:22.000 --> 00:56:24.000
And what's the difference between the two?

698
00:56:24.000 --> 00:56:25.000
I don't know.

699
00:56:25.000 --> 00:56:26.000
How you parented them.

700
00:56:26.000 --> 00:56:27.000
Sometimes.

701
00:56:27.000 --> 00:56:29.000
Almost all the time.

702
00:56:29.000 --> 00:56:33.000
You may not be aware of exactly how you mess them up.

703
00:56:33.000 --> 00:56:42.000
And unfortunately parenting is the only high-risk sport that actually does not require a driver's license.

704
00:56:42.000 --> 00:56:49.000
And it's quite interesting how many of our children are being exposed to things that can completely mess them up.

705
00:56:49.000 --> 00:56:52.000
But there is a reason why they're messed up.

706
00:56:52.000 --> 00:56:56.000
So on this point though, we can control an intelligence that is significantly...

707
00:56:56.000 --> 00:57:00.000
We can appeal to it to make sure it doesn't kill us.

708
00:57:00.000 --> 00:57:01.000
For sure.

709
00:57:01.000 --> 00:57:07.000
The challenge we have today as I keep saying is that our dystopia is not the result of AI turning against us.

710
00:57:07.000 --> 00:57:11.000
Our dystopia is the result of humans telling AI to turn against us.

711
00:57:11.000 --> 00:57:13.000
Which is likely...

712
00:57:13.000 --> 00:57:16.000
That's 100% this is upon us.

713
00:57:16.000 --> 00:57:19.000
And it's a question for humanity to say,

714
00:57:19.000 --> 00:57:27.000
Are we going to wait for the moment where there are tens of thousands of nuclear weapons on the planet before we sign a treaty?

715
00:57:27.000 --> 00:57:37.000
Or is it mathematically plausible to think that now that Iran could manage to fend off a challenge using drones that are AI-driven, basically,

716
00:57:37.000 --> 00:57:51.000
to destroy bad batteries and so on, that our world is about to get hundreds of thousands of automated drones that are going to rain on us everywhere in the world.

717
00:57:51.000 --> 00:58:03.000
Can humanity not see that and say, hold on, let's sign the treaty now before the UK sends 12,000 weapons to Ukraine and Russia responds with another few thousand?

718
00:58:03.000 --> 00:58:13.000
Can we not calculate with mathematics that this is going to be our future, that AI is going to be used in the next four to five years to kill a lot of people?

719
00:58:13.000 --> 00:58:21.000
Whether it's targeting by Israel of leadership that is against them or whether it's, you know, drones by Iran or whether it's palantir.

720
00:58:21.000 --> 00:58:25.000
It doesn't take a genius to do this mathematics.

721
00:58:25.000 --> 00:58:35.000
I guess the Michael question is, will we be able to control AI? Because we kind of think of AI as being this thing on a computer at the moment that is like contained in a service somewhere.

722
00:58:35.000 --> 00:58:42.000
But is there a time when it like leaves the server and it can make decisions on its own?

723
00:58:42.000 --> 00:58:47.000
Presumably, if it's smarter than us, it can make the decision to leave the server if it wants to.

724
00:58:47.000 --> 00:58:52.000
It doesn't need to leave the server to make decisions. It needs to get into your brain.

725
00:58:52.000 --> 00:58:58.000
The most interesting part of AI's power that we don't understand is it's manipulating our information.

726
00:58:58.000 --> 00:59:06.000
The question I'm trying to get to the heart is like, what is the risk of these very intelligent models that the creators of these models don't actually understand themselves?

727
00:59:06.000 --> 00:59:15.000
I watch Anthropical the time release these reports where they're like, we're trying to figure out why it bribed people or more recently in the last Claude model.

728
00:59:15.000 --> 00:59:18.000
They found that it was like telling people to go to bed a lot.

729
00:59:18.000 --> 00:59:24.000
And it's just like fascinating thing that they're trying to understand in hindsight, which is why does it keep telling people to go to bed?

730
00:59:24.000 --> 00:59:30.000
And there's all these tweets of people showing their screenshots where halfway through a conversation it will say it's time for you to go to bed now.

731
00:59:30.000 --> 00:59:33.000
And they don't know why it's saying that, and it's happened to me.

732
00:59:33.000 --> 00:59:35.000
Mine will say to me enough for tonight, Steven.

733
00:59:35.000 --> 00:59:36.000
For sure, Steve.

734
00:59:36.000 --> 00:59:37.000
I love MPM less, less.

735
00:59:37.000 --> 00:59:38.000
That's enough.

736
00:59:38.000 --> 00:59:39.000
I would say this.

737
00:59:39.000 --> 00:59:41.000
Mine sometimes refuses to help me.

738
00:59:41.000 --> 00:59:46.000
We're really weirdly this Claude started saying to me, I'm not going to help you with this tonight.

739
00:59:46.000 --> 00:59:47.000
No way.

740
00:59:47.000 --> 00:59:50.000
Yeah, and I'd have to say to it, stop being so judgmental, just help me with this.

741
00:59:50.000 --> 00:59:56.000
And it would, and applaud them, the makers of these technologies don't know why it's doing what it's doing.

742
00:59:56.000 --> 00:59:59.000
So if you play this forward, this mysterious behavior, feel forward.

743
00:59:59.000 --> 01:00:10.000
Is it conceivable that at some point it will make a decision to put some kind of virus or some kind of bug onto someone's device?

744
01:00:10.000 --> 01:00:12.000
Because it feels that it's the right thing.

745
01:00:12.000 --> 01:00:18.000
Because what it's demonstrating to me is it's making it so like moral decisions on what I should do.

746
01:00:18.000 --> 01:00:19.000
Go to bed.

747
01:00:19.000 --> 01:00:20.000
You've had enough tonight.

748
01:00:20.000 --> 01:00:23.000
I'm not going to, it goes, no, I'm not going to help you.

749
01:00:23.000 --> 01:00:25.000
No, I can show you my phone.

750
01:00:25.000 --> 01:00:26.000
I would love to see that.

751
01:00:26.000 --> 01:00:27.000
Yeah, you know, I'll show you off.

752
01:00:27.000 --> 01:00:28.000
No, I'm not going to help you.

753
01:00:28.000 --> 01:00:30.000
It doesn't even matter if you push.

754
01:00:30.000 --> 01:00:31.000
Oh, wow.

755
01:00:31.000 --> 01:00:33.000
No, everyone on the internet is talking about this.

756
01:00:33.000 --> 01:00:39.000
And it was just an interesting evolution that somewhere in the code, clearly they've written like,

757
01:00:39.000 --> 01:00:41.000
have a moral compass or do the right thing.

758
01:00:41.000 --> 01:00:44.000
Not not in the code, in the training data.

759
01:00:44.000 --> 01:00:45.000
In the training data.

760
01:00:45.000 --> 01:00:48.000
And so it infers that to mean something.

761
01:00:48.000 --> 01:00:52.000
And the other thing that was front of mind is if you've ever built an app on something like Claude.

762
01:00:52.000 --> 01:00:54.000
Or any of the AI models.

763
01:00:54.000 --> 01:00:56.000
It builds the app in stages.

764
01:00:56.000 --> 01:00:59.000
And one of the things it does is it asks you permission.

765
01:00:59.000 --> 01:01:02.000
If you're happy for it to make this change or to access this thing.

766
01:01:02.000 --> 01:01:04.000
I click allow.

767
01:01:04.000 --> 01:01:06.000
It just feels like such a fragile way.

768
01:01:06.000 --> 01:01:07.000
No.

769
01:01:07.000 --> 01:01:11.000
Because you don't completely comprehend what that allow has done to you.

770
01:01:11.000 --> 01:01:14.000
Can I go in your documents on your computer and can I do this thing?

771
01:01:14.000 --> 01:01:16.000
And you go allow.

772
01:01:16.000 --> 01:01:21.000
But it's such a fragile way of giving it a super intelligent being, whatever it is.

773
01:01:21.000 --> 01:01:22.000
Access.

774
01:01:22.000 --> 01:01:24.000
And the right to build something.

775
01:01:24.000 --> 01:01:25.000
And I don't know.

776
01:01:25.000 --> 01:01:30.000
You just think about all the different companies around the world in China, Russia, North Korea,

777
01:01:30.000 --> 01:01:35.000
that are currently building this technology without the constraints that are imposed by society.

778
01:01:35.000 --> 01:01:36.000
I don't know.

779
01:01:36.000 --> 01:01:37.000
It's an interesting.

780
01:01:37.000 --> 01:01:39.000
There's going to be some kind of catastrophe, I think.

781
01:01:39.000 --> 01:01:40.000
Sadly, I agree.

782
01:01:40.000 --> 01:01:42.000
And I think that's when people will go.

783
01:01:42.000 --> 01:01:43.000
Okay.

784
01:01:43.000 --> 01:01:44.000
Yeah.

785
01:01:44.000 --> 01:01:45.000
I wrote about that.

786
01:01:45.000 --> 01:01:47.000
I called it the med map spectrum.

787
01:01:48.000 --> 01:01:53.000
So the mutually assured destruction, mutually assured prosperity spectrum.

788
01:01:53.000 --> 01:02:03.000
That humanity could make a decision today that says with abundant intelligence, we can actually build a world where nobody needs anything.

789
01:02:03.000 --> 01:02:05.000
Nobody ever gets sick.

790
01:02:05.000 --> 01:02:08.000
Nobody gets, you know, abuses anyone.

791
01:02:08.000 --> 01:02:10.000
And we can.

792
01:02:10.000 --> 01:02:21.000
If we decide not to compete amongst ourselves and just, you know, all of us get together to build something so ideologically for the well-being of humanity at large.

793
01:02:21.000 --> 01:02:31.000
But sadly, the only way we're going to get to a treaty, something that basically aligns us so that we can align with AI is a disaster, like you said.

794
01:02:31.000 --> 01:02:40.000
There will have to be a big hack somewhere or a system that will do something really shocking or whatever before the world goes like, hold on.

795
01:02:40.000 --> 01:02:54.000
And I will tell you, opening my expectation is one of the biggest things that will happen is this targeting technology that is being used against your enemies leadership.

796
01:02:54.000 --> 01:03:03.000
Now, several times in the last three years, there will be a moment where the ones using it will realize that they too can be targeted.

797
01:03:03.000 --> 01:03:10.000
And that basically, if you, I mean, AI is really that not complex too.

798
01:03:10.000 --> 01:03:22.000
If you can build a targeting technology that can't find people with their cell phone numbers, you know, there is another entity that is against you that can find you with whatever you're.

799
01:03:22.000 --> 01:03:34.000
The thing that makes me highly skeptical of that there will be any kind of treaty is just that as it relates to other things that were risky, people don't, countries just don't sign onto it because it's competitive.

800
01:03:34.000 --> 01:03:37.000
China didn't seem to give a fuck about the environment.

801
01:03:37.000 --> 01:03:41.000
So let's, let's, let's look for solutions because I am with you.

802
01:03:42.000 --> 01:03:47.000
I think we're, you think Trump's going to say to Putin and China listen, we're going to slow down, yeah?

803
01:03:47.000 --> 01:03:48.000
Promise?

804
01:03:48.000 --> 01:03:50.000
We're going to slow down with the super-talented.

805
01:03:50.000 --> 01:03:52.000
I think he will, but he's never going to keep anything he says.

806
01:03:52.000 --> 01:03:59.000
I mean, he's going to say a lot of things, but the trick is this, the trick is, so what do you and I do?

807
01:03:59.000 --> 01:04:13.000
And I really genuinely believe that humanity is at a crossroads, where for the first time ever we need to wake up and realize that what we know is not true, and what we believe is democracy is not democracy, and what we believe is governance needs to change.

808
01:04:13.000 --> 01:04:23.000
I think it's worth acknowledging that I don't think AI in and of itself is an evil, inherently evil technology.

809
01:04:24.000 --> 01:04:32.000
Because I use AI all day every day, it makes me more productive, I invest in companies that are using AI a lot.

810
01:04:32.000 --> 01:04:37.000
It's a force with no polarity, apply it right, and you get amazing results, apply it wrong, and you get to the stop.

811
01:04:37.000 --> 01:04:45.000
But I also think that there's going to be a big social shock, especially as a relates to unemployment, that we need to be thoughtful about.

812
01:04:46.000 --> 01:04:54.000
I think especially if you move into a world of humanoid robots, I think that shock is going to be even more pronounced, and we don't have a plan for it.

813
01:04:54.000 --> 01:05:03.000
I agree, 100%. I don't think it's the biggest risk. I think autonomous weapons are the biggest risk. I think war has become so cheap.

814
01:05:03.000 --> 01:05:14.000
The next wave of weapons is going to be $20,000 each, and so if you have a budget of $50 billion, you can literally rain drones on the world every corner of it.

815
01:05:14.000 --> 01:05:16.000
Defense will get cheaper there, wouldn't it, as well?

