#160 Exploring the Uncharted Territory of AI with Garik Tate

1
00:00:00,900 --> 00:00:06,600
From the heart of Dubai where
tomorrow is being built today to

2
00:00:06,600 --> 00:00:12,600
the world, welcome to the CTO
show with limit here.

3
00:00:12,700 --> 00:00:17,600
We redefined technology and
reimagine possibilities with

4
00:00:17,600 --> 00:00:23,300
mehmed delve into the riveting
Realms of AI cybersecurity and

5
00:00:23,300 --> 00:00:28,900
digital technology experience
the thrilling highs and lows of

6
00:00:28,900 --> 00:00:31,700
startups.
Immerse yourself in the spirit

7
00:00:31,700 --> 00:00:34,400
of Entrepreneurship and
witnessed the future of

8
00:00:34,400 --> 00:00:36,700
business.
Innovation, being written in

9
00:00:36,700 --> 00:00:40,700
real time.
Now, without further Ado, let's

10
00:00:40,700 --> 00:00:47,100
tune in and explore the future.
Hello, and welcome to a new

11
00:00:47,100 --> 00:00:48,900
episode of the CTO show with
mammoths.

12
00:00:48,900 --> 00:00:51,200
My name is Muhammad.
And I mean each episode, as you

13
00:00:51,208 --> 00:00:54,100
know, I cover different topics
from emerging Technologies,

14
00:00:54,100 --> 00:00:57,600
like, artificial intelligence,
AI, cybersecurity, digital

15
00:00:57,600 --> 00:01:00,000
transformation, Quantum
Computing.

16
00:01:00,000 --> 00:01:04,000
And also at the same time,
recently, I'm doing more

17
00:01:04,000 --> 00:01:08,000
interviews with subject matter,
experts startup, Founders and

18
00:01:08,000 --> 00:01:10,800
entrepreneurs.
And he today, I have Gary who's

19
00:01:10,800 --> 00:01:15,100
joining me garak?
I would let you tell the

20
00:01:15,100 --> 00:01:16,800
audience.
What are you joining from and

21
00:01:16,800 --> 00:01:20,000
what you do?
Yeah, thanks for having me met

22
00:01:20,000 --> 00:01:24,100
and so happy to be here and I'm
calling in from the Philippines.

23
00:01:24,100 --> 00:01:26,100
So, I'm originally from the, the
USA.

24
00:01:26,100 --> 00:01:31,000
My basis is here in the
Philippines and I, my AI

25
00:01:31,000 --> 00:01:34,700
futurist and CEO of a
hollowed-out team at Valhalla.

26
00:01:34,700 --> 00:01:37,500
We specialize in working with
Purpose, Driven tech companies,

27
00:01:37,800 --> 00:01:40,000
helping them build out their
development teams and their

28
00:01:40,000 --> 00:01:42,700
products while leveraging the
power of AI.

29
00:01:43,100 --> 00:01:46,200
You could say, our expertise
lies at the intersection of a AI

30
00:01:46,200 --> 00:01:50,600
IQ and EQ.
Wow, that's fantastic.

31
00:01:50,800 --> 00:01:55,500
Now, in the first place, how did
you first become interested in

32
00:01:55,500 --> 00:02:00,100
Ai and what led you to become an
AI futurist?

33
00:02:01,700 --> 00:02:06,500
So, the I think it all started
when I was quite young as, as

34
00:02:06,500 --> 00:02:11,300
most things do, there was a I
was thinking about this guy.

35
00:02:11,300 --> 00:02:14,100
I get this question asked
normally and in the very

36
00:02:14,100 --> 00:02:17,100
beginning my answer was kind of
flat.

37
00:02:17,100 --> 00:02:19,200
Like, I don't know.
It just was always obvious.

38
00:02:19,200 --> 00:02:21,200
Like it was, it was the thing.
I don't know what sparked it,

39
00:02:21,600 --> 00:02:25,500
but I think, I think it started.
And, you know, same way that a

40
00:02:25,500 --> 00:02:28,100
lot of us, kind of, nerds get
into things that we start with

41
00:02:28,100 --> 00:02:30,200
sci-fi started with the books,
we read.

42
00:02:30,200 --> 00:02:37,400
And, and There's booking a
recursion and it the more I dug

43
00:02:37,400 --> 00:02:41,700
into this idea of building
intelligence from scratch.

44
00:02:42,100 --> 00:02:45,200
The more I started obsessing
over it on pretty much every

45
00:02:45,200 --> 00:02:52,700
level you can imagine because It
strikes me as a very Grand

46
00:02:53,300 --> 00:02:57,300
project, that is really at the
very tip of the iceberg of

47
00:02:57,300 --> 00:02:59,700
everything that it means to be
human for the last many

48
00:02:59,700 --> 00:03:03,300
thousands of years and were,
it's pushing the limits of our

49
00:03:03,300 --> 00:03:05,100
Hardware.
It's pushing the limits of our

50
00:03:05,100 --> 00:03:06,900
philosophy.
It's pushing the limits of our

51
00:03:06,900 --> 00:03:08,300
ethics.
It's pushing the limits of our

52
00:03:08,300 --> 00:03:12,300
industry and it's all coming
together in a way that just

53
00:03:12,400 --> 00:03:14,700
makes right now.
A very exciting time to be

54
00:03:14,708 --> 00:03:18,600
alive.
And so I have been obsessing

55
00:03:18,700 --> 00:03:22,100
This for as long as I can
remember, it looks like a long.

56
00:03:22,200 --> 00:03:26,000
I would say, help session.
I can see this also get it.

57
00:03:26,300 --> 00:03:31,800
Now from what you are seeing,
what is the current state of AI

58
00:03:32,200 --> 00:03:36,600
and because you mentioned, you
work with a lot of your

59
00:03:36,600 --> 00:03:39,600
customers to see how they can
implement this.

60
00:03:39,600 --> 00:03:45,000
So, where do you think the
future holds in terms of, you

61
00:03:45,000 --> 00:03:48,600
know, future of business, and
the impact on society?

62
00:03:49,900 --> 00:03:54,700
Yeah, there's a fantastic quote
of the, the future is here.

63
00:03:54,700 --> 00:03:56,700
It's just not evenly
distributed.

64
00:03:56,900 --> 00:04:04,500
So so a lot of incredible
Innovations have have arrived.

65
00:04:04,500 --> 00:04:07,600
And I think we're entrepreneurs
gonna be making a lot of money,

66
00:04:07,600 --> 00:04:09,900
is the same way they've always
made money, which is going to be

67
00:04:09,900 --> 00:04:12,900
on The Cutting Edge and they're
going to be playing Arbitrage

68
00:04:12,900 --> 00:04:14,800
with the future.
They're going to be figuring out

69
00:04:14,800 --> 00:04:16,100
where the puck is going in
there.

70
00:04:16,100 --> 00:04:17,800
I'll be moving there there
first.

71
00:04:18,100 --> 00:04:23,400
And and I'm advising companies
into thinking about how to

72
00:04:23,400 --> 00:04:26,800
predict what's going to be
happening with AI and where it's

73
00:04:26,800 --> 00:04:30,900
going.
There's quite a few models I

74
00:04:30,900 --> 00:04:34,200
give them but but one that I
want to share here to get

75
00:04:34,200 --> 00:04:38,200
started with is that the the
closest analogy of where we are

76
00:04:38,200 --> 00:04:42,900
at right now to history.
That's always a good foundation.