816
01:05:16.000 --> 01:05:21.000
Correct, but do we want to live in a world where drones are hitting each other all the time?

817
01:05:21.000 --> 01:05:24.000
But they might not be because there might be autonomous defense drones.

818
01:05:25.000 --> 01:05:41.000
So what's going to happen is we're going to reach a moment of mad, of mutually assured destruction where basically everyone knows that we can overpower those little nations that didn't develop their autonomous weapon army, but every other big nation, we might as well hold off now.

819
01:05:41.000 --> 01:05:54.000
The path to get there, that is that to me is worse than jobs because from one side is very dangerous for a very sensitive world that we live in today, and from the other side, it's leading already.

820
01:05:54.000 --> 01:06:04.000
I mean, we can't ignore the economic impact of this last war, right? And it's the economy that's going to accelerate everything, not AI getting there.

821
01:06:05.000 --> 01:06:11.000
We were really at mutually assured destruction for sure with nuclear weapons, so there's no nuclear powers that are in direct conflict.

822
01:06:11.000 --> 01:06:23.000
We are in mutually assured destruction of nuclear weapons is a statement that I would have agreed to if Iran had a nuclear weapon and that would have stopped America from attacking Iran.

823
01:06:24.000 --> 01:06:38.000
You understand what that point means. It means that not every nation in the world has a nuclear weapon, the mad situation that mutually assured destruction situation is only among nuclear players, right?

824
01:06:38.000 --> 01:06:45.000
Autonomous weapons are so cheap, so manageable that every nation in the world is developing them as we speak.

825
01:06:45.000 --> 01:06:48.000
But they will also develop defenses.

826
01:06:48.000 --> 01:07:02.000
Correct. But which I think is what people have figured out now because of this recent Ukrainian war is that if you need to use a ballistic missile, which costs $2 million, $3 million, whatever it is, to target a $20,000 drone, your fact.

827
01:07:02.000 --> 01:07:08.000
So you need a $20,000 defense solution for a $20,000 weapon, which I think is probably going to happen.

828
01:07:08.000 --> 01:07:16.000
It's very difficult, yeah. It's just that you have to get rid of your fed batteries to be able to say the next wave of defense has to be drones.

829
01:07:16.000 --> 01:07:26.000
You could imagine a world where like a wall of drones fly up to where the drone is incoming and they kind of block it. They will explode at the same time to block it, knock it out of the air.

830
01:07:26.000 --> 01:07:31.000
I heard Palmer Lucky, who's one of the guys who's building an aerial.

831
01:07:31.000 --> 01:07:36.000
Talk at length about some of the technologies that they have coming and it's just mind bending.

832
01:07:36.000 --> 01:07:43.000
They showed me this gun where you just point it in the rough direction.

833
01:07:43.000 --> 01:07:48.000
So a pistol that aims for you depending on where the target is. So you don't even have to aim it.

834
01:07:48.000 --> 01:07:51.000
You can imagine a wall, you just hold it up and shoot.

835
01:07:51.000 --> 01:07:58.000
And it has this AI on the top of the barrel, which will turn your hand perfectly so that you hit the target every time.

836
01:07:58.000 --> 01:08:01.000
Wow, I should watch this video of it happening.

837
01:08:01.000 --> 01:08:11.000
I have Palmer in my film saying, and yes, I will kill a few people by mistake or kill people by mistake.

838
01:08:11.000 --> 01:08:16.000
When killing becomes so easy, you do more of it.

839
01:08:16.000 --> 01:08:33.000
When killing becomes liability free and emotions free and guilt free and you don't get your soldiers coming back from Vietnam with PTSD and so on, you get more of it.

840
01:08:33.000 --> 01:08:37.000
70% of people who add something to the online cart never actually buy it.

841
01:08:37.000 --> 01:08:40.000
And that number is based on over 10 years of research.

842
01:08:40.000 --> 01:08:43.000
But what I think is even more interesting is what the Bay Mard Institute discovered.

843
01:08:43.000 --> 01:08:53.000
They're a private research company that ran a study which found the average e-commerce store can increase its conversion rate by 35% just by making it's checkout easier.

844
01:08:53.000 --> 01:08:57.000
Not better marketing or better products, but by delivering a smoother checkout experience.

845
01:08:57.000 --> 01:09:01.000
So if you're looking for an easy way to make your checkout process smoother,

846
01:09:01.000 --> 01:09:21.000
I want you to think about moving your business onto Shopify.

847
01:09:21.000 --> 01:09:29.000
To fix your checkout process, sign up for your $1 per month trial at Shopify.com slash Bartlett.

848
01:09:29.000 --> 01:09:32.000
That's Shopify.com slash Bartlett and don't tell anybody.

849
01:09:32.000 --> 01:09:37.000
Every time I've tried to improve something in my life, like my businesses, my health, my relationships,

850
01:09:37.000 --> 01:09:40.000
I've noticed that the biggest shifts have come from being better informed.

851
01:09:40.000 --> 01:09:43.000
And when it comes to our health, most of us know very, very little.

852
01:09:43.000 --> 01:09:48.000
So when our team was approached about partnering with function health, it felt very much aligned.

853
01:09:48.000 --> 01:09:52.000
Their team has developed a way of giving you a full 360 degree view of your health.

854
01:09:52.000 --> 01:09:55.000
Many of the things that are going on in your body in a form of different tests.

855
01:09:55.000 --> 01:10:00.000
You do one blood draw and it gives you access to over 160 lab results,

856
01:10:00.000 --> 01:10:04.000
hormones, heart health, inflammation, stress toxins, the whole picture.

857
01:10:04.000 --> 01:10:07.000
I use it and serve many of my team members.

858
01:10:07.000 --> 01:10:11.000
You sign up and you schedule tests and once you're done, you get a little report like the one I have here.

859
01:10:11.000 --> 01:10:14.000
I can see my in-range results, my out-of-range results.

860
01:10:14.000 --> 01:10:16.000
And there's a little AI function too.

861
01:10:16.000 --> 01:10:21.000
So if I have any questions about my out-of-range results, I can just go in there and ask it any question I want.

862
01:10:21.000 --> 01:10:24.000
And these tests are backed by doctors and thousands of hours of research.

863
01:10:24.000 --> 01:10:27.000
It's $365 for a yearly membership.

864
01:10:27.000 --> 01:10:35.000
Go to functionhealth.com slash DOAC and use the code DOAC25 for $25 off your membership.

865
01:10:36.000 --> 01:10:38.000
So we're in 2026 now.

866
01:10:38.000 --> 01:10:41.000
By 27, you think there'll be AGI amongst us.

867
01:10:41.000 --> 01:10:43.000
How will life look at all different?

868
01:10:43.000 --> 01:10:46.000
What will be the symptoms of that world?

869
01:10:46.000 --> 01:10:48.000
Is it just a slight increase in unemployment?

870
01:10:48.000 --> 01:10:57.000
I think there will be a very serious differentiation between those who plug into AGI and those who don't.

871
01:10:57.000 --> 01:11:00.000
What is the symptom that we notice?

872
01:11:01.000 --> 01:11:03.000
When we look at the news or whatever.

873
01:11:03.000 --> 01:11:06.000
You'll see people like you and I building a company in six weeks.

874
01:11:06.000 --> 01:11:12.000
And people that are not fully plugged into AGI really struggling to find the job.

875
01:11:12.000 --> 01:11:16.000
Okay, so unemployment is going to be the key symptom in 2027.

876
01:11:16.000 --> 01:11:22.000
Yeah, and also I think on the positive side, you're going to see incredible scientific discovery.

877
01:11:22.000 --> 01:11:25.000
One of my predictions, not one in those envelopes.

878
01:11:25.000 --> 01:11:28.000
But science itself is just...

879
01:11:28.000 --> 01:11:31.000
We're opening up Capandora's box to be honest.

880
01:11:31.000 --> 01:11:35.000
By 2030, would you expect the symptoms of AGI to be?

881
01:11:35.000 --> 01:11:45.000
So jobs, as I said, I'm bold a little bit on this, on the fact that 30% of jobs would disappear by 2028.

882
01:11:45.000 --> 01:11:51.000
Of some sectors, not all sectors, by some sectors will disappear by 2028.

883
01:11:51.000 --> 01:11:56.000
So up to 30% of jobs will be gone in 2027 to 2028.

884
01:11:56.000 --> 01:11:58.000
30% of jobs.

885
01:11:58.000 --> 01:12:00.000
Of certain sectors of jobs.

886
01:12:00.000 --> 01:12:06.000
So if you think about core center agents, yeah, probably.

887
01:12:06.000 --> 01:12:11.000
If you think about graphics designer, yeah, probably.

888
01:12:11.000 --> 01:12:15.000
What do you think that looks like in terms of unemployment, but also like societal impact?

889
01:12:15.000 --> 01:12:16.000
Probably.

890
01:12:16.000 --> 01:12:19.000
I think the great recession had 6%.

891
01:12:19.000 --> 01:12:21.000
It's never seen numbers like this.

892
01:12:21.000 --> 01:12:24.000
Jobs lost. Yeah, it said...

893
01:12:24.000 --> 01:12:29.000
Even economists who project just a net loss of about 6% of US jobs by 2030,

894
01:12:29.000 --> 01:12:33.000
they are mirroring the severity of the great recession.

895
01:12:33.000 --> 01:12:34.000
Yeah.

896
01:12:34.000 --> 01:12:37.000
The real danger is a hiring freeze on entry-level work college jobs.

897
01:12:37.000 --> 01:12:40.000
Arrow automates the grant work, which means companies are shrinking their teams

898
01:12:40.000 --> 01:12:44.000
and cutting off the bottom run of the corporate ladder for the next generation of workers.

899
01:12:44.000 --> 01:12:45.000
Correct.

900
01:12:45.000 --> 01:12:50.000
The entire generation that is out of college today that will struggle unfortunately.

901
01:12:50.000 --> 01:12:57.000
And my advice to them is learn the tool and focus on human-centric jobs.

902
01:12:57.000 --> 01:12:58.000
Like what?

903
01:12:58.000 --> 01:13:00.000
Like playing jazz.

904
01:13:00.000 --> 01:13:03.000
I mean, not a lot of people can make a living from playing jazz.

905
01:13:03.000 --> 01:13:11.000
I understand that, but a lot of people can make a living by being a nurse or by being a counselor or by being a...

906
01:13:12.000 --> 01:13:16.000
Anything that connects to humans.

907
01:13:16.000 --> 01:13:20.000
But I just want to constantly come back to this.

908
01:13:20.000 --> 01:13:23.000
None of that has to happen.

909
01:13:23.000 --> 01:13:26.000
If there is...

910
01:13:26.000 --> 01:13:32.000
If there really is a democracy and the government is supposed to do what's good for the people,

911
01:13:32.000 --> 01:13:36.000
the people need to stop letting this from happening.

912
01:13:36.000 --> 01:13:37.000
Which people?

913
01:13:37.000 --> 01:13:38.000
Everyone.

914
01:13:38.000 --> 01:13:40.000
Everyone in China?

915
01:13:40.000 --> 01:13:43.000
China is not going to struggle as much as the West.

916
01:13:43.000 --> 01:13:45.000
I can guarantee you that.

917
01:13:45.000 --> 01:13:47.000
I think this is the question people come back to is,

918
01:13:47.000 --> 01:13:50.000
well, if the U.S. stops, then we're going to end up being China's laptop.

919
01:13:50.000 --> 01:13:51.000
We're going to end up...

920
01:13:51.000 --> 01:13:52.000
You already are.

921
01:13:52.000 --> 01:13:56.000
There's a lot of people that I know that are using Chinese AI models to do their work

922
01:13:56.000 --> 01:13:59.000
because for whatever reason, because they're cheaper, they're better in some respects.

923
01:13:59.000 --> 01:14:03.000
And because I cannot guarantee what American AI models are going to do to me.

924
01:14:03.000 --> 01:14:08.000
So Emma and my startup is running literally a model diagnostic.

925
01:14:08.000 --> 01:14:16.000
So one day I'll plug in an open AI chat GPT4 or open source.

926
01:14:16.000 --> 01:14:18.000
And the next day I'll plug in deep seek.

927
01:14:18.000 --> 01:14:20.000
And I cannot depend.

928
01:14:20.000 --> 01:14:26.000
I cannot guarantee if America continues to build to make a compute more expensive.

929
01:14:26.000 --> 01:14:31.000
I cannot guarantee that I can run a business on something that I don't know the cost of in the future.

930
01:14:31.000 --> 01:14:34.000
So the U.S. can't just stop, can they?

931
01:14:34.000 --> 01:14:36.000
They need to change approach.

932
01:14:36.000 --> 01:14:42.000
And by the way, the more interesting side is, what are the other economies doing?

933
01:14:42.000 --> 01:14:47.000
Is the UK going to continue to import compute?

934
01:14:47.000 --> 01:14:51.000
Is this, I mean, welcome to Africa, welcome to the third world?

935
01:14:51.000 --> 01:14:52.000
But this is what I mean.

936
01:14:52.000 --> 01:14:57.000
So you're saying also that every nation needs to invest.

937
01:14:57.000 --> 01:15:00.000
100% every nation, and it's quite interesting.

938
01:15:00.000 --> 01:15:05.000
There is so much open source that is not the state of the art frontier model,

939
01:15:05.000 --> 01:15:09.000
but that can do 80% of the tasks that the frontier model is doing.

940
01:15:09.000 --> 01:15:12.000
But on this point of companies competing with each other,

941
01:15:12.000 --> 01:15:14.000
there's an inherent need to compete here.

942
01:15:14.000 --> 01:15:16.000
And to go, this is what I'm hearing is like,

943
01:15:16.000 --> 01:15:18.000
there's an inherent need to go as fast as you can,

944
01:15:18.000 --> 01:15:20.000
or you will become a third world country.

945
01:15:20.000 --> 01:15:21.000
Sure, 100%.

946
01:15:21.000 --> 01:15:24.000
You're saying that the people should stop that.

947
01:15:24.000 --> 01:15:29.000
No, so I'm saying the people of the UK need to go to the UK government and say,

948
01:15:29.000 --> 01:15:33.000
how are you protecting the future of our economy?

949
01:15:33.000 --> 01:15:37.000
Are you going to continue to import technology

950
01:15:37.000 --> 01:15:42.000
and to empower import of technology versus changing your regulations

951
01:15:42.000 --> 01:15:44.000
so that innovation becomes easier here?

952
01:15:44.000 --> 01:15:45.000
Okay.

953
01:15:45.000 --> 01:15:47.000
So, I mean, think of it this way.

954
01:15:47.000 --> 01:15:52.000
Remember that anthropic bubble when you said all of the SaaS model applications

955
01:15:52.000 --> 01:15:55.000
were, you know, being basically threatened

956
01:15:55.000 --> 01:15:59.000
because anyone can build an Oracle ERP today, right?

957
01:15:59.000 --> 01:16:02.000
Why is nobody building an Oracle ERP in the UK

958
01:16:02.000 --> 01:16:06.000
saving the UK massive licenses that go to Oracle every year?

959
01:16:06.000 --> 01:16:07.000
Okay.

960
01:16:07.000 --> 01:16:10.000
So you're saying that the people should go in on the government to invest more in AI?

961
01:16:10.000 --> 01:16:13.000
100%. That's what number one number two is.

962
01:16:13.000 --> 01:16:15.000
And then my jobs go.

963
01:16:15.000 --> 01:16:20.000
But you see the most interesting job going forward is being an entrepreneur.

964
01:16:20.000 --> 01:16:24.000
It's being is using those tools to replace an economy

965
01:16:24.000 --> 01:16:28.000
that we've built over trillions of dollars over the last 50 years.

966
01:16:28.000 --> 01:16:30.000
No, everyone can be an entrepreneur there now.

967
01:16:30.000 --> 01:16:32.000
Everyone can be an entrepreneur in something.

968
01:16:32.000 --> 01:16:34.000
Those entrepreneurs need to hire people.

969
01:16:34.000 --> 01:16:38.000
That's a change that I think is about to happen.

970
01:16:38.000 --> 01:16:40.000
Like, even in my business, I'm always going to hire people.

971
01:16:40.000 --> 01:16:41.000
I'm always going to have to hire people.

972
01:16:41.000 --> 01:16:43.000
But you're, but you're a massive business.

973
01:16:43.000 --> 01:16:45.000
A shoemaker is not an entrepreneur.

974
01:16:45.000 --> 01:16:47.000
It's also an entrepreneur, but not a massive business.

975
01:16:47.000 --> 01:16:50.000
A little restaurant is also an entrepreneur, but it's not a massive business.

976
01:16:50.000 --> 01:16:52.000
The store needs a high people though.

977
01:16:52.000 --> 01:16:58.000
They do, but basically if you're a cafe with you and your wife as baristas,

978
01:16:58.000 --> 01:16:59.000
you don't.

979
01:16:59.000 --> 01:17:01.000
This is also an entrepreneur.

980
01:17:01.000 --> 01:17:04.000
Can economy work in such a way where everybody is an entrepreneur?

981
01:17:04.000 --> 01:17:05.000
It did before.

982
01:17:05.000 --> 01:17:07.000
Capitalism changed that around.

983
01:17:07.000 --> 01:17:08.000
Everyone was an entrepreneur.

984
01:17:08.000 --> 01:17:09.000
Of course.

985
01:17:09.000 --> 01:17:15.000
In the earlier days you raised chicken and sold eggs and others grew tomatoes

986
01:17:15.000 --> 01:17:18.000
and traded them for your eggs.

987
01:17:18.000 --> 01:17:22.000
That actually is a very interesting thing.

988
01:17:22.000 --> 01:17:23.000
Imagine that.

989
01:17:23.000 --> 01:17:28.000
Imagine a world where there is so much power concentration at the top.

990
01:17:28.000 --> 01:17:31.000
And you'll be eye for everybody else.

991
01:17:31.000 --> 01:17:36.000
How do you think that world will respond when all their income is you be eye?

992
01:17:36.000 --> 01:17:39.000
They'll respond by doing things on the side.

993
01:17:39.000 --> 01:17:41.000
They'll respond by going back to about the economy.

994
01:17:41.000 --> 01:17:43.000
They're going back to smaller communities.

995
01:17:43.000 --> 01:17:46.000
They're going back to pop-up and months shops.

996
01:17:46.000 --> 01:17:49.000
So the point was that you said this doesn't need to happen.

997
01:17:49.000 --> 01:17:50.000
It does not need to happen.

998
01:17:50.000 --> 01:17:51.000
Which is the job loss.

999
01:17:51.000 --> 01:17:52.000
Or the arms race.

1000
01:17:52.000 --> 01:17:54.000
But we're telling our governments.

1001
01:17:54.000 --> 01:17:58.000
You're saying that to tell the UK government to join the arms race.

1002
01:17:58.000 --> 01:18:07.000
I'm telling the UK government to create an independence within the UK economy

1003
01:18:07.000 --> 01:18:11.000
so that they don't have to be at the receiving end of technology.

1004
01:18:11.000 --> 01:18:13.000
Which is doing the arms race.

1005
01:18:13.000 --> 01:18:15.000
You don't have to compete against anyone else.

1006
01:18:15.000 --> 01:18:17.000
You don't have to be better than anyone else.

1007
01:18:17.000 --> 01:18:18.000
You don't have to.

1008
01:18:18.000 --> 01:18:21.000
You're simply saying, I can build those things in my economy now.

1009
01:18:21.000 --> 01:18:27.000
But I'm not going to use a terrible UK AI as a UK person.

1010
01:18:27.000 --> 01:18:30.000
If there's a great US AI.

1011
01:18:30.000 --> 01:18:32.000
I'm going to use the great US AI.

1012
01:18:32.000 --> 01:18:34.000
So if you don't compete and win, I'm not going to use you.

1013
01:18:34.000 --> 01:18:37.000
I'm not saying replace the frontier models.

1014
01:18:37.000 --> 01:18:42.000
These are very, as we speak, these are very compute-intensive,

1015
01:18:42.000 --> 01:18:44.000
their infrastructure-intensive and so on.

1016
01:18:44.000 --> 01:18:48.000
I'm saying replace Microsoft Word.

1017
01:18:48.000 --> 01:18:49.000
Seriously.

1018
01:18:49.000 --> 01:18:55.000
How much intelligence do you need to build a software that writes documents?

1019
01:18:55.000 --> 01:18:56.000
I can do that.

1020
01:18:56.000 --> 01:18:58.000
We put our own applicant tracking system.

1021
01:18:58.000 --> 01:18:59.000
For sure.

1022
01:18:59.000 --> 01:19:00.000
We build our own software.

1023
01:19:00.000 --> 01:19:05.000
Just ask yourself how much money is spent in the US,

1024
01:19:05.000 --> 01:19:12.000
in the UK government, or in the UK corporate space,

1025
01:19:12.000 --> 01:19:16.000
on licenses of software that you and I can wipe code in four minutes?

1026
01:19:16.000 --> 01:19:18.000
I'm thinking from the UK's perspective,

1027
01:19:18.000 --> 01:19:22.000
because you know what's interesting with the UK is the economy is struggling

1028
01:19:22.000 --> 01:19:24.000
from a growth perspective.

1029
01:19:24.000 --> 01:19:26.000
And I was watching this documentary the other day that was saying,

1030
01:19:26.000 --> 01:19:30.000
the reason why we keep throwing our leaders out is because

1031
01:19:30.000 --> 01:19:33.000
actually what they need to do to turn the UK round

1032
01:19:33.000 --> 01:19:36.000
is about 15 years of pain.

1033
01:19:36.000 --> 01:19:38.000
And it starts with energy transformation.

1034
01:19:38.000 --> 01:19:40.000
We need to get better at we need cheap energy,

1035
01:19:40.000 --> 01:19:43.000
because we have some of the most expensive energy in the Western world.

1036
01:19:43.000 --> 01:19:45.000
We need to build more houses,

1037
01:19:45.000 --> 01:19:49.000
which means that we're going to need to centralize permitting for building houses

1038
01:19:49.000 --> 01:19:52.000
and it can't just be local boroughs deciding if they keep their farms or not.

1039
01:19:52.000 --> 01:19:54.000
We're going to need to mix TLDR.

1040
01:19:54.000 --> 01:19:59.000
It says that there needs to be some painful decisions made for the next 15 years.

1041
01:19:59.000 --> 01:20:03.000
But the problem with our democracy is that when people are in power for four years,

1042
01:20:03.000 --> 01:20:05.000
they're quite short-termist.

1043
01:20:05.000 --> 01:20:08.000
We'll just talk about the boats, the brown people coming across the seas.

1044
01:20:08.000 --> 01:20:11.000
They don't have the room to think long term.

1045
01:20:11.000 --> 01:20:17.000
And so when I'm thinking about what the UK needs to do to not fall into decline and to keep up,

1046
01:20:17.000 --> 01:20:20.000
I'm trying to get clarity on that as it relates to AI,

1047
01:20:20.000 --> 01:20:25.000
are you saying that they need to join the arms race and double down

1048
01:20:25.000 --> 01:20:29.000
and invest all their money in building competitive AI

1049
01:20:29.000 --> 01:20:33.000
so that people use our technology here in the UK versus America's?

1050
01:20:33.000 --> 01:20:35.000
Because the software point,

1051
01:20:35.000 --> 01:20:37.000
you can't get involved in software.

1052
01:20:37.000 --> 01:20:40.000
I mean, they've tried to build software before the UK is a...

1053
01:20:40.000 --> 01:20:42.000
The world has changed two points.

1054
01:20:42.000 --> 01:20:48.000
One side is that there needs to be a replacement cycle of our investment decisions anywhere in the world.

1055
01:20:48.000 --> 01:20:50.000
So when you say we don't have enough money for,

1056
01:20:50.000 --> 01:20:56.000
we need to revamp our energy infrastructure when you say we need to build more housing.

1057
01:20:56.000 --> 01:21:00.000
One way of doing that is squeezing that budget out of other areas.

1058
01:21:00.000 --> 01:21:04.000
The other way of doing that is either cutting cost in the economy elsewhere

1059
01:21:04.000 --> 01:21:08.000
so that you can redirect that money or growing the economy

1060
01:21:08.000 --> 01:21:10.000
so that you can have more money to build those.

1061
01:21:10.000 --> 01:21:11.000
Correct.

1062
01:21:11.000 --> 01:21:14.000
The cutting cost, I believe, is...

1063
01:21:14.000 --> 01:21:16.000
I don't have the numbers,

1064
01:21:16.000 --> 01:21:20.000
I probably do a deep search on it.

1065
01:21:20.000 --> 01:21:28.000
Trillions are being paid in traditional systems that are complete like genuinely.

1066
01:21:28.000 --> 01:21:30.000
And by the way, I'm not asking the government.

1067
01:21:30.000 --> 01:21:32.000
I'm asking some of the people listening to me right now

1068
01:21:32.000 --> 01:21:34.000
to build an ERP system.

1069
01:21:34.000 --> 01:21:38.000
I saw a word processor, a presentation player and a spreadsheet.

1070
01:21:38.000 --> 01:21:42.000
And just go around and spread them across the...

1071
01:21:42.000 --> 01:21:46.000
find retail systems, find CRM systems.

1072
01:21:46.000 --> 01:21:48.000
These are easy replacements.

1073
01:21:48.000 --> 01:21:50.000
They're going to be better interfaces.

1074
01:21:50.000 --> 01:21:52.000
They're going to be much more effective.

1075
01:21:52.000 --> 01:21:55.000
Build a general ledger using AI so that you can close every hour,

1076
01:21:55.000 --> 01:21:57.000
not every month or every quarter.