77
00:04:42,900 --> 00:04:47,900
It started with is at the turn
of the 20th century as

78
00:04:48,800 --> 00:04:52,000
Electricity was being
distributed throughout

79
00:04:52,000 --> 00:04:53,800
civilization.
So think of Thomas Edison and

80
00:04:53,800 --> 00:05:01,000
light bulb think of the the
power grid throughout Europe

81
00:05:01,000 --> 00:05:06,400
throughout the this throughout
the west where the

82
00:05:06,400 --> 00:05:10,400
transformation meant that any
technology we had before, we

83
00:05:10,400 --> 00:05:13,600
could add power to it through
the form of electricity.

84
00:05:13,600 --> 00:05:19,700
So a hammer became a jackhammer
a, you know, No a a carriage

85
00:05:19,700 --> 00:05:23,300
became a car originally.
The first cars were electric a

86
00:05:23,600 --> 00:05:26,000
screwdriver, you know, became a
power tool.

87
00:05:26,000 --> 00:05:30,800
So on so forth, and what we're
doing right now is something

88
00:05:31,100 --> 00:05:34,200
similar where.
At this point we are adding

89
00:05:34,700 --> 00:05:38,500
intelligence to just about
anything that you could imagine

90
00:05:39,000 --> 00:05:43,100
and because the intelligence at
this point has an advanced, the

91
00:05:43,108 --> 00:05:47,200
point where it can emulate human
communication.

92
00:05:48,400 --> 00:05:51,400
Literally, let the answer is it,
like everything is going to be

93
00:05:51,400 --> 00:05:53,500
changed.
So what I suggest people to be

94
00:05:53,500 --> 00:05:56,200
thinking about, is to be taking
their subject.

95
00:05:56,200 --> 00:05:59,400
Matter expertise, and this is,
you know, same with most

96
00:05:59,400 --> 00:06:00,800
Innovations, right?
Is it.

97
00:06:00,800 --> 00:06:03,300
Take your subject matter.
Expertise takes what you find

98
00:06:03,300 --> 00:06:09,300
unique, what makes you unique?
And to figure out how a i is

99
00:06:09,300 --> 00:06:12,000
going to be entering into it.
How is what you're going to be

100
00:06:12,000 --> 00:06:19,800
doing going to be augmented by
it and If that is a little bit

101
00:06:19,800 --> 00:06:23,800
too broad than the the way that
you break it down, is you think

102
00:06:23,800 --> 00:06:27,500
through some sort of value added
chain, I'm going to assume that

103
00:06:27,500 --> 00:06:30,100
you either an entrepreneur that
has a business or going into a

104
00:06:30,108 --> 00:06:34,400
business and you break down that
that value chain of what that

105
00:06:34,400 --> 00:06:36,400
business would look like, or
what that process is.

106
00:06:36,800 --> 00:06:41,100
And then think through every
step of that value chain is both

107
00:06:41,100 --> 00:06:43,400
adding value.
But also subtracting value

108
00:06:43,500 --> 00:06:46,200
because costs accrue friction is
added.

109
00:06:46,300 --> 00:06:49,400
Miscommunications could be added
but Also theoretical value is

110
00:06:49,400 --> 00:06:51,800
added.
So, you know, in e-commerce to

111
00:06:51,800 --> 00:06:54,100
there's a point where they see
the advertisement and they click

112
00:06:54,100 --> 00:06:56,800
on it, then they put in the
order button or so on and so

113
00:06:56,800 --> 00:06:58,700
forth.
At every one of those stages,

114
00:06:58,700 --> 00:07:03,600
you can add in different forms
of AI, to increase the

115
00:07:03,600 --> 00:07:06,900
efficiency and decrease the, the
cost.

116
00:07:07,500 --> 00:07:11,100
And I guess I'll leave it there
for now.

117
00:07:11,100 --> 00:07:14,200
I think I mean there's also some
interesting comments about what

118
00:07:14,200 --> 00:07:16,900
are the recent Innovations in Ai
and what's coming down the

119
00:07:16,900 --> 00:07:18,900
pipeline?
But I'll, I'll leave it there

120
00:07:18,900 --> 00:07:22,100
for now.
Yeah, we come back there but

121
00:07:22,100 --> 00:07:26,400
just, you know, like being the
devil's advocate here, right?

122
00:07:26,400 --> 00:07:31,700
So someone might say, because
you mentioned a very important,

123
00:07:33,000 --> 00:07:37,300
I would say phrase there and you
said it's like the moment is

124
00:07:37,300 --> 00:07:40,800
here and, you know, you did.
Similarity with previous things.

125
00:07:41,100 --> 00:07:46,200
Now someone might say getting
but you know the AI Is Not A New

126
00:07:46,200 --> 00:07:48,300
Concept.
AI has been with us now for

127
00:07:48,300 --> 00:07:54,900
sometimes why now all the hype
is happening around AI.

128
00:07:54,900 --> 00:08:00,200
Is it because You know, the
generative AI what we are seeing

129
00:08:00,200 --> 00:08:02,900
with tools, like, Chad GPT from
open-air.

130
00:08:02,900 --> 00:08:07,000
I bought from Google meet
journey, and all these tools

131
00:08:07,300 --> 00:08:12,800
that why the hype happened now
and not before, knowing that,

132
00:08:12,800 --> 00:08:16,900
and I covered this in one of my
episode of the history of a i AI

133
00:08:17,300 --> 00:08:20,900
you know, the concept goes back
to very old times and then

134
00:08:20,900 --> 00:08:25,400
Computing I mean in modern age
has its start in the 50s.

135
00:08:25,400 --> 00:08:28,300
You know, in all this movement
that happened Din the

136
00:08:28,400 --> 00:08:33,299
universities across the u.s.,
but why now is the moment of AI?

137
00:08:34,500 --> 00:08:38,500
So in in my opinion, that there
is a certain level of

138
00:08:38,500 --> 00:08:43,799
arbitrariness to it because the
technology that underpins a lot

139
00:08:43,799 --> 00:08:47,900
of the recent, you know, gizmos
and things that are being

140
00:08:47,900 --> 00:08:52,100
distributed masses, to be bar,
chatter BT, Etc, were based on

141
00:08:52,100 --> 00:08:55,600
an innovation in 2017 from a
paper called.

142
00:08:55,600 --> 00:08:59,600
Attention is all you need.
Which was kind of a precursor

143
00:08:59,600 --> 00:09:03,200
paper to the Transformer.
And so, the Transformers, it's a

144
00:09:03,208 --> 00:09:10,900
fairly technical concept that
was an innovation on to our

145
00:09:10,900 --> 00:09:14,100
existing, you know,
understanding of how to use

146
00:09:14,100 --> 00:09:16,700
neural Nets, and how to use
machine learning and the

147
00:09:16,700 --> 00:09:20,700
fundamental, well, actually, I'd
be very curious momentum.

148
00:09:20,700 --> 00:09:23,600
If you're familiar with
Transformers what you, what, you

149
00:09:23,900 --> 00:09:26,400
perceive your perspective would
be on what they added to the

150
00:09:26,400 --> 00:09:32,300
game, but From what?
Has been impactful for me about

151
00:09:32,300 --> 00:09:36,400
Transformers, is it?
They unlocked a much greater

152
00:09:36,400 --> 00:09:39,400
degree of parallel processing,
amongst other things which

153
00:09:39,400 --> 00:09:41,900
basically meant that we could
be.