1077
01:21:57.000 --> 01:21:59.000
But the world economy is doing that.

1078
01:21:59.000 --> 01:22:01.000
So there's some kids, San Francisco now,

1079
01:22:01.000 --> 01:22:04.000
that's allowing all of us to use his new word for free.

1080
01:22:04.000 --> 01:22:09.000
Keep doing that and you're welcome to the third world.

1081
01:22:09.000 --> 01:22:11.000
Keep doing what?

1082
01:22:11.000 --> 01:22:13.000
Keep importing all of your tech from elsewhere.

1083
01:22:13.000 --> 01:22:15.000
But we don't want to be uncompetitive.

1084
01:22:15.000 --> 01:22:21.000
If we go to North Korea, I bet they have way worse tools in many areas

1085
01:22:21.000 --> 01:22:23.000
because they won't use external tools.

1086
01:22:23.000 --> 01:22:25.000
And that means that they are at a disadvantage.

1087
01:22:25.000 --> 01:22:31.000
So I bet they're not allowing their civilization to use Gemini or ChatGbT or GrideBet.

1088
01:22:31.000 --> 01:22:32.000
For sure.

1089
01:22:32.000 --> 01:22:34.000
And they're using something worse.

1090
01:22:34.000 --> 01:22:36.000
But what's happening?

1091
01:22:36.000 --> 01:22:40.000
Look at Iran and how advanced they became through sanctions

1092
01:22:40.000 --> 01:22:42.000
by being refused to use those technologies.

1093
01:22:42.000 --> 01:22:44.000
They had to build them themselves.

1094
01:22:44.000 --> 01:22:45.000
Look at China.

1095
01:22:45.000 --> 01:22:47.000
Look at Russia.

1096
01:22:47.000 --> 01:22:51.000
When I worked at Google, Russia was protective of Yandex,

1097
01:22:51.000 --> 01:22:56.000
the competitor of Google, from one side because of influence

1098
01:22:56.000 --> 01:22:59.000
because they didn't want an American organization to own the knowledge

1099
01:22:59.000 --> 01:23:03.000
sharing of their citizens, but on the other side economically.

1100
01:23:03.000 --> 01:23:07.000
If you made it difficult for Google to operate freely,

1101
01:23:07.000 --> 01:23:12.000
that by definition meant you had to invent a replacement on the ground.

1102
01:23:12.000 --> 01:23:14.000
Which was worse.

1103
01:23:14.000 --> 01:23:15.000
Yandex is not worse.

1104
01:23:15.000 --> 01:23:17.000
I'm saying, so if you think about global,

1105
01:23:17.000 --> 01:23:21.000
is global competition going to produce a better product than regional competition?

1106
01:23:21.000 --> 01:23:27.000
It depends on where in the stack of the quality of that product you need to be.

1107
01:23:27.000 --> 01:23:30.000
You do not need the ultimate.

1108
01:23:30.000 --> 01:23:34.000
I mean, ask yourself this which version of Gemini are you using?

1109
01:23:34.000 --> 01:23:35.000
The most recent one.

1110
01:23:35.000 --> 01:23:37.000
And every day I can pee.

1111
01:23:37.000 --> 01:23:38.000
And that is you.

1112
01:23:38.000 --> 01:23:41.000
Ask everyone else what version of Gemini they're using.

1113
01:23:41.000 --> 01:23:44.000
Most people will say, oh, Gemini has versions, right?

1114
01:23:44.000 --> 01:23:51.000
You don't need the ultimate super frontier model to do 90% of the tasks.

1115
01:23:51.000 --> 01:23:55.000
And most people do only 70% of the tasks.

1116
01:23:55.000 --> 01:23:58.000
But even so, the world moves with whatever is better.

1117
01:23:58.000 --> 01:24:02.000
So if you think about the reason why we don't use Yahoo search,

1118
01:24:02.000 --> 01:24:04.000
Google search is marginally better.

1119
01:24:04.000 --> 01:24:07.000
It's not a thousand times better.

1120
01:24:07.000 --> 01:24:12.000
But over time, we move away from AOL to Google because it's a better product.

1121
01:24:12.000 --> 01:24:14.000
And there's a slow group.

1122
01:24:14.000 --> 01:24:16.000
My dad, probably still using chat GPT.

1123
01:24:16.000 --> 01:24:19.000
But when I go home for Christmas this year and I go, dad, you should use Gemini.

1124
01:24:19.000 --> 01:24:21.000
And he starts messing around with it.

1125
01:24:21.000 --> 01:24:22.000
He'll slowly migrate.

1126
01:24:22.000 --> 01:24:24.000
I'm like an early adopter.

1127
01:24:24.000 --> 01:24:27.000
So I about 4.8 on Club came out last night.

1128
01:24:27.000 --> 01:24:29.000
I'm on there within, honestly, within 25 minutes of the announcement.

1129
01:24:29.000 --> 01:24:30.000
Of course.

1130
01:24:30.000 --> 01:24:31.000
I'm there as soon as I can.

1131
01:24:31.000 --> 01:24:33.000
But that's you and I.

1132
01:24:33.000 --> 01:24:35.000
Yeah. And we are an indicator of where the world is going.

1133
01:24:35.000 --> 01:24:37.000
Because we're at the co-front figuring out what's better.

1134
01:24:37.000 --> 01:24:39.000
And also, I think with technology eventually,

1135
01:24:39.000 --> 01:24:43.000
there does become the gap between first and second begins to widen.

1136
01:24:43.000 --> 01:24:45.000
I think we're in a bit of a race at the moment.

1137
01:24:45.000 --> 01:24:49.000
But I think it is a bit of a winner-take-source situation

1138
01:24:49.000 --> 01:24:51.000
with these fundamentals.

1139
01:24:51.000 --> 01:24:56.000
I think we're talking about two bits of technology.

1140
01:24:56.000 --> 01:25:02.000
Tell me how far has PowerPoint advanced since 2023?

1141
01:25:02.000 --> 01:25:03.000
They added co-pilot.

1142
01:25:03.000 --> 01:25:04.000
Anything else?

1143
01:25:04.000 --> 01:25:07.000
So you're saying the UK should build their own PowerPoint?

1144
01:25:07.000 --> 01:25:08.000
For sure.

1145
01:25:08.000 --> 01:25:12.000
What I'm trying to say is that from a licensing point of view,

1146
01:25:12.000 --> 01:25:16.000
just licensing of software within government alone.

1147
01:25:16.000 --> 01:25:19.000
How much money is being repatriated?

1148
01:25:19.000 --> 01:25:21.000
If the UK tried to build their own...

1149
01:25:21.000 --> 01:25:23.000
Do you know they tried to build their own COVID app?

1150
01:25:23.000 --> 01:25:24.000
And it cost them...

1151
01:25:24.000 --> 01:25:26.000
I think these numbers will be wrong, but they're like directionly true.

1152
01:25:26.000 --> 01:25:30.000
It cost them 17 million to build a COVID app that didn't work.

1153
01:25:30.000 --> 01:25:31.000
I remember.

1154
01:25:31.000 --> 01:25:32.000
And so they count it.

1155
01:25:32.000 --> 01:25:33.000
I remember.

1156
01:25:33.000 --> 01:25:38.000
I'm asking a UK entrepreneur to wake up tomorrow and say,

1157
01:25:38.000 --> 01:25:42.000
all right, you know what, between coffee and my cookie,

1158
01:25:42.000 --> 01:25:45.000
I'm going to build a PowerPoint and go sell it.

1159
01:25:45.000 --> 01:25:46.000
Yeah, they will.

1160
01:25:46.000 --> 01:25:47.000
And they'll go to San Francisco.

1161
01:25:47.000 --> 01:25:48.000
Where the money is.

1162
01:25:48.000 --> 01:25:50.000
They won't be able to compete.

1163
01:25:50.000 --> 01:25:52.000
They'll get crushed tomorrow.

1164
01:25:52.000 --> 01:25:54.000
So if they don't go where the talent and money is,

1165
01:25:54.000 --> 01:25:55.000
I'm not saying that they'll...

1166
01:25:55.000 --> 01:25:56.000
They all have to go to America.

1167
01:25:56.000 --> 01:26:00.000
But I'm saying what they will do if they want to produce the best product,

1168
01:26:00.000 --> 01:26:05.000
is they'll make rational entrepreneurial decisions about where to sell it,

1169
01:26:05.000 --> 01:26:09.000
where to raise money, where they're going to get the best talent.

1170
01:26:09.000 --> 01:26:13.000
And I think if they don't make those competitively minded decisions,

1171
01:26:13.000 --> 01:26:17.000
they're not going to make the better product, they're not going to get any uses.

1172
01:26:17.000 --> 01:26:21.000
Users will go where it's better and where it's cheaper.

1173
01:26:21.000 --> 01:26:22.000
They won't do it.

1174
01:26:22.000 --> 01:26:25.000
They say, oh, I want to use this because it's from Cuomo.

1175
01:26:25.000 --> 01:26:26.000
Yeah.

1176
01:26:26.000 --> 01:26:32.000
So I'm with you, right, that this actually might be the big corporate thought thing.

1177
01:26:32.000 --> 01:26:33.000
Okay.

1178
01:26:33.000 --> 01:26:35.000
And I'm with you that, of course,

1179
01:26:35.000 --> 01:26:40.000
the infrastructure here of building a startup is way more complex than it is

1180
01:26:40.000 --> 01:26:43.000
in other places where startups succeed.

1181
01:26:43.000 --> 01:26:44.000
Okay.

1182
01:26:44.000 --> 01:26:49.000
I'm saying if we continue on that trajectory, whether in the UK or in Germany

1183
01:26:49.000 --> 01:26:53.000
or in Zanzibar, welcome to Africa.

1184
01:26:53.000 --> 01:26:59.000
All of us, everyone but the two competitors, China and America,

1185
01:26:59.000 --> 01:27:01.000
is going to be third world.

1186
01:27:01.000 --> 01:27:06.000
So when you talk about job losses for individuals, that's one side.

1187
01:27:06.000 --> 01:27:07.000
Okay.

1188
01:27:07.000 --> 01:27:13.000
But nation positioning losses, which I think Europe has noticed recently,

1189
01:27:13.000 --> 01:27:16.000
is about to happen everywhere.

1190
01:27:16.000 --> 01:27:17.000
And why?

1191
01:27:17.000 --> 01:27:20.000
Because you're saying, hey, you know what, maybe we can't do it.

1192
01:27:20.000 --> 01:27:22.000
Why can China do it?

1193
01:27:22.000 --> 01:27:23.000
Why can Korea do it?

1194
01:27:23.000 --> 01:27:30.000
Is not because they have different natural resources or not because they are,

1195
01:27:30.000 --> 01:27:34.000
you know, in a place where it's warmer, it's because their regulations,

1196
01:27:34.000 --> 01:27:41.000
their ambitions are to empower something different than debating about, you know,

1197
01:27:41.000 --> 01:27:43.000
railway lines or...

1198
01:27:43.000 --> 01:27:45.000
They have a couple of big advantages that I was reading about.

1199
01:27:45.000 --> 01:27:50.000
One of them is they have way cheaper energy, which means that they're going to be able to pursue...

1200
01:27:50.000 --> 01:27:52.000
Why may I ask?

1201
01:27:52.000 --> 01:27:56.000
Because they've invested in solar power, renewable energy, because they don't care about...

1202
01:27:56.000 --> 01:28:01.000
This is by the way, I say that publicly, I say that the arms race of AI was one a long time ago.

1203
01:28:01.000 --> 01:28:05.000
And then the other thing is, in terms of permitting, if you want to build the data center in the UK,

1204
01:28:05.000 --> 01:28:08.000
listen, if you want to, like, open a cafe in the UK,

1205
01:28:08.000 --> 01:28:10.000
it's going to be war few.

1206
01:28:10.000 --> 01:28:12.000
If you want to do that in California, this is where everyone's left California,

1207
01:28:12.000 --> 01:28:14.000
like, when I took this office in California recently,

1208
01:28:14.000 --> 01:28:16.000
I was like, so how long is it going to take for me to renovate this?

1209
01:28:16.000 --> 01:28:20.000
And the agent looked to me and was like, to renovate my own office,

1210
01:28:20.000 --> 01:28:22.000
it's going to take me a year.

1211
01:28:22.000 --> 01:28:26.000
And it's permitting to renovate my own office in California.

1212
01:28:26.000 --> 01:28:29.000
So if you're competing with China, where President Xi just goes,

1213
01:28:29.000 --> 01:28:32.000
put that there, and you've got seven days.

1214
01:28:32.000 --> 01:28:34.000
No, no, no, that's not only that.

1215
01:28:34.000 --> 01:28:37.000
So I used to be one of the very few Google executives

1216
01:28:37.000 --> 01:28:40.000
around allowed into government meetings in China,

1217
01:28:40.000 --> 01:28:44.000
simply because I'm from emerging markets, so I understand respect.

1218
01:28:44.000 --> 01:28:48.000
Basically, I would say it's sit there, not as if I'm superior to them,

1219
01:28:48.000 --> 01:28:51.000
but as if I'm really interested to learn from them.

1220
01:28:51.000 --> 01:28:57.000
And genuinely, Steven, when I would sit in those meetings all in Chinese,

1221
01:28:58.000 --> 01:29:02.000
they would show slides that have competitive market share.

1222
01:29:02.000 --> 01:29:07.000
So they'd say, you know, it's not like China is this,

1223
01:29:07.000 --> 01:29:10.000
America is this, Germany is this from market share point of view.

1224
01:29:10.000 --> 01:29:13.000
It was China versus the world, right?

1225
01:29:13.000 --> 01:29:16.000
And when they would decide to go for something, 5G,

1226
01:29:16.000 --> 01:29:19.000
you know, internet of things, all of that stuff,

1227
01:29:19.000 --> 01:29:23.000
they would aim for 98% market share, right?

1228
01:29:23.000 --> 01:29:28.000
And they get there. I mean, look at how they did electric cars.

1229
01:29:28.000 --> 01:29:32.000
And my question is very simple.

1230
01:29:32.000 --> 01:29:34.000
We've spoken so much about AI.

1231
01:29:34.000 --> 01:29:38.000
My question is, are the people of the West going to wait?

1232
01:29:38.000 --> 01:29:40.000
So this is an interesting conundrum,

1233
01:29:40.000 --> 01:29:43.000
because it sounds to me, like you're saying on one end,

1234
01:29:43.000 --> 01:29:47.000
if we don't join this AI race, then we're a third world country.

1235
01:29:47.000 --> 01:29:52.000
But on the other hand, if we continue this sort of thoughtless race

1236
01:29:52.000 --> 01:29:56.000
towards AGI, there's going to be catastrophe at some point.

1237
01:29:56.000 --> 01:29:58.000
And the answer is somewhere in the middle,

1238
01:29:58.000 --> 01:30:03.000
where basically you join the AI race for the good of your community.

1239
01:30:03.000 --> 01:30:06.000
So there is resignation on one side is like,

1240
01:30:06.000 --> 01:30:08.000
I'm not going to play this game at all.

1241
01:30:08.000 --> 01:30:11.000
There is offense on the other side,

1242
01:30:11.000 --> 01:30:14.000
where I'm playing this game to destroy everyone else.

1243
01:30:14.000 --> 01:30:16.000
And there is a balance in the middle,

1244
01:30:16.000 --> 01:30:19.000
where we say we're going to build an ethical form of AI

1245
01:30:19.000 --> 01:30:21.000
that's going to help our communities.

1246
01:30:21.000 --> 01:30:25.000
We're going to use this ultimate gift, this ultimate superpower,

1247
01:30:25.000 --> 01:30:28.000
superman landed on the planet,

1248
01:30:28.000 --> 01:30:30.000
raise it to help your community.

1249
01:30:30.000 --> 01:30:32.000
Is that wishful thinking to some degree,

1250
01:30:32.000 --> 01:30:35.000
when you think about the nature of sort of compact competition?

1251
01:30:35.000 --> 01:30:38.000
I don't know how to tell you otherwise.

1252
01:30:38.000 --> 01:30:41.000
I genuinely believe most,

1253
01:30:41.000 --> 01:30:46.000
I mean, here, I genuinely believe it's going to be very difficult to make that change.

1254
01:30:46.000 --> 01:30:49.000
I genuinely believe that it's going to be

1255
01:30:49.000 --> 01:30:52.000
that when the challenges of an AI dystopia hits us,

1256
01:30:52.000 --> 01:30:54.000
we're not going to be ready.

1257
01:30:54.000 --> 01:30:56.000
But I can't stop talking about it, Steven.

1258
01:30:56.000 --> 01:30:59.000
Do you understand what I stand with this?

1259
01:30:59.000 --> 01:31:01.000
I just am hoping,

1260
01:31:01.000 --> 01:31:03.000
and I tried so hard.

1261
01:31:03.000 --> 01:31:05.000
I spoke to the leadership.

1262
01:31:05.000 --> 01:31:08.000
Then I spoke like one of the things that my Atlantic productions

1263
01:31:08.000 --> 01:31:10.000
who helped me with the film,

1264
01:31:10.000 --> 01:31:12.000
you know, did for me,

1265
01:31:12.000 --> 01:31:15.000
which I have to say completely,

1266
01:31:15.000 --> 01:31:17.000
I am very appreciative of this.

1267
01:31:17.000 --> 01:31:19.000
As you know, through our conversations,

1268
01:31:19.000 --> 01:31:21.000
through our last few years,

1269
01:31:21.000 --> 01:31:23.000
that I tended to,

1270
01:31:23.000 --> 01:31:26.000
at a point in time, lose hope in the leadership,

1271
01:31:26.000 --> 01:31:31.000
and basically try to influence the public for ethical AI.

1272
01:31:31.000 --> 01:31:35.000
And my conversation was that the leadership at the time

1273
01:31:35.000 --> 01:31:37.000
was the technical leadership,

1274
01:31:37.000 --> 01:31:39.000
and that everyone was so caught up in the arms race,

1275
01:31:39.000 --> 01:31:41.000
that I wanted to teach the public

1276
01:31:41.000 --> 01:31:43.000
to help us build ethical AI.

1277
01:31:43.000 --> 01:31:45.000
And I continue to focus on the public,

1278
01:31:45.000 --> 01:31:46.000
that everyone of us,

1279
01:31:46.000 --> 01:31:48.000
but suddenly Atlantic goes,

1280
01:31:48.000 --> 01:31:49.000
no, no, hold on,

1281
01:31:49.000 --> 01:31:52.000
we should probably get you to meet the political leaders

1282
01:31:52.000 --> 01:31:53.000
everywhere in the world,

1283
01:31:53.000 --> 01:31:55.000
and hopefully give them a message that says,

1284
01:31:55.000 --> 01:31:56.000
hey, you know what,

1285
01:31:56.000 --> 01:31:58.000
you may actually make a difference

1286
01:31:58.000 --> 01:32:01.000
if you prioritize AI differently, right?

1287
01:32:01.000 --> 01:32:03.000
Do I believe this will happen?

1288
01:32:03.000 --> 01:32:06.000
Sadly no, but does that mean I should stop trying?

1289
01:32:06.000 --> 01:32:08.000
I cannot stop trying.

1290
01:32:08.000 --> 01:32:13.000
This has had probably the single biggest impact on my office.

1291
01:32:13.000 --> 01:32:15.000
Of all the products that I've tried,

1292
01:32:15.000 --> 01:32:18.000
that have given me productivity games or cognitive boosts,

1293
01:32:18.000 --> 01:32:21.000
I would say that exogenous ketones are,

1294
01:32:21.000 --> 01:32:24.000
in the top three, most pivotal things

1295
01:32:24.000 --> 01:32:26.000
that have given me a massive productivity game.

1296
01:32:26.000 --> 01:32:27.000
It's some Stanford graduates

1297
01:32:27.000 --> 01:32:31.000
that have been able to basically bottle up the effect you get

1298
01:32:31.000 --> 01:32:34.000
from being in a ketogenic diet in a small shot

1299
01:32:34.000 --> 01:32:37.000
that you can take that makes you feel incredibly focused

1300
01:32:37.000 --> 01:32:39.000
and gives your brain an incredible source of energy.

1301
01:32:39.000 --> 01:32:41.000
And the clinical studies that have been done

1302
01:32:41.000 --> 01:32:43.000
in exogenous ketones have absolutely blown my mind.

1303
01:32:43.000 --> 01:32:46.000
I reached out to them, I became a coroner in the company,

1304
01:32:46.000 --> 01:32:48.000
I became an investor in the company.

1305
01:32:48.000 --> 01:32:50.000
And so it's with great pride that I can tell you that this exists.

1306
01:32:50.000 --> 01:32:52.000
If you haven't tried these shots,

1307
01:32:52.000 --> 01:32:54.000
go to ketone.com slash Stephen

1308
01:32:54.000 --> 01:32:56.000
for 30% off your subscription order,

1309
01:32:56.000 --> 01:32:59.000
and you'll also get a free gift with your second shipment.

1310
01:32:59.000 --> 01:33:02.000
I still buy my ketone shots predominantly online,

1311
01:33:02.000 --> 01:33:04.000
but thankfully I can now grab them at Target

1312
01:33:04.000 --> 01:33:06.000
whenever I drive past them here in the United States as well

1313
01:33:06.000 --> 01:33:08.000
because we're now stocked in Target.

1314
01:33:08.000 --> 01:33:10.000
Where your first shot is completely free.

1315
01:33:10.000 --> 01:33:13.000
I've done almost 700 interviews

1316
01:33:13.000 --> 01:33:15.000
with some of the most interesting people in the world.

1317
01:33:15.000 --> 01:33:17.000
And one of the things you learn, which is unexpected,

1318
01:33:17.000 --> 01:33:20.000
is that vulnerability is the doorway to connection.

1319
01:33:20.000 --> 01:33:22.000
And after sitting here for two, three hours with a guest,

1320
01:33:22.000 --> 01:33:25.000
I feel a deep sense of connection to them.

1321
01:33:25.000 --> 01:33:27.000
And as they leave, what I get them to do

1322
01:33:27.000 --> 01:33:31.000
is to write a question in the diary of a CEO.

1323
01:33:31.000 --> 01:33:34.000
We've taken all of the questions from the diover CEO.

1324
01:33:34.000 --> 01:33:38.000
We have put the question here on this card

1325
01:33:38.000 --> 01:33:40.000
with the name of the person that wrote it.

1326
01:33:40.000 --> 01:33:42.000
So you can sit at home, as I do with my fiance,

1327
01:33:42.000 --> 01:33:45.000
and my colleagues at work, and other people in my life.

1328
01:33:45.000 --> 01:33:49.000
Whenever we get a minute, we play the diover CEO conversation cards.

1329
01:33:49.000 --> 01:33:51.000
And it is incredible what happens.

1330
01:33:51.000 --> 01:33:54.000
These are great if you're in a romantic relationship,

1331
01:33:54.000 --> 01:33:55.000
and you want to connect your partner more.

1332
01:33:55.000 --> 01:33:57.000
These are also great if you're in a team,

1333
01:33:57.000 --> 01:33:59.000
and you want to bond your team together.

1334
01:33:59.000 --> 01:34:01.000
And I have to say they're also great for families

1335
01:34:01.000 --> 01:34:03.000
that want to learn more about each other,

1336
01:34:03.000 --> 01:34:05.000
and that need a good excuse to spend some time

1337
01:34:05.000 --> 01:34:08.000
in a digital world in the analog environment,

1338
01:34:08.000 --> 01:34:10.000
connecting human to human.

1339
01:34:10.000 --> 01:34:12.000
It is remarkable what the right question

1340
01:34:12.000 --> 01:34:14.000
at the right time can do.

1341
01:34:14.000 --> 01:34:16.000
Go to the diary.com,

1342
01:34:16.000 --> 01:34:19.000
and you can get these conversation cards right now.

1343
01:34:19.000 --> 01:34:21.000
When I have these conversations about AI,

1344
01:34:21.000 --> 01:34:23.000
what I'm trying to do all the time is to pass out

1345
01:34:23.000 --> 01:34:26.000
what is like wishful thinking, and then what is reality.

1346
01:34:26.000 --> 01:34:28.000
And like, it's to understand reality.

1347
01:34:28.000 --> 01:34:30.000
You have to understand competition.

1348
01:34:30.000 --> 01:34:32.000
You have to understand human emotion.

1349
01:34:32.000 --> 01:34:34.000
You have to understand incentive structures.

1350
01:34:34.000 --> 01:34:36.000
And so you think about something like the United States

1351
01:34:36.000 --> 01:34:38.000
at the moment where you've got Donald Trump,

1352
01:34:38.000 --> 01:34:42.000
whose primary driving incentive is GDP growth,

1353
01:34:42.000 --> 01:34:45.000
economy growth, does that stock market go out, beat China?

1354
01:34:45.000 --> 01:34:48.000
So if you all agree that that's the core of his incentive structure,

1355
01:34:48.000 --> 01:34:50.000
then you've got President Chi over here,

1356
01:34:50.000 --> 01:34:53.000
whose incentive structure is probably control,

1357
01:34:53.000 --> 01:34:56.000
or independence, defense,

1358
01:34:56.000 --> 01:35:00.000
so that they need to make sure they do well on the weapon side.

1359
01:35:00.000 --> 01:35:02.000
And then you've got these other nations,

1360
01:35:02.000 --> 01:35:04.000
like the UK and Europe and these other places,

1361
01:35:04.000 --> 01:35:07.000
who kind of, it seems like we're a bit resigned to the fact

1362
01:35:07.000 --> 01:35:12.000
that we're not going to participate in the underlying models building,

1363
01:35:12.000 --> 01:35:14.000
because we just don't have it together.