154
00:09:43,900 --> 00:09:47,700
We could do horizontal scaling
with our machine models in a way

155
00:09:47,700 --> 00:09:50,800
that we couldn't do before.
So all of a sudden it became a

156
00:09:50,800 --> 00:09:57,600
lot more tenable for us to
consume 10% of the internet in

157
00:09:57,600 --> 00:10:03,200
order to create a model, like
gbt 3.5, which was the chat, CBT

158
00:10:03,600 --> 00:10:06,800
and, and it's still billions of
tokens or however many it was.

159
00:10:07,300 --> 00:10:10,400
So, the Transformers had a lot
of innovations.

160
00:10:10,500 --> 00:10:14,700
That was one of them.
Another was that it Transformers

161
00:10:14,700 --> 00:10:22,100
allows you to chunk a piece of
input data and to add you

162
00:10:22,108 --> 00:10:27,500
consume that input data to tell
its context.

163
00:10:27,500 --> 00:10:31,900
So so what I mean by that is the
original paper attention, all

164
00:10:31,900 --> 00:10:37,000
you need was was focusing on how
to translate languages.

165
00:10:37,000 --> 00:10:40,700
So the example I gave was from
from French to English and

166
00:10:40,700 --> 00:10:44,900
various phrases in French, you
know, will have Different

167
00:10:44,900 --> 00:10:47,200
sentence structure so it's not
one-to-one.

168
00:10:47,400 --> 00:10:52,600
You know, translation you'll
have Yoda speak and the this the

169
00:10:52,600 --> 00:10:54,200
grammar structure will be will
be thrown off.

170
00:10:54,400 --> 00:10:58,900
So the the Transformers were
able to more efficiently discern

171
00:10:58,900 --> 00:11:02,600
that by being able to tell how
the words existed and came

172
00:11:02,600 --> 00:11:06,800
together into sentences.
So, long story short, is that it

173
00:11:06,800 --> 00:11:10,400
could, it could consume larger
pieces of data and approach

174
00:11:10,400 --> 00:11:13,400
something much closer like
human.

175
00:11:13,700 --> 00:11:15,800
Ish understanding.
I wouldn't say that they are

176
00:11:15,800 --> 00:11:18,300
anything like human
consciousness but they could,

177
00:11:18,300 --> 00:11:22,600
they could better understand
larger volumes of data and parse

178
00:11:22,600 --> 00:11:25,000
the patterns between them.
So those were the two things.

179
00:11:25,000 --> 00:11:30,400
So why did it not launch off in
2017 late 2017?

180
00:11:30,400 --> 00:11:34,900
When, when that paper came out,
you know, I think it took about

181
00:11:36,600 --> 00:11:40,900
For four and a half years before
enough Engineers really

182
00:11:40,900 --> 00:11:44,500
understood the potential and
really started to build out the

183
00:11:44,500 --> 00:11:46,100
infrastructure to make chatty
Beauty.

184
00:11:46,100 --> 00:11:50,600
And then the quote-unquote hype
has been that Chad gbt was

185
00:11:50,600 --> 00:11:53,400
released and now anybody had
access to it for free.

186
00:11:53,400 --> 00:11:58,700
They took a gander with having a
billion dollars of funding from

187
00:11:58,700 --> 00:12:00,400
Microsoft.
Now most of that was in server

188
00:12:00,400 --> 00:12:04,500
server tokens but putting that
aside they had a big investment

189
00:12:04,500 --> 00:12:08,600
enough To go Mass public and you
know, the future is here.

190
00:12:08,600 --> 00:12:10,600
It's just not evenly
distributed, people didn't know

191
00:12:10,600 --> 00:12:13,000
this technology.
Really technology was out there

192
00:12:13,000 --> 00:12:16,700
and so it took that long before
it hit mainstream and here we

193
00:12:16,700 --> 00:12:20,000
are.
Yeah, I remember and this is now

194
00:12:20,000 --> 00:12:21,600
I'm not the devil advocate
anymore.

195
00:12:22,200 --> 00:12:28,000
I am who I am.
Usually, I think I started to

196
00:12:28,000 --> 00:12:34,600
see like discussions online and
then, I can't remember exactly

197
00:12:34,600 --> 00:12:40,000
how Came up with because at
first there was the API of what

198
00:12:40,000 --> 00:12:44,900
people knows now, as Chad GPT.
So yeah, open AI, they release

199
00:12:44,900 --> 00:12:49,400
the closed, beta or closed
platform.

200
00:12:49,700 --> 00:12:53,300
And at that time you know like
they put some use cases there.

201
00:12:53,300 --> 00:12:57,200
And I was thinking like okay
this is really interesting, I'm

202
00:12:57,200 --> 00:13:00,300
not a coder but by any means but
of course I understand I have

203
00:13:00,300 --> 00:13:03,400
the knowledge.
Okay so they did then API that

204
00:13:03,400 --> 00:13:06,300
you can give some input and then
It gives you some output that's

205
00:13:06,300 --> 00:13:09,700
cook interesting.
And then you know in December or

206
00:13:09,700 --> 00:13:13,500
November I believe end of
November when they release chat

207
00:13:13,500 --> 00:13:17,100
GPT right?
Oh man.

208
00:13:17,100 --> 00:13:20,300
Like like yeah.
This is something that we never

209
00:13:20,300 --> 00:13:24,000
saw before and now to your point
Gary and this way now I'm not

210
00:13:24,000 --> 00:13:27,800
the devil advocate and I
repeated multiple times y.

211
00:13:27,800 --> 00:13:32,700
It is the moment because you
know, the anyone that is

212
00:13:32,700 --> 00:13:37,500
familiar with computer science.
knows that, you know, natural

213
00:13:37,700 --> 00:13:43,200
language processing and knows
that, you know, having ability

214
00:13:43,200 --> 00:13:47,500
of a machine to understand the
human like it's like the

215
00:13:47,700 --> 00:13:53,600
Paramount I would say theorem
Theory whatever you want to want

216
00:13:53,600 --> 00:13:58,000
problem that everyone wanted to
solve and you know this

217
00:13:58,000 --> 00:14:02,900
breakthrough because You know,
I'm like you interested in AI

218
00:14:02,900 --> 00:14:07,000
from long time and the key for
doing other things with a, I

219
00:14:07,000 --> 00:14:09,700
start by letting the machine
understand actually, what you

220
00:14:09,700 --> 00:14:12,400
want.
So, if I can let the machine,

221
00:14:12,400 --> 00:14:15,500
understand what I'm thinking
about, I can later go and tell

222
00:14:15,500 --> 00:14:19,200
the machine to go and do it,
because it can understand me

223
00:14:19,700 --> 00:14:23,300
now.
Take taking this from this

224
00:14:23,300 --> 00:14:28,100
point, I want to understand from
you and for the audience, of

225
00:14:28,100 --> 00:14:35,600
course, how this will affect the
future of jobs careers and our

226
00:14:35,600 --> 00:14:42,400
lives from your point of view.
Yeah, I think that there's as

227
00:14:42,400 --> 00:14:44,100
always is with large amounts of
change.

228
00:14:44,100 --> 00:14:49,800
There's a lot of fear, there's a
lot of uncertainty and it's

229
00:14:49,800 --> 00:14:54,100
becomes quite hard to to, you
know, for individuals who

230
00:14:54,100 --> 00:14:56,400
haven't been thinking about this
stuff to predict what's going to

231
00:14:56,400 --> 00:15:03,500
happen next and I I think that
the, the two things that I talk

232
00:15:03,500 --> 00:15:07,900
with people about who are
worried about AI taking their

233
00:15:07,900 --> 00:15:11,500
jobs, disrupting their
livelihood, is that first of

234
00:15:11,500 --> 00:15:19,100
all, AI is not going to replace.
Anybody it's instead, it's going

235
00:15:19,100 --> 00:15:23,800
to amplify some people to be so
hyper productive that they're

236
00:15:23,800 --> 00:15:25,900
going to replace people.
So, in other words, your jobs

237
00:15:25,900 --> 00:15:28,300
not only taken by an AI.
It's not be taken by someone

238
00:15:28,400 --> 00:15:30,100
who's using a, I better than
you.

239
00:15:31,200 --> 00:15:35,200
Yeah.
And so in some Industries there

240
00:15:35,200 --> 00:15:39,700
are they are more constrained by
Supply than demand what I mean

241
00:15:39,700 --> 00:15:44,700
by that is if if you talk with
people in the tech world, you

242
00:15:44,700 --> 00:15:46,500
know, we kind of need ten times
more.

243
00:15:46,600 --> 00:15:48,300
More coders than we have right
now.

244
00:15:48,300 --> 00:15:52,200
Like like this, the constraint
has been on the supply side.

245
00:15:52,200 --> 00:15:55,800
We want more code.
We I would also argue that we

246
00:15:55,800 --> 00:15:58,500
want more art in the world.
That there are a lot of things

247
00:15:58,500 --> 00:16:01,600
that we have taken for granted
the supply and demand curve are

248
00:16:01,600 --> 00:16:04,100
where they're at, but give it a
few years and we're probably

249
00:16:04,100 --> 00:16:08,400
going to really find that the
demand has increased

250
00:16:08,400 --> 00:16:11,900
exponentially but of course,
there are other roles one that

251
00:16:11,900 --> 00:16:16,100
comes to mind is I believe
McDonald's has recently started

252
00:16:16,100 --> 00:16:21,400
testing I for there for taking
orders through at drive-throughs

253
00:16:21,500 --> 00:16:25,600
and things like that and so it's
like is that going to increase

254
00:16:25,600 --> 00:16:30,800
exponentially as well?
There there are some historical

255
00:16:30,800 --> 00:16:36,400
examples such as with bank,
tellers back in the 1980s.

256
00:16:37,200 --> 00:16:39,500
When the ATM was introduced.
We thought bank, tellers were

257
00:16:39,508 --> 00:16:43,200
going to go out of business.
They they were gone drop down by

258
00:16:43,200 --> 00:16:47,500
like 1/5 because 4/5 of their
time was spent just punching

259
00:16:47,500 --> 00:16:53,400
papers and, and doing things at
ATMs can do, but when ATMs were

260
00:16:53,400 --> 00:16:56,300
popularized, in fact, bank,
tellers jobs and not go down

261
00:16:56,300 --> 00:16:57,800
there.
She went up because then the

262
00:16:57,800 --> 00:17:01,900
bank's wanted to create more
Banks, they want to create more

263
00:17:01,900 --> 00:17:04,700
outlets.
And so the bank teller job was

264
00:17:04,700 --> 00:17:06,400
not was not affect the way we
expected.

265
00:17:06,400 --> 00:17:09,700
So I think there's going to be a
lot of surprises in this

266
00:17:09,700 --> 00:17:17,700
journey, but By taking advantage
of this technology and not

267
00:17:17,700 --> 00:17:21,000
seeing as a race with Silicon
Valley, but instead, just erase

268
00:17:21,000 --> 00:17:24,300
with your industry, as you
already were doing, you already

269
00:17:24,300 --> 00:17:26,300
were competing with people in
your field.

270
00:17:26,300 --> 00:17:30,900
Now it's just there's, there's a
new, a new tool in the tool

271
00:17:30,900 --> 00:17:32,500
belt.
That's a much healthier way to

272
00:17:32,500 --> 00:17:34,100
look at it.
The other piece of advice, I

273
00:17:34,100 --> 00:17:39,500
give people is that when you are
in a state of fear, we know this

274
00:17:39,500 --> 00:17:44,900
with with Neuroscience that the
blood Pools away from your

275
00:17:44,900 --> 00:17:49,900
neocortex and your 32 percent,
less capable of making creative

276
00:17:49,900 --> 00:17:52,400
decisions and and higher-order
executive decisions.

277
00:17:52,700 --> 00:17:59,300
And so I really encourage people
to You know, it's real advice to

278
00:17:59,300 --> 00:18:02,100
look on the bright side and to
get excited for this technology

279
00:18:02,100 --> 00:18:06,800
is going.
If you are if you're scared, if

280
00:18:06,800 --> 00:18:09,100
you're playing on defense, if
you're going on the passive and

281
00:18:09,100 --> 00:18:12,600
try and like whether this out,
then you're not going to be at

282
00:18:12,600 --> 00:18:15,500
the level that we really need
you to be.

283
00:18:16,300 --> 00:18:20,800
And so those are generally how I
think about most people's

284
00:18:20,800 --> 00:18:22,900
tactics for, for weathering
things.

285
00:18:23,500 --> 00:18:25,800
Yeah.
On this point.

286
00:18:26,800 --> 00:18:30,100
What?
I would say, like, feeding

287
00:18:31,200 --> 00:18:36,100
complaining, I don't know, like,
all this work that I can think

288
00:18:36,100 --> 00:18:39,100
about Shawn - would not change
anything.

289
00:18:39,100 --> 00:18:42,900
Like even we saw some of the big
names even in Silicon Valley.

290
00:18:42,900 --> 00:18:46,800
They signed a paper for
stopping, you know, the research

291
00:18:46,800 --> 00:18:49,900
and on AI, some people ask me,
what do you think?

292
00:18:50,000 --> 00:18:55,600
I said, okay, look if openly I
or Google, they don't go and

293
00:18:55,600 --> 00:19:00,200
make it, someone else will come.
Make it because, you know, it's

294
00:19:00,200 --> 00:19:03,600
matter of as we call it in the
tech world.

295
00:19:03,600 --> 00:19:06,300
And, you know, in the Consulting
world, you need a proof of

296
00:19:06,300 --> 00:19:08,800
concept, right?
So you need to proof of concept.

297
00:19:08,800 --> 00:19:12,900
And once you have the proof of
concept, succeeded.

298
00:19:12,900 --> 00:19:16,300
So we know we can do this the
Genies out of the bottle, so to

299
00:19:16,300 --> 00:19:17,700
speak.
Exactly.

300
00:19:17,700 --> 00:19:20,100
So if Mammoth will not do it,
get it tomorrow.

301
00:19:20,100 --> 00:19:22,300
We'll go and do it.
If Gary doesn't do it, someone

302
00:19:22,300 --> 00:19:26,100
else will come and do it because
we know that it can also

303
00:19:26,300 --> 00:19:28,800
collecting Yeah it's gonna take
some time with them.