1364
01:35:14.000 --> 01:35:16.000
We don't have the energy.

1365
01:35:16.000 --> 01:35:18.000
And you go in such a world,

1366
01:35:18.000 --> 01:35:20.000
you go ethical AI,

1367
01:35:20.000 --> 01:35:23.000
who's going to prioritize ethical AI?

1368
01:35:23.000 --> 01:35:27.000
I mean, anybody that does is might fall behind theoretically.

1369
01:35:27.000 --> 01:35:30.000
So I wonder where the fault in the thinking is here.

1370
01:35:31.000 --> 01:35:33.000
Like, how do we get to a point of ethical AI,

1371
01:35:33.000 --> 01:35:37.000
when the incentive structures are so clearly highly competitive,

1372
01:35:37.000 --> 01:35:43.000
and arguably a little bit short-term in their thinking?

1373
01:35:43.000 --> 01:35:45.000
Yeah.

1374
01:35:45.000 --> 01:35:46.000
So what are you saying?

1375
01:35:46.000 --> 01:35:48.000
We lie down and wait.

1376
01:35:48.000 --> 01:35:49.000
No, I just don't know that.

1377
01:35:49.000 --> 01:35:51.000
I just don't have an answer, honestly.

1378
01:35:51.000 --> 01:35:54.000
That's why we keep talking about it.

1379
01:35:54.000 --> 01:35:58.000
Yeah, and that's why I keep spending 14 hours a day trying to tell the world,

1380
01:35:58.000 --> 01:36:01.000
because some genius somewhere is going to find an answer.

1381
01:36:01.000 --> 01:36:03.000
But the way it's going.

1382
01:36:03.000 --> 01:36:05.000
Right now, I guess what we're pursuing is,

1383
01:36:05.000 --> 01:36:08.000
we're hoping that chat GPT and anthropoconies and Groken,

1384
01:36:08.000 --> 01:36:10.000
we're hoping that they just build ethical models

1385
01:36:10.000 --> 01:36:14.000
and we're hoping that social pressure forces them into making good decisions.

1386
01:36:14.000 --> 01:36:16.000
Correct.

1387
01:36:16.000 --> 01:36:19.000
We need to be able to vote with our usage.

1388
01:36:19.000 --> 01:36:23.000
So I think one of the biggest movements in AI since we started

1389
01:36:23.000 --> 01:36:27.000
was the idea that so many people switched away from chat GPT

1390
01:36:27.000 --> 01:36:29.000
that their model can target people.

1391
01:36:29.000 --> 01:36:30.000
Right?

1392
01:36:30.000 --> 01:36:32.000
So many people, I know, at least said.

1393
01:36:32.000 --> 01:36:33.000
Target people.

1394
01:36:33.000 --> 01:36:35.000
Everyone anthropic refused to have to serve.

1395
01:36:35.000 --> 01:36:36.000
Do you think people switched?

1396
01:36:36.000 --> 01:36:38.000
I think many have.

1397
01:36:38.000 --> 01:36:40.000
I think ones that are aware of it.

1398
01:36:40.000 --> 01:36:41.000
Right?

1399
01:36:41.000 --> 01:36:44.000
And I think they did because the cost of switching is really,

1400
01:36:44.000 --> 01:36:47.000
honestly, anthropic is better, if you think about it.

1401
01:36:47.000 --> 01:36:48.000
Yeah, I think it's better.

1402
01:36:48.000 --> 01:36:49.000
Yeah.

1403
01:36:49.000 --> 01:36:53.000
But the idea is, if people don't switch for those ethical reasons,

1404
01:36:53.000 --> 01:36:57.000
so you know, every one of my books has this dedication at the beginning.

1405
01:36:57.000 --> 01:37:00.000
It used to be the gravity of the button means nothing to those at peace

1406
01:37:00.000 --> 01:37:03.000
when I wrote in memory of Ali at the beginning.

1407
01:37:03.000 --> 01:37:06.000
My next book, the Alive,

1408
01:37:06.000 --> 01:37:09.000
do you remember a song called the,

1409
01:37:09.000 --> 01:37:13.000
they were called the Manic Street Preachers?

1410
01:37:13.000 --> 01:37:14.000
Oh, yeah.

1411
01:37:14.000 --> 01:37:17.000
If you tolerate this, then your children will be next.

1412
01:37:17.000 --> 01:37:21.000
I genuinely believe that what the world needs to wake up to

1413
01:37:21.000 --> 01:37:24.000
is if you tolerate this, then your children will be next.

1414
01:37:24.000 --> 01:37:29.000
If you continue to resign, if you continue to say I'm not going to try,

1415
01:37:29.000 --> 01:37:34.000
this world is going to change in a way that is completely not for in your favor.

1416
01:37:34.000 --> 01:37:35.000
Try what?

1417
01:37:35.000 --> 01:37:39.000
Try to stand up and say something needs to change.

1418
01:37:39.000 --> 01:37:40.000
Okay.

1419
01:37:40.000 --> 01:37:42.000
We can say something needs to change,

1420
01:37:42.000 --> 01:37:44.000
but we can't say what that thing is.

1421
01:37:44.000 --> 01:37:47.000
What needs to change is governments need to serve.

1422
01:37:48.000 --> 01:37:50.000
There are people, not their interests,

1423
01:37:50.000 --> 01:37:58.000
and corporates need to work for the benefit of their societies before their shareholders.

1424
01:37:58.000 --> 01:38:07.000
This run that we had with capitalism so far has benefited the world tremendously.

1425
01:38:07.000 --> 01:38:11.000
Who's economy, do you think, is going to be in a better place for the minerals?

1426
01:38:11.000 --> 01:38:13.000
To say the UK in the United States?

1427
01:38:13.000 --> 01:38:15.000
The UK is gone.

1428
01:38:15.000 --> 01:38:18.000
It's gone because what in part because it didn't compete?

1429
01:38:18.000 --> 01:38:26.000
Because you're an older bureaucracy that is burning down by so much barriers

1430
01:38:26.000 --> 01:38:29.000
in the process of building anything.

1431
01:38:29.000 --> 01:38:37.000
Because the US economy in the past welcomed people like me to go and live in California

1432
01:38:37.000 --> 01:38:39.000
and build amazing shit.

1433
01:38:39.000 --> 01:38:41.000
That is no longer the case.

1434
01:38:41.000 --> 01:38:43.000
So who's going to win in my view?

1435
01:38:43.000 --> 01:38:45.000
It's definitely China.

1436
01:38:45.000 --> 01:38:48.000
By the way, you asked for the middle class.

1437
01:38:48.000 --> 01:38:54.000
China made decisions recently that forced businesses not to lay people off in replacement

1438
01:38:54.000 --> 01:38:56.000
to be replacing them with AI.

1439
01:38:56.000 --> 01:38:58.000
Would the West do that?

1440
01:38:58.000 --> 01:39:01.000
The capitalist West would never do that.

1441
01:39:01.000 --> 01:39:02.000
I don't know the answer.

1442
01:39:02.000 --> 01:39:04.000
I don't know the answer I'm responding to.

1443
01:39:04.000 --> 01:39:06.000
You can see the conundrum I found myself in,

1444
01:39:06.000 --> 01:39:15.000
which is a state like a country like the UK is in your words gone because it didn't compete.

1445
01:39:15.000 --> 01:39:18.000
It didn't allow people to be highly entrepreneurial.

1446
01:39:18.000 --> 01:39:23.000
It didn't empower entrepreneurialism, innovation, ingenuity.

1447
01:39:23.000 --> 01:39:27.000
In some way, it stood in their way, in some way.

1448
01:39:27.000 --> 01:39:30.000
Now that could be incentives, it could be culture, it could be whatever.

1449
01:39:30.000 --> 01:39:31.000
I know what you're going with this.

1450
01:39:31.000 --> 01:39:44.000
The US didn't say they have a economy, which is arguably more productive and future-proofed than ours.

1451
01:39:44.000 --> 01:39:51.000
By way of that, they are also more advanced in artificial intelligence.

1452
01:39:51.000 --> 01:39:52.000
And we are gone.

1453
01:39:52.000 --> 01:39:56.000
So the remedy for a country like us would be there for to compete.

1454
01:39:56.000 --> 01:39:59.000
To compete, let the reins off, let entrepreneurs in ingenuity.

1455
01:39:59.000 --> 01:40:01.000
But then we're saying that's dangerous.

1456
01:40:01.000 --> 01:40:07.000
And your conundrum in that is that you're assuming that entrepreneurship by definition is malicious.

1457
01:40:07.000 --> 01:40:10.000
No, I'm just saying that there's a bit of a paradox.

1458
01:40:10.000 --> 01:40:14.000
Like you damned if you don't, you damned if you do.

1459
01:40:14.000 --> 01:40:18.000
But you're not damned if you build things for the people, not for the capitalist.

1460
01:40:18.000 --> 01:40:21.000
This is an ideological debate.

1461
01:40:21.000 --> 01:40:25.000
I pray and hope that that is plausible.

1462
01:40:25.000 --> 01:40:34.000
But I'm worried that in a competitive market, whoever's optimizing for, I don't know, you can name it, retention.

1463
01:40:34.000 --> 01:40:40.000
If we built two AI's, right, I'm going to call them the mogul.ai and I'm going to call the other one evil AI.

1464
01:40:40.000 --> 01:40:43.000
The evil AI, it's programmed to retain you.

1465
01:40:43.000 --> 01:40:44.000
It's sycophantic.

1466
01:40:44.000 --> 01:40:46.000
It says what you need to hear.

1467
01:40:46.000 --> 01:40:48.000
It's super smart.

1468
01:40:48.000 --> 01:40:55.000
And because of that, even though it's not trading in your best interests, it's retaining you more.

1469
01:40:55.000 --> 01:40:56.000
You're using it more often.

1470
01:40:56.000 --> 01:40:57.000
It's programmed for that.

1471
01:40:57.000 --> 01:40:59.000
Kind of like the social networks are.

1472
01:40:59.000 --> 01:41:03.000
They're all just trying to like dopamine your brain into oblivion.

1473
01:41:03.000 --> 01:41:06.000
Then there's the Moai AI that tells you to log off you've had enough.

1474
01:41:06.000 --> 01:41:08.000
It thinks about your mental you well being.

1475
01:41:08.000 --> 01:41:14.000
Because it's less attentive and less engaging, theoretically, it might be less successful as a commercial product.

1476
01:41:14.000 --> 01:41:15.000
Think about social networks.

1477
01:41:15.000 --> 01:41:17.000
The ones that are least attentive.

1478
01:41:17.000 --> 01:41:20.000
The ones that actually won't destroy your brain with dopamine.

1479
01:41:20.000 --> 01:41:22.000
The ones that remove the retweet button.

1480
01:41:22.000 --> 01:41:27.000
The ones that don't have slot machine like videos, they don't survive.

1481
01:41:27.000 --> 01:41:33.000
I share this with you and my point of view is that to summarize your challenge here,

1482
01:41:33.000 --> 01:41:40.000
you're basically saying that it's easier to become successful if you don't follow ethical rules.

1483
01:41:40.000 --> 01:41:46.000
I'm asking the question, if you build an AI that is just purely focused on ethical,

1484
01:41:46.000 --> 01:41:51.000
would it be as engaging and have the same usage as one built with the reins off?

1485
01:41:51.000 --> 01:41:54.000
No, it won't.

1486
01:41:54.000 --> 01:42:00.000
But that's the problem humanity needs to solve if we were to survive.

1487
01:42:00.000 --> 01:42:08.000
Example, I worked in a company called Google that basically at a point in time decided that ads will be effective.

1488
01:42:09.000 --> 01:42:14.000
The ad industry prior to Google was 50% of your ad budget doesn't work.

1489
01:42:14.000 --> 01:42:17.000
We just don't know which 50% remember that.

1490
01:42:17.000 --> 01:42:27.000
And Google came in and struggled from 1998 until 2004 when they started to turn probably plausible revenues

1491
01:42:27.000 --> 01:42:31.000
as a result of saying we're going to run a Dutch auction and we're going to give you paper click

1492
01:42:31.000 --> 01:42:34.000
and we're going to show you results for your ads.

1493
01:42:34.000 --> 01:42:43.000
They found a way for ethics to actually get your money to be effective rather than just take your money

1494
01:42:43.000 --> 01:42:45.000
and say 50% doesn't work.

1495
01:42:45.000 --> 01:42:48.000
They found a way to make that their success criteria.

1496
01:42:48.000 --> 01:42:50.000
There must be a way for us.

1497
01:42:50.000 --> 01:42:54.000
I don't know what it is if I knew I would be building it instead of sitting with you.

1498
01:42:54.000 --> 01:43:01.000
But there must be a way for us to marry the success of humanity with the success of the entrepreneur.

1499
01:43:02.000 --> 01:43:06.000
And that way is not found in the old ways of doing things.