304
00:19:28,800 --> 00:19:31,800
Probably I'm not sure.
But with all the processing

305
00:19:31,800 --> 00:19:35,800
speed you know like building a
large language model.

306
00:19:35,900 --> 00:19:37,600
Yeah.
I think it's time it needs all

307
00:19:37,600 --> 00:19:40,500
the city's resources but you
know like I believe in six

308
00:19:40,500 --> 00:19:43,600
months time we can see another
open a i for example, right?

309
00:19:43,600 --> 00:19:47,200
But the can yeah the same thing.
Have you followed the the leak

310
00:19:47,200 --> 00:19:51,400
with with a llama from from from
the the Facebook a I have you

311
00:19:51,400 --> 00:19:53,000
followed this story?
Oh yeah.

312
00:19:53,000 --> 00:19:54,000
Yes.
Yeah.

313
00:19:54,700 --> 00:19:59,000
So once that so Book had their
own large language model.

314
00:19:59,200 --> 00:20:01,800
They called it llama and it was
it was leaked out.

315
00:20:03,000 --> 00:20:06,200
But something really interesting
has which I did not predict

316
00:20:06,600 --> 00:20:11,500
occurred which is that once it
leaked out to the masses, the

317
00:20:11,500 --> 00:20:14,300
open source Community has taken
it and run with it and they have

318
00:20:14,300 --> 00:20:17,800
found Justin certain
efficiencies of how to create

319
00:20:17,800 --> 00:20:22,700
new large language models with
with using the Llama framework

320
00:20:23,000 --> 00:20:25,800
that are orders of magnitude
more efficient.

321
00:20:25,900 --> 00:20:29,500
So while it cost a Million
dollars in processing to create

322
00:20:29,500 --> 00:20:34,200
chat CBT.
We are, we are now creating new

323
00:20:34,200 --> 00:20:38,500
large language models that are
as good as G BT 3 not as good as

324
00:20:38,500 --> 00:20:45,100
GP for yet, but are as good as G
PT 3, which it was is pretty

325
00:20:45,100 --> 00:20:48,800
much good enough on orders of
magnitude less.

326
00:20:48,800 --> 00:20:53,600
I think Google came out this
paper chinchilla which, which

327
00:20:53,600 --> 00:20:55,900
had some of the major
breakthroughs that helped the

328
00:20:55,900 --> 00:20:58,000
open source community and then
And there's a few other

329
00:20:59,300 --> 00:21:02,600
breakthroughs, they had as well.
But at this point what from what

330
00:21:02,608 --> 00:21:07,200
I'm seeing is that people with a
beefy enough home set up and a

331
00:21:07,200 --> 00:21:11,800
few few days of downtime can be
creating these large language

332
00:21:11,800 --> 00:21:15,600
models, which I do not think I
thought the constraint was going

333
00:21:15,600 --> 00:21:18,000
to be a half a billion dollars
in a server Farm.

334
00:21:18,500 --> 00:21:22,000
But this point true that the
genes outside the bottle and if

335
00:21:22,000 --> 00:21:25,000
anyone's interested in that, I
would suggest the Google paper

336
00:21:25,200 --> 00:21:28,800
called I think it was called We
have no moat and it was a leaked

337
00:21:28,800 --> 00:21:31,100
Google Document.
If you Google we have no motor

338
00:21:31,100 --> 00:21:34,300
Google and neither does open
a.i. you'll see a link Google

339
00:21:34,300 --> 00:21:36,200
document that goes in a lot more
detail on that.

340
00:21:36,800 --> 00:21:38,700
Yeah.
So surprisingly and, you know,

341
00:21:38,700 --> 00:21:42,700
this is one and information to
give to the audience if you are

342
00:21:42,700 --> 00:21:47,200
watching or listening to us.
So it's my, it's my top hit

343
00:21:48,400 --> 00:21:50,500
episode.
Actually on the show, it get the

344
00:21:50,500 --> 00:21:54,000
highest number of listening and
I didn't put on YouTube because

345
00:21:54,000 --> 00:21:57,100
was a solo recorded without
camera, but you know everyone.

346
00:21:57,400 --> 00:22:01,800
Started to also ping me about it
when I covered Auto GP T and

347
00:22:01,800 --> 00:22:05,000
this is exactly, you know, like
they used actually the Llama

348
00:22:05,600 --> 00:22:09,500
open source, right?
So so to do a few things and

349
00:22:10,300 --> 00:22:12,300
it's interesting.
Now you mentioned getting

350
00:22:12,300 --> 00:22:14,900
something, and I'll ask you this
question, not from technical

351
00:22:14,900 --> 00:22:19,300
perspective to just also shed
some light on the, you know,

352
00:22:19,300 --> 00:22:21,800
like the future of work that we
talked about.

353
00:22:21,800 --> 00:22:26,100
Now, you always separated
between GPT 3.5 and GPT for,

354
00:22:26,200 --> 00:22:27,000
right?
Yeah.

355
00:22:27,300 --> 00:22:33,300
The reason is and I did also a
like a dataview kind of between

356
00:22:33,300 --> 00:22:40,400
birth and GPT for and always I
repeat myself that to me GPT for

357
00:22:40,400 --> 00:22:44,100
has the ability and even if you
have access to GPT for today,

358
00:22:44,500 --> 00:22:47,300
you know, like they are not sure
if they still show it or not

359
00:22:48,300 --> 00:22:50,100
open.
They are they show you something

360
00:22:50,100 --> 00:22:55,600
like a score for the model?
So for example, GPT 3.5 is fast

361
00:22:55,800 --> 00:22:59,200
GPT for like Less fast.
I mean slow down a little bit

362
00:22:59,400 --> 00:23:01,500
and then you have reasoning,
right?

363
00:23:01,500 --> 00:23:07,400
So the reasoning part.
Now from an AI futurist

364
00:23:07,400 --> 00:23:10,900
perspective.
Like, when we say reasoning, we

365
00:23:10,900 --> 00:23:15,400
mean someone that has the
ability to come up with

366
00:23:15,400 --> 00:23:18,000
something that doesn't exist
with a.

367
00:23:18,000 --> 00:23:21,300
I be able to do that as a great
question.

368
00:23:23,900 --> 00:23:27,400
You know it says it's for this
one could guess what?

369
00:23:28,300 --> 00:23:36,900
So so I guess I guess bottom
line I would say Yes, because we

370
00:23:36,900 --> 00:23:40,000
just see how fast were
advancing.

371
00:23:40,000 --> 00:23:41,600
And how many breakthroughs were
making?

372
00:23:43,300 --> 00:23:46,900
How far is that in the future to
come up with true creativity?

373
00:23:47,100 --> 00:23:49,400
It's hard to say.
But I will, I will talk a little

374
00:23:49,400 --> 00:23:51,800
bit about what's the gap between
where we are.

375
00:23:51,800 --> 00:23:54,200
And that point so I think we're
going to cross it.

376
00:23:54,900 --> 00:23:57,500
We're probably on across it this
Century but I want to talk a

377
00:23:57,508 --> 00:23:59,500
little bit about what that Gap
is.

378
00:23:59,500 --> 00:24:04,300
So if you go to chat CBT and you
ask it, What's two plus two?

379
00:24:04,300 --> 00:24:08,300
It will give you the answer for.
But if you ask it a very complex

380
00:24:08,300 --> 00:24:12,900
math or not complex but a very
long math problem like what's,

381
00:24:12,900 --> 00:24:16,800
you know, five thousand trillion
400, you know, so and so forth

382
00:24:17,300 --> 00:24:19,600
times and equally large
ridiculous number.

383
00:24:19,700 --> 00:24:21,100
So number.
So it's a math problem.

384
00:24:21,100 --> 00:24:25,700
That is not very popular online.
It will give you a wrong number.

385
00:24:26,100 --> 00:24:28,600
And, and if you punch into a
calculator, you will verify that

386
00:24:28,600 --> 00:24:32,400
the wrong number.
But the first few digits will be

387
00:24:32,400 --> 00:24:35,700
correct.
The last few digits will be

388
00:24:35,700 --> 00:24:38,300
correct of the new number.
Yeah.

389
00:24:38,700 --> 00:24:42,300
So what what it's doing there
is, it's recognizing patterns,

390
00:24:42,700 --> 00:24:44,100
it's not actually calculating
the numbers in.

391
00:24:44,108 --> 00:24:49,600
This is kind of a weird, a weird
thing to get your head around

392
00:24:49,600 --> 00:24:53,600
that when you're asking these
questions, it's not reasoning

393
00:24:53,600 --> 00:24:55,700
through the problem.
It's not calculating it even

394
00:24:55,700 --> 00:24:57,500
though it's a computer, you
expect a computer to calculate

395
00:24:57,500 --> 00:25:01,900
it for, for Christ's sake, but
it's just recognizing patterns

396
00:25:01,900 --> 00:25:03,700
in the internet.
So if you ask What's two plus

397
00:25:03,700 --> 00:25:06,400
two?
It will have millions of

398
00:25:06,400 --> 00:25:09,500
examples of people saying two
plus two equals four so it knows

399
00:25:09,500 --> 00:25:11,400
the answer is 4.
But if you ask it a question, it

400
00:25:11,408 --> 00:25:14,000
hasn't seen before even a math
question that you could reason

401
00:25:14,000 --> 00:25:18,800
through with enough time it will
give a wrong answer but it will

402
00:25:18,800 --> 00:25:22,300
get parts of it correct.
Because it has already

403
00:25:22,300 --> 00:25:25,700
recognized a pattern that to get
the first few digits.

404
00:25:25,700 --> 00:25:29,800
And the last few digits of a
very complex multiplication, it

405
00:25:29,800 --> 00:25:32,500
only has to look at the first
few and last few digits of of

406
00:25:32,500 --> 00:25:35,400
the two What do you call them
component?

407
00:25:35,500 --> 00:25:39,200
Component numbers.
So, it is the pattern

408
00:25:39,200 --> 00:25:43,200
recognizing has gotten us
further than we thought it.

409
00:25:43,200 --> 00:25:46,300
But it only has the illusion
that actually has a model of

410
00:25:46,300 --> 00:25:48,200
reality.
So the way that I think about

411
00:25:48,200 --> 00:25:51,500
this is that they're us humans.
When we are reading through

412
00:25:51,500 --> 00:25:55,000
things, we reason things both
bottom up where we have

413
00:25:55,000 --> 00:25:59,400
empirical historical data of
what we've seen and and we have

414
00:25:59,400 --> 00:26:03,100
examples, but we also think
through things top down you can

415
00:26:03,200 --> 00:26:05,000
Imagine something you've never
seen before.

416
00:26:05,400 --> 00:26:08,300
Yeah.
And right now, these

417
00:26:08,300 --> 00:26:13,000
Technologies, they and I would
encourage people to play around

418
00:26:13,300 --> 00:26:15,400
with Chad, you beauty is do, is
you punch in some of these

419
00:26:15,400 --> 00:26:16,700
numbers?
It's quite Illuminating.

420
00:26:18,500 --> 00:26:24,700
What this technology is, is it,
it has a illusion of top-down

421
00:26:24,700 --> 00:26:29,600
thinking and it's, um, it's done
that only through bottom up.

422
00:26:29,800 --> 00:26:34,200
And so, I think that as we are
developing Better and better

423
00:26:34,200 --> 00:26:38,600
Technologies for these a is to
have internal simulated versions

424
00:26:38,600 --> 00:26:41,200
of reality.
So we're creating like these

425
00:26:41,900 --> 00:26:45,100
virtual Minds like physics
simulations and other things

426
00:26:45,600 --> 00:26:48,700
that are a little bit more
obeying, the laws of physics.

427
00:26:48,700 --> 00:26:52,800
So to speak, I think we're going
to get closer to an AI That's

428
00:26:52,800 --> 00:26:55,400
able to do something a little
more top-down and that's going

429
00:26:55,400 --> 00:26:58,000
to be quite exciting.
I think that's going to be what

430
00:26:58,000 --> 00:27:00,900
the next few Innovations are.
Transformers, they've taken us

431
00:27:01,200 --> 00:27:03,900
further than we ever thought
they would but they are Going to

432
00:27:04,000 --> 00:27:06,300
peek out, and we need a few more
Innovations.

433
00:27:06,300 --> 00:27:10,400
A few new truly creative new
Innovations and no one's talking

434
00:27:10,400 --> 00:27:12,300
about yet.
We need a few more of those

435
00:27:12,300 --> 00:27:15,200
before really going to crack
that problem, but the question

436
00:27:15,200 --> 00:27:19,600
got it is, don't you think that
always, we need someone to punch

437
00:27:19,600 --> 00:27:25,400
this as we call it prompt that
who gonna give the prompt today?

438
00:27:25,400 --> 00:27:28,800
I is it like an other AI that is
giving that prompt how it will

439
00:27:28,800 --> 00:27:35,600
work.
So you're asking in like a

440
00:27:35,608 --> 00:27:38,600
hundred years from now.
What's going to be the process

441
00:27:38,600 --> 00:27:43,900
of maximizing a eyes utility?
Yeah, that let me give you a

442
00:27:43,900 --> 00:27:48,100
more like straightforward
example.

443
00:27:49,100 --> 00:27:55,200
So assuming that a, I will have
the power to come up with

444
00:27:55,700 --> 00:28:00,100
completely new business model or
a completely new.

445
00:28:00,200 --> 00:28:02,000
Software.
Let's make it more simple.

446
00:28:02,500 --> 00:28:07,800
Okay, so someone need to tell
whatever, you know, the model

447
00:28:07,800 --> 00:28:11,300
they are using that.
I need a software that I believe

448
00:28:11,300 --> 00:28:12,900
it should do.
One, two, three, four.

449
00:28:12,900 --> 00:28:16,100
And then the AI will go and
write the code.

450
00:28:16,500 --> 00:28:21,600
They I will do the UI they I
will do the ux but who is giving

451
00:28:21,600 --> 00:28:25,100
this ignite to start you know
the process.

452
00:28:26,800 --> 00:28:30,500
Yeah, you know.
So if we're talking about a I

453
00:28:30,500 --> 00:28:34,200
having greater and greater
impact into the real world, so

454
00:28:34,200 --> 00:28:37,300
it was like the idea is you ask,
Chad, your beauty make me a

455
00:28:37,300 --> 00:28:39,300
million-dollar company.
So it's based like you know find

456
00:28:39,300 --> 00:28:41,300
a way to put a million dollars
in my bank account right now and

457
00:28:41,300 --> 00:28:42,100
make it happen.
Yeah.