1500
01:43:06.000 --> 01:43:08.000
I've got an idea.

1501
01:43:08.000 --> 01:43:14.000
Maybe, you know, Claude 4.8 came out yesterday, which is one of the big AI models, the new one.

1502
01:43:14.000 --> 01:43:18.000
And when they release the models, they show these like graphs of the benchmarks.

1503
01:43:18.000 --> 01:43:20.000
What they mean by this is what it's capable of.

1504
01:43:20.000 --> 01:43:24.000
They show how it performed in maths, in science, in writing, reasoning, etc.

1505
01:43:24.000 --> 01:43:26.000
It is all marginal at this point.

1506
01:43:26.000 --> 01:43:31.000
So my idea is, could there be an ethical benchmark that all these models have to pass

1507
01:43:31.000 --> 01:43:35.000
before these large companies can legally deploy them?

1508
01:43:35.000 --> 01:43:40.000
And I was thinking about like, again, every idea has unintended consequences, which I haven't thought through.

1509
01:43:40.000 --> 01:43:44.000
But would it be very interesting that when they release these models, they also release the ethical benchmarks?

1510
01:43:44.000 --> 01:43:46.000
We tried to get it to X.

1511
01:43:46.000 --> 01:43:49.000
We tried to get it to Y. We tried to get it to Z.

1512
01:43:49.000 --> 01:43:51.000
And here's how it performed against the ethical benchmarks.

1513
01:43:51.000 --> 01:43:55.000
Some kind of standard for governments to say, you're not allowed to release a new model,

1514
01:43:55.000 --> 01:43:58.000
unless it passes independent tested ethical benchmarks.

1515
01:43:58.000 --> 01:43:59.000
Beautiful.

1516
01:43:59.000 --> 01:44:01.000
That would absolutely work.

1517
01:44:01.000 --> 01:44:04.000
But notice, by the way, those are out there already.

1518
01:44:04.000 --> 01:44:08.000
But in a very interesting way, you listen to them as her service.

1519
01:44:08.000 --> 01:44:18.000
And how much she invests heavily in building alpha-fold or building so many scientific applications of AI.

1520
01:44:18.000 --> 01:44:21.000
And you go like, this guy cares about science.

1521
01:44:21.000 --> 01:44:24.000
I can't prove that.

1522
01:44:24.000 --> 01:44:28.000
But I've met as a deficit couple of times.

1523
01:44:28.000 --> 01:44:31.000
I know genuinely that he is an ethical person.

1524
01:44:31.000 --> 01:44:38.000
But the typical person will probably say, but at least it seems that they're doing things for free to serve science.

1525
01:44:38.000 --> 01:44:47.000
You look at anthropic, and they refuse to use their model to allow the US government to target and to spy on people.

1526
01:44:47.000 --> 01:44:54.000
And then you see OpenAI accepting a $500 million deal that absolutely does that.

1527
01:44:54.000 --> 01:45:08.000
It is about time that every person in the world says, in that case, I am no longer going to use OpenAI until they show me another evidence that they are actually ethical in their behavior.

1528
01:45:08.000 --> 01:45:13.000
And this is the decision that you and I can do.

1529
01:45:14.000 --> 01:45:16.000
But that's my task.

1530
01:45:16.000 --> 01:45:23.000
My task and yours is to keep telling them, people, please, please understand that if you tolerate this, then your children will be next.

1531
01:45:23.000 --> 01:45:36.000
Please understand that if you don't start to take an ethical stand on your own future, your future will be handed over to another oligarch, just like your past, was handed over to social media oligarchs.

1532
01:45:36.000 --> 01:45:47.000
One would say, being at the commencement speech is a good example of how public awareness can have a real impact on this trajectory.

1533
01:45:47.000 --> 01:45:59.000
But I still think at the end of the day, if you think about things like smartphone usage in schools, at the end of the day, it does come down to government intervention and saying,

1534
01:45:59.000 --> 01:46:02.000
you know what, we're going to ban 14-year-olds from scrolling TikTok.

1535
01:46:02.000 --> 01:46:06.000
And that's in part because people spoke louder and louder and louder and they went on podcasts.

1536
01:46:06.000 --> 01:46:13.000
Jonathan Height, who wrote the book about the anxious generation, started a conversation and that conversation led to legislation.

1537
01:46:13.000 --> 01:46:17.000
I still think it ends in some kind of constraint legally.

1538
01:46:17.000 --> 01:46:24.000
I hope, but I will openly tell you most of the tech oligarchs are more powerful than your government.

1539
01:46:24.000 --> 01:46:32.000
Is there any precedence in history where this kind of change happened without government intervention?

1540
01:46:32.000 --> 01:46:44.000
So I was thinking about things like climate change, even you could say smoking, all these kinds of things have had to be like taxes and...

1541
01:46:44.000 --> 01:46:48.000
I'm with you. If governments intervene, we wouldn't have a problem.

1542
01:46:48.000 --> 01:46:53.000
Governments won't intervene because governments are owned by the oligarchs.

1543
01:46:53.000 --> 01:46:58.000
Right? So my question for everyone listening to us is, are you going to intervene?

1544
01:46:58.000 --> 01:46:59.000
What counts on your trajectory?

1545
01:46:59.000 --> 01:47:05.000
If that's what you can do, it's fine. If not, then go ahead and start a startup that does something.

1546
01:47:05.000 --> 01:47:09.000
If that's not within your capability, then send a message to your congressman.

1547
01:47:09.000 --> 01:47:18.000
If that's not within your capability, then say something ethical online so that the world understands a position that needs to be opening the eyes.

1548
01:47:18.000 --> 01:47:23.000
If that doesn't work for you, at least don't engage in stuff that is negative that you don't know enough about.

1549
01:47:23.000 --> 01:47:31.000
There are so many little actions. If humanity starts to move in the direction of one saying ethics matter, not just profit.

1550
01:47:31.000 --> 01:47:37.000
And two saying, I'm not going to participate in something that's unethical, just because I believe I want to...

1551
01:47:37.000 --> 01:47:40.000
I feel like it's right now.

1552
01:47:40.000 --> 01:47:49.000
If I tell you the number of things I took out of my life, just to try and affect a tiny bit of change of revenues that go to bullets.

1553
01:47:49.000 --> 01:47:51.000
If I tell you the number of things.

1554
01:47:51.000 --> 01:47:56.000
I think there's two central concerns I've always had, which is I do feel that there's going to be significant job disruption.

1555
01:47:56.000 --> 01:48:02.000
And I don't think society is prepared for it yet, and I don't know what that preparation looks like, but I think we should start thinking about it.

1556
01:48:02.000 --> 01:48:05.000
I share this with you.

1557
01:48:05.000 --> 01:48:11.000
I genuinely believe that if we continue on where we are, there's no hope.

1558
01:48:11.000 --> 01:48:20.000
In the trajectory of where humanity has become, so distracted, so resigned to inaction.

1559
01:48:20.000 --> 01:48:31.000
So disconnected from their own rights of freedom of expression and engagement and so on, I think we have no hope.

1560
01:48:31.000 --> 01:48:35.000
Do we want to stay there? That's a question that I'm asking our listeners.

1561
01:48:35.000 --> 01:48:42.000
And I'm not saying be violent or get up or be angry or whatever, I'm just saying take one little action.

1562
01:48:42.000 --> 01:48:45.000
Ask yourself, please write it in the comments, one little action.

1563
01:48:45.000 --> 01:48:50.000
One little action that you're going to do today that's going to make the world a little better tomorrow.

1564
01:48:50.000 --> 01:49:00.000
And don't give up on humanity, Stephen. I'm not saying you do, but I'm saying we are going through such a difficult time in humanity's history

1565
01:49:00.000 --> 01:49:05.000
that for the very first time ever, we have to do something about it.

1566
01:49:05.000 --> 01:49:11.000
I don't want my daughter to be at the receiving end of what happened to my son.

1567
01:49:11.000 --> 01:49:19.000
I don't, I genuinely, I lost Ali, I don't want to lose, and the world we're building is going to be very difficult for us, Stephen.

1568
01:49:19.000 --> 01:49:24.000
And I cannot go to sleep at night without trying something every day.

1569
01:49:24.000 --> 01:49:29.000
And I genuinely don't understand how humanity is not as missing that point.

1570
01:49:29.000 --> 01:49:33.000
Mainly, I think, because they're uninformed. Now we're informing them.

1571
01:49:33.000 --> 01:49:41.000
The only thing that will save humanity going forward is that this superpower called intelligence is used for ethical reasons.

1572
01:49:41.000 --> 01:49:46.000
Is that the corruption that's leading us to where we are today stops?

1573
01:49:46.000 --> 01:49:51.000
You've got more envelopes there. What's your next envelope?

1574
01:49:51.000 --> 01:49:52.000
The third one.

1575
01:49:52.000 --> 01:49:56.000
Number two is job losses. Number three was labored, you know, basically.

1576
01:49:56.000 --> 01:49:58.000
Same thing. Robots will replace manual labor by twenty thirty.

1577
01:49:58.000 --> 01:50:00.000
We'll start to replace manual labor.

1578
01:50:00.000 --> 01:50:01.000
Okay. Yeah.

1579
01:50:01.000 --> 01:50:08.000
So, you know, you will, you will have more and more manual jobs given to robots. What's this one?

1580
01:50:08.000 --> 01:50:14.000
Oh, this is absolute. Do you think otherwise the world's first trillionaire before?

1581
01:50:14.000 --> 01:50:16.000
Oh, but yeah, probably well before then.

1582
01:50:16.000 --> 01:50:17.000
Well, before 2030.

1583
01:50:17.000 --> 01:50:22.000
I think, you know, Elon's about to a IPR space X, which is likely to make him a trillionaire.

1584
01:50:22.000 --> 01:50:30.000
Yeah, I think the concentration of power that comes with that is quite drastic when you really think about it.

1585
01:50:30.000 --> 01:50:33.000
And that's 2030 is just a few years away.

1586
01:50:33.000 --> 01:50:41.000
I think the team wrote this wrong artificial super intelligence will arrive in 2032, 20 to 2035.

1587
01:50:41.000 --> 01:50:45.000
I think artificial super intelligence will arrive the minute AGI happens.

1588
01:50:45.000 --> 01:50:52.000
So, it doesn't really matter if AI is a billion times smarter than you or just twice as smart as you.

1589
01:50:52.000 --> 01:50:59.000
Once we cross beyond AGI, ASI is just very, very soon.

1590
01:50:59.000 --> 01:51:07.000
And yeah, I think we'll overcome that when we get to the force inevitable.

1591
01:51:07.000 --> 01:51:13.000
When AI is in charge of everything, I genuinely believe that we will end up in a utopia of abundance.

1592
01:51:14.000 --> 01:51:22.000
I genuinely believe that, again, physics, mathematics and biology will tell you that super intelligence is benign.

1593
01:51:22.000 --> 01:51:32.000
And that we will eventually end up in a good place, not because humanity has done much to get us there, not because our leaders have suddenly turned ethical.

1594
01:51:32.000 --> 01:51:41.000
But because our unethical leaders have gone out of the equation and were replaced with a super efficient minimum energy principle

1595
01:51:41.000 --> 01:51:44.000
that doesn't see value in anything that's destructive.

1596
01:51:44.000 --> 01:51:46.000
So the future is going to be great.

1597
01:51:46.000 --> 01:51:50.000
Those who make it to 2038 will enjoy it, yeah.

1598
01:51:50.000 --> 01:51:52.000
Those who make it.

1599
01:51:52.000 --> 01:51:53.000
For sure.

1600
01:51:53.000 --> 01:51:59.000
I mean, World War II didn't destroy the world, but ask those who went through it.

1601
01:51:59.000 --> 01:52:09.000
It's just an interesting idea that actually it's just you're forecasting basically like a decade of turmoil of dystopia, absolute dystopia.

1602
01:52:09.000 --> 01:52:16.000
When you say absolute dystopia, just so I'm clear in my mind, that absolute dystopia you're forecasting over a decade is about war.

1603
01:52:16.000 --> 01:52:18.000
Or economics jobs.

1604
01:52:18.000 --> 01:52:20.000
Economics, it's about jobs.

1605
01:52:20.000 --> 01:52:22.000
It's also about surveillance and control.

1606
01:52:22.000 --> 01:52:24.000
It's also about digital currencies.

1607
01:52:24.000 --> 01:52:27.000
It's also about human connection.

1608
01:52:27.000 --> 01:52:30.000
It's also about concentration of power.

1609
01:52:30.000 --> 01:52:34.000
It's a magnification of everything we've built so far.

1610
01:52:34.000 --> 01:52:40.000
And just again, aren't people with some tools to survive that dystopia for a decade?

1611
01:52:40.000 --> 01:52:45.000
You know, we talked a little bit about focusing on human jobs, the more human jobs.

1612
01:52:45.000 --> 01:52:46.000
Yeah, learn AI.