458
00:28:42,800 --> 00:28:48,700
And you if everybody has access
to that then then a lot of

459
00:28:48,700 --> 00:28:50,900
things are gonna start to break
down at least in terms of like

460
00:28:50,900 --> 00:28:55,000
what is the definition of some
of the things like if if

461
00:28:55,000 --> 00:28:58,700
everybody has infinite money
than money is less valuable,

462
00:28:58,700 --> 00:29:00,700
right?
And we're still in a physical

463
00:29:00,700 --> 00:29:03,300
universe.
So you know what we're aiming

464
00:29:03,300 --> 00:29:08,600
for is a post-scarcity
civilization where where you

465
00:29:08,608 --> 00:29:10,000
know, people are all given
enough.

466
00:29:10,000 --> 00:29:15,200
And and, you know, I think that
Star Trek Vision still is at the

467
00:29:15,200 --> 00:29:17,700
heart of a lot of nerds like
you.

468
00:29:17,700 --> 00:29:22,000
And I trying to make the world a
better place, but, you know, as

469
00:29:22,000 --> 00:29:24,100
an AI futurist.
And I'm looking at this stuff,

470
00:29:24,100 --> 00:29:29,100
practically we Need to keep in
mind that there are these

471
00:29:29,100 --> 00:29:31,100
Technologies are existing in a
physical universe.

472
00:29:31,600 --> 00:29:39,200
And so if we are Asking it
questions like give me

473
00:29:39,200 --> 00:29:42,500
comparative power over other
humans like make a trillion

474
00:29:42,500 --> 00:29:44,300
dollar company.
Hmm.

475
00:29:44,300 --> 00:29:49,600
Bit If everybody has it then it
won't work because it's

476
00:29:49,600 --> 00:29:52,400
comparative value and
Technologies democratized.

477
00:29:52,800 --> 00:29:55,200
So then the real game we have to
play the infinite game where

478
00:29:55,200 --> 00:30:00,600
everybody has enough or be one
person has the only access to

479
00:30:00,600 --> 00:30:03,600
Super Ai and then they ask and
make a trillion-dollar thing and

480
00:30:03,600 --> 00:30:05,900
they're a eyes more powerful.
So, it works.

481
00:30:06,700 --> 00:30:14,100
Yeah, so it's one of those two
scenarios and I think that It's

482
00:30:14,100 --> 00:30:16,600
a really beautiful thing where
we are right now we and I hope

483
00:30:16,600 --> 00:30:19,500
it plays out that the
philosophy.

484
00:30:19,500 --> 00:30:22,300
That's underpinning.
A lot of these companies is it

485
00:30:22,300 --> 00:30:25,600
we want to democratize this
company, we don't want a few

486
00:30:26,500 --> 00:30:30,100
Super Players to do it.
And so that was the idea that

487
00:30:30,100 --> 00:30:32,300
opened my eyes.
I was founded with now, is it

488
00:30:32,300 --> 00:30:34,700
still living up to its ideals?
You know, I think there's a

489
00:30:34,708 --> 00:30:39,600
debate to be had about that, but
principles are still very much

490
00:30:39,600 --> 00:30:43,300
in the culture of the open
source market of Allah.

491
00:30:43,500 --> 00:30:45,600
People work at open.
A I, you know, we're all just

492
00:30:45,600 --> 00:30:47,200
human beings at the end of the
day.

493
00:30:48,500 --> 00:30:52,500
And so I think, I think it's
really important, we keep that

494
00:30:53,900 --> 00:30:58,000
Vision and that inspiration and
that ideal at the Forefront of

495
00:30:58,000 --> 00:31:00,000
this movement.
So what that guy said?

496
00:31:01,600 --> 00:31:04,000
Yes and I can see got it.
Like you are on the positive

497
00:31:04,000 --> 00:31:09,400
side of the thinking, right?
So you're optimistic Yeah, I

498
00:31:09,400 --> 00:31:11,300
would, I would say that
generally speaking.

499
00:31:13,000 --> 00:31:18,900
The, the realization that that's
allowed me to be optimistic, is

500
00:31:18,900 --> 00:31:23,700
to look at the writings of
earlier centuries and to see

501
00:31:23,700 --> 00:31:27,000
what blind spots that people had
when they were.

502
00:31:27,000 --> 00:31:34,500
So, so confident, and to see how
much better we are off right

503
00:31:34,500 --> 00:31:38,700
now, and to see how many of
those pessimists were wrong has

504
00:31:38,800 --> 00:31:43,300
So as I think, luckily happen
enough times that there's a

505
00:31:43,300 --> 00:31:46,500
pattern, we can recognize and
that pattern can be extrapolated

506
00:31:46,500 --> 00:31:48,200
without just Faith.
Like like, you know, I think

507
00:31:48,200 --> 00:31:51,700
that there's questions we need
to have about gray goo and what,

508
00:31:51,700 --> 00:31:54,400
if it isn't super, replicator
Ai, and, and Cetera.

509
00:31:54,500 --> 00:31:57,600
Anything that there's there's a
rational logical reason to be

510
00:31:57,800 --> 00:32:01,300
optimistic even about these
more, esoteric problems.

511
00:32:01,300 --> 00:32:05,100
But fundamentally I think
there's a lot to be optimistic

512
00:32:05,100 --> 00:32:08,300
about on many dimensions.
I'm on your side and just, you

513
00:32:08,300 --> 00:32:10,300
know, too.
I would say complement, what you

514
00:32:10,300 --> 00:32:15,600
mentioned garak.
I believe I will because you,

515
00:32:15,800 --> 00:32:19,500
you touch on that and you said,
maybe either we all be in an

516
00:32:19,500 --> 00:32:22,300
abundance.
And I believe this is this is

517
00:32:22,300 --> 00:32:24,000
the case.
I believe all of us will be in

518
00:32:24,000 --> 00:32:27,600
abundance, life will not be the
same.

519
00:32:27,600 --> 00:32:31,900
We are living it now, of course
it will take maybe hundreds of

520
00:32:31,900 --> 00:32:36,100
years, I'm not sure I'm hoping
that it's less whereas because

521
00:32:36,300 --> 00:32:39,600
the reason I'm saying this
because if anyone You know, and

522
00:32:39,700 --> 00:32:42,500
we saw life examples, maybe it
was like a joke.

523
00:32:42,500 --> 00:32:46,000
We saw this guy who went viral
on Twitter and he did something

524
00:32:46,200 --> 00:32:50,300
he called it, the hustle GPT.
Right.

525
00:32:50,300 --> 00:32:52,900
But guess what?
I have people that I know that

526
00:32:53,100 --> 00:32:55,700
we follow each other's.
Even we were in cohort together.

527
00:32:56,000 --> 00:32:59,400
They did it.
I mean okay it was a small scale

528
00:32:59,600 --> 00:33:03,400
but they did something and get
them 500 bucks and you know they

529
00:33:03,400 --> 00:33:05,500
said okay let's try to get 5,000
bucks.

530
00:33:05,700 --> 00:33:08,300
Of course if now you go and type
okay I need this site has some

531
00:33:08,300 --> 00:33:11,100
that Make me 1 million dollar,
it will not work.

532
00:33:11,100 --> 00:33:13,300
And actually chat GPT will
answer you.