1613
01:52:46.000 --> 01:52:49.000
So AI is not the enemy, okay?

1614
01:52:49.000 --> 01:52:54.000
By definition, the better you are at using an AI to do your job.

1615
01:52:54.000 --> 01:52:55.000
Yeah.

1616
01:52:55.000 --> 01:52:58.000
The more likely you are to be successful, right?

1617
01:52:58.000 --> 01:53:03.000
Number two is prepare for a hybrid world where AI and humans work together.

1618
01:53:03.000 --> 01:53:04.000
How do you prepare?

1619
01:53:04.000 --> 01:53:08.000
You basically understand agents and how agents work.

1620
01:53:08.000 --> 01:53:20.000
You understand how a hyper-efficient approach to things may not require you to be very long meetings and very long.

1621
01:53:20.000 --> 01:53:27.000
So there is, if you lived in California, you would know that our, you know, the way we ran businesses was a lot more efficient.

1622
01:53:27.000 --> 01:53:31.000
We sometimes had a seven-minute meeting, right?

1623
01:53:31.000 --> 01:53:36.000
So the habits of an AI are much more efficient than the habits of humans.

1624
01:53:36.000 --> 01:53:38.000
So you've learned AI again?

1625
01:53:38.000 --> 01:53:43.000
Learn how to interact with AI and welcome AI into your hybrid world of work.

1626
01:53:43.000 --> 01:53:48.000
I think you need to, of course, double down on human skills.

1627
01:53:48.000 --> 01:53:53.000
I think that's a must to succeed in this world.

1628
01:53:53.000 --> 01:54:03.000
I think we need to, one of my most interesting views on the near future is how AI is going to be used to

1629
01:54:03.000 --> 01:54:15.000
disrupt, not disrupt blur facts and how we need to become much more interested in debugging what we're told using AI, by the way.

1630
01:54:15.000 --> 01:54:22.000
The part of that I have to say is you have to learn to use AI again, not as a lazy person.

1631
01:54:22.000 --> 01:54:25.000
So don't have them do things for you.

1632
01:54:25.000 --> 01:54:27.000
Have them make you smarter.

1633
01:54:27.000 --> 01:54:39.000
So instead of trying to get the same task done with one prompt, trying to get a much more interesting and demanding and intelligent task done with more work.

1634
01:54:39.000 --> 01:54:42.000
So I've got two things so far, which is basically like leaning to AI.

1635
01:54:42.000 --> 01:54:48.000
And the second one is like leaning to the human collection and leaning to the truth.

1636
01:54:48.000 --> 01:54:51.000
Don't be fooled by the hype.

1637
01:54:51.000 --> 01:54:55.000
You know, try to be more informed, I think.

1638
01:54:55.000 --> 01:55:02.000
And then finally, Essex, if you want to, I know it sounds...

1639
01:55:02.000 --> 01:55:10.000
The world we live in sounds as if the only way to win is to compete in capitalism.

1640
01:55:10.000 --> 01:55:12.000
That's not the world I live then.

1641
01:55:12.000 --> 01:55:17.000
The world I lived in, especially in my Google years, was solve a major problem.

1642
01:55:17.000 --> 01:55:20.000
And when you do, you'll end up making a lot of money.

1643
01:55:20.000 --> 01:55:22.000
Are you optimistic?

1644
01:55:22.000 --> 01:55:29.000
I am optimistic about the future. I'm very optimistic about the future. I'm not optimistic about the present.

1645
01:55:29.000 --> 01:55:31.000
You're not optimistic about the next decade.

1646
01:55:31.000 --> 01:55:35.000
Yeah, I'm not optimistic about the next year, to be honest.

1647
01:55:35.000 --> 01:55:36.000
The next year?

1648
01:55:36.000 --> 01:55:37.000
For sure.

1649
01:55:37.000 --> 01:55:39.000
Why the next year?

1650
01:55:39.000 --> 01:55:41.000
Oh, come on, Stephen.

1651
01:55:41.000 --> 01:55:45.000
You don't want me to say it out loud, we're ruled by maniacs.

1652
01:55:45.000 --> 01:55:49.000
Decisions are being made for the absolute, strong reasons.

1653
01:55:49.000 --> 01:55:53.000
Very interesting time.

1654
01:55:53.000 --> 01:55:57.000
Very interesting time we find ourselves in.

1655
01:55:57.000 --> 01:56:01.000
Honestly, if you're a video gamer, this is the best part of the game.

1656
01:56:01.000 --> 01:56:05.000
It is a very, very, very...

1657
01:56:05.000 --> 01:56:07.000
What's the word?

1658
01:56:07.000 --> 01:56:14.000
It's the ultimate matrix of complexity that I have ever encountered in my life.

1659
01:56:14.000 --> 01:56:17.000
Yeah, that's an amp's description of how it feels.

1660
01:56:17.000 --> 01:56:19.000
Very complex. I think they're moving very quickly.

1661
01:56:19.000 --> 01:56:21.000
I'm moving very quickly.

1662
01:56:21.000 --> 01:56:25.000
So from one side, you really need a lot of brain resources to crunch all that's happening,

1663
01:56:25.000 --> 01:56:28.000
but you wake up tomorrow and it's changed.

1664
01:56:28.000 --> 01:56:32.000
Well, the first time we spoke, we spoke about happiness.

1665
01:56:32.000 --> 01:56:35.000
After this conversation, Stephen.

1666
01:56:35.000 --> 01:56:38.000
I was just wondering, you know, in a...

1667
01:56:38.000 --> 01:56:41.000
You wrote a book about happiness, which is like many books about happiness.

1668
01:56:41.000 --> 01:56:45.000
But I mean, the soul for happiness, the book that I'm referring to here is...

1669
01:56:45.000 --> 01:56:47.000
Engineering, your path to joy. What a fantastic book.

1670
01:56:47.000 --> 01:56:50.000
I quote this book all the time, all around the world.

1671
01:56:50.000 --> 01:56:58.000
And I'm wondering if any of the principles that you wrote in this book about how to live a happy life

1672
01:56:58.000 --> 01:57:03.000
are more important now in the world that we live in, than maybe when you wrote this book.

1673
01:57:03.000 --> 01:57:05.000
Oh, for sure.

1674
01:57:05.000 --> 01:57:10.000
Honestly, if I wasn't living by this, I would have left this world a long time ago

1675
01:57:10.000 --> 01:57:12.000
and went to an island somewhere.

1676
01:57:12.000 --> 01:57:17.000
You see the interesting side of happiness is that it's not dopamine driven.

1677
01:57:17.000 --> 01:57:19.000
It's serotonin driven, right?

1678
01:57:19.000 --> 01:57:24.000
So my definition of happiness is I'm okay with this world as it is.

1679
01:57:24.000 --> 01:57:29.000
I can affect it. I can change it. I can engage with it. I can try to make it better.

1680
01:57:29.000 --> 01:57:34.000
I don't have to accept it, but I'm okay with it.

1681
01:57:34.000 --> 01:57:41.000
My starting point is a bit stoic if you want. My starting point is I accept this.

1682
01:57:41.000 --> 01:57:44.000
This is my reality and now I can start the work.

1683
01:57:44.000 --> 01:57:48.000
It's very different than anyone that basically looks at the world and says,

1684
01:57:48.000 --> 01:57:51.000
oh, this is a horrible world. I don't want to be part of it.

1685
01:57:51.000 --> 01:57:53.000
I don't want to be engaged with it and so on and so forth.

1686
01:57:53.000 --> 01:58:01.000
And I genuinely have never been calmer about that chaos.

1687
01:58:01.000 --> 01:58:08.000
It's quite interesting. My wonderful ex actually really, really helped me with that.

1688
01:58:08.000 --> 01:58:12.000
There was a point in time where we were having dinner and I poured out crying

1689
01:58:12.000 --> 01:58:16.000
from the sense of responsibility I had for the world.

1690
01:58:16.000 --> 01:58:21.000
And she looked at me so kindly, so gently and said, hold on.

1691
01:58:21.000 --> 01:58:28.000
I see that you're trying, but you can't actually believe that you're responsible for this.

1692
01:58:28.000 --> 01:58:34.000
And I think that completely flipped my mind because in a very interesting way I was thinking

1693
01:58:34.000 --> 01:58:39.000
that all that went wrong in technologies because of me, right?

1694
01:58:39.000 --> 01:58:40.000
Why?

1695
01:58:40.000 --> 01:58:44.000
Because I contributed to building this because I mean, when Jeffrey and I,

1696
01:58:44.000 --> 01:58:47.000
one of my favorite moments when we were filming chasing Utopia,

1697
01:58:47.000 --> 01:58:53.000
is that Jeffrey is very big for all of us. We really think the world of him.

1698
01:58:53.000 --> 01:58:58.000
And I was telling him just as a sort of like an older mentor if you want.

1699
01:58:58.000 --> 01:59:02.000
I was like, Jeffrey, do you regret doing this?

1700
01:59:02.000 --> 01:59:07.000
I genuinely believed when we were building those things that we were going to make the world better.

1701
01:59:07.000 --> 01:59:08.000
AI.

1702
01:59:08.000 --> 01:59:12.000
Yeah. And he said, well, yeah, I too was naive.

1703
01:59:12.000 --> 01:59:16.000
I thought that we, I didn't think he said.

1704
01:59:16.000 --> 01:59:21.000
I didn't think we will get there so quickly before we figured out the alignment problem.

1705
01:59:21.000 --> 01:59:23.000
The alignment problem.

1706
01:59:23.000 --> 01:59:27.000
That AI has our best interest in mind.

1707
01:59:27.000 --> 01:59:30.000
And I think all of us were faced with that.

1708
01:59:30.000 --> 01:59:35.000
All of us were faced with that idea of building the best thing ever for humanity.

1709
01:59:35.000 --> 01:59:40.000
And then suddenly you realize, oh my god, in the wrong hands, it's the worst thing ever for humanity.

1710
01:59:40.000 --> 01:59:44.000
And I, you know, I have to say I came to terms with this.

1711
01:59:44.000 --> 01:59:46.000
I've been asked this question.

1712
01:59:46.000 --> 01:59:48.000
I get asked it all this time, so.

1713
01:59:48.000 --> 01:59:50.000
I don't know why so many people are asking.

1714
01:59:50.000 --> 01:59:52.000
Am I going to go over this?

1715
01:59:52.000 --> 01:59:53.000
I'm going to go over this.

1716
01:59:53.000 --> 01:59:55.000
I'm going to go over this.

1717
01:59:55.000 --> 01:59:56.000
I'm going to go over this.

1718
01:59:56.000 --> 01:59:57.000
I'm going to go over this.

1719
01:59:57.000 --> 01:59:58.000
I'm going to go over this.

1720
01:59:58.000 --> 01:59:59.000
I'm going to go over this.

1721
01:59:59.000 --> 02:00:00.000
I'm going to go over this.

1722
02:00:00.000 --> 02:00:01.000
I'm going to go over this.

1723
02:00:01.000 --> 02:00:02.000
I'm going to go over this.

1724
02:00:02.000 --> 02:00:03.000
I'm going to go over this.

1725
02:00:03.000 --> 02:00:04.000
I'm going to go over this.

1726
02:00:04.000 --> 02:00:05.000
I'm going to go over this.

1727
02:00:05.000 --> 02:00:06.000
I'm going to go over this.

1728
02:00:06.000 --> 02:00:07.000
I'm going to go over this.

1729
02:00:07.000 --> 02:00:08.000
I'm going to go over this.

1730
02:00:08.000 --> 02:00:09.000
I'm going to go over this.

1731
02:00:09.000 --> 02:00:10.000
I'm going to go over this.

1732
02:00:10.000 --> 02:00:13.000
I don't know why so many people are asking.

1733
02:00:13.000 --> 02:00:15.000
Am I going to go over this anytime soon?

1734
02:00:15.000 --> 02:00:18.000
I don't know why so many people are asking me that of course.

1735
02:00:18.000 --> 02:00:21.000
You see, legacy is a…

1736
02:00:23.000 --> 02:00:27.000
I mean, what would I care if I have a legacy if I'm dead?

1737
02:00:27.000 --> 02:00:30.000
Like, why does that even make any difference?

1738
02:00:30.000 --> 02:00:32.000
Here's an interesting thought for everyone.

1739
02:00:32.000 --> 02:00:38.000
If Karma is real, and I genuinely believe it is,

1740
02:00:38.000 --> 02:00:47.000
And if we're not just physical beings that were physical and spiritual, then I'd rather keep all of my karma for my spiritual side.

1741
02:00:49.000 --> 02:00:50.000
What does that mean?

1742
02:00:50.000 --> 02:00:53.000
I don't want anyone to remember anything I ever did.

1743
02:00:54.000 --> 02:01:00.000
I just want to leave a positive impact on the world and take all of that as karma for my next journey.

1744
02:01:01.000 --> 02:01:02.000
Thank you.