533
00:33:13,300 --> 00:33:15,500
I think they tweaked something
and to tell if there you are,

534
00:33:15,500 --> 00:33:19,900
man, I'm just, I language model
and I cannot give you any

535
00:33:20,100 --> 00:33:24,400
Financial advice.
But I mean, this showed us

536
00:33:24,400 --> 00:33:26,200
again, I call it proof of
concept, right?

537
00:33:26,200 --> 00:33:28,700
So there's something called
proof of concept that.

538
00:33:28,700 --> 00:33:32,200
Yes, a, I can generate ideas.
Let me give you one more

539
00:33:32,600 --> 00:33:35,100
example.
And actually, I was planning to

540
00:33:35,100 --> 00:33:40,700
do even a Hey, they're more like
maybe walk through about it two

541
00:33:40,700 --> 00:33:44,800
days back and you know like you
and I we share this geeky thing.

542
00:33:44,800 --> 00:33:48,600
So I looked at you know by
office here, I have a Raspberry

543
00:33:48,600 --> 00:33:53,700
Pi and they have a Arduino as a
city sitting here and you know,

544
00:33:53,700 --> 00:33:58,700
all of a sudden I said, hey, let
me go to chat GPT and ask, you

545
00:33:58,700 --> 00:34:01,600
know, give me some cool idea
that no one may be have done it.

546
00:34:01,700 --> 00:34:05,400
Of course, up to the knowledge
that they had up to 20 21, and

547
00:34:05,400 --> 00:34:07,800
guess what?
It gave me Really cool ideas

548
00:34:07,800 --> 00:34:11,199
like maybe no one will be
interested in, but when I start

549
00:34:11,199 --> 00:34:15,600
to read, oh look at this like,
you know, it's suggesting me,

550
00:34:15,900 --> 00:34:18,699
you know, the ideas are just me,
the components that I need to

551
00:34:18,699 --> 00:34:23,900
have giving me step by step, you
know, guide and actually it's

552
00:34:23,900 --> 00:34:26,400
funny enough.
It's told me I'd and hey, like

553
00:34:26,400 --> 00:34:30,100
this is just, you know, a fun
project, do it yourself.

554
00:34:30,100 --> 00:34:33,300
Kind of thing, even subject, you
know, I said, okay, what can I

555
00:34:33,308 --> 00:34:36,000
do?
He said, yeah, like, you can put

556
00:34:36,000 --> 00:34:38,699
it into a website.
Fight for hobbyists who would

557
00:34:38,699 --> 00:34:41,199
like so you can put this idea
that I'm giving you.

558
00:34:42,000 --> 00:34:44,500
Okay.
So this is the augmentation guys

559
00:34:44,500 --> 00:34:45,900
that, you know, I'm enjoying
now.

560
00:34:45,900 --> 00:34:49,300
So if you are always thinking of
this kind of take my job, you

561
00:34:49,300 --> 00:34:51,300
know, like this will end the
world.

562
00:34:51,300 --> 00:34:53,600
No nothing like this.
Nothing like this will happen,

563
00:34:53,699 --> 00:34:57,000
actually, you just need to have
fun and actually I was not very

564
00:34:57,000 --> 00:35:00,300
fan of you know, like reels and
Tick-Tock and all these things.

565
00:35:00,300 --> 00:35:04,200
But now when you go that side
you see like there are a bunch

566
00:35:04,200 --> 00:35:08,500
of people young people who are
Sugar are getting educated on AI

567
00:35:08,500 --> 00:35:11,800
and they are using it somehow
either to create content or to

568
00:35:11,800 --> 00:35:13,800
do something, which is very,
very positive.

569
00:35:14,100 --> 00:35:19,900
Now, we wanted to say something
and we kept it aside for the use

570
00:35:19,900 --> 00:35:28,500
cases Garrick like the most, I
mean close ones that we can see

571
00:35:28,600 --> 00:35:32,100
in the near future like where we
going to see the changes fast.

572
00:35:32,100 --> 00:35:34,600
And when I say fast we're not
talking about yes.

573
00:35:34,600 --> 00:35:36,500
Maybe in the coming six to 12
months.

574
00:35:36,800 --> 00:35:40,500
And then down the road how you
can break down that to us.

575
00:35:41,400 --> 00:35:45,900
Yeah.
So one thing that's going to be

576
00:35:45,900 --> 00:35:49,700
coming out very soon is I think
we're going to have a lot of

577
00:35:49,708 --> 00:35:54,500
companies that are going to be
producing better interfaces with

578
00:35:54,800 --> 00:36:00,500
with where this technology.
So if you start if you start

579
00:36:00,500 --> 00:36:03,600
using chatter BT not just for
fun but actually for your work,

580
00:36:03,800 --> 00:36:09,600
you'll often find that the You
know, the hard part was figuring

581
00:36:09,600 --> 00:36:12,600
out how to describe the problem,
who would have thought at the

582
00:36:12,608 --> 00:36:14,100
questions was harder than the
answer.

583
00:36:15,000 --> 00:36:17,800
I asked you to make a big
report, but telling it what to

584
00:36:17,800 --> 00:36:21,300
make a report on, who's it for,
what's the relevant, you know,

585
00:36:21,900 --> 00:36:24,600
factors?
That's the hard part.

586
00:36:24,700 --> 00:36:26,300
So I think we're going to be
creating better and better

587
00:36:26,300 --> 00:36:31,400
interfaces to reduce the effort
of those parts.

588
00:36:32,400 --> 00:36:35,700
So there's big communities right
now that are trading and

589
00:36:35,700 --> 00:36:39,300
selling.
Prompts the, they call

590
00:36:39,300 --> 00:36:41,700
themselves prompts engineers and
prompt marketplaces.

591
00:36:42,000 --> 00:36:45,900
I think that that is just the
very, very early days.

592
00:36:45,900 --> 00:36:49,200
And probably will look nothing
like a Marketplace in the coming

593
00:36:49,200 --> 00:36:53,200
years, but I think a lot of
people going to make a pretty

594
00:36:53,200 --> 00:36:55,500
penny, creating better
interfaces.

595
00:36:55,500 --> 00:37:00,400
I'm thinking of like, Jasper, Ai
and writer Ai, and those tools,

596
00:37:00,400 --> 00:37:03,400
it's really just as using chat,
CBT on the back end, but there

597
00:37:03,600 --> 00:37:06,600
they created an interface.
Makes it easier to extract.

598
00:37:06,800 --> 00:37:12,800
Good copy, that's number one.
Number two, is that the the a

599
00:37:12,800 --> 00:37:16,400
eyes that are being produced by
like barred with Google and

600
00:37:16,400 --> 00:37:19,200
tragic Beauty with opening a, I
are very general in their

601
00:37:19,200 --> 00:37:27,100
knowledge and a go to Tech.
Go to Startup is going to be to

602
00:37:28,300 --> 00:37:31,400
create a specialized AI that
performs very well in a

603
00:37:31,408 --> 00:37:34,700
subfield.
So think of illegal AI right now

604
00:37:34,700 --> 00:37:40,400
that the Is AI hallucinates and
you can't have that in the legal

605
00:37:40,400 --> 00:37:41,800
profession or a medical
profession.

606
00:37:41,800 --> 00:37:45,500
So by creating finely-tuned a is
that make less mistakes and are

607
00:37:45,500 --> 00:37:49,000
less generic?
I think a lot of value is going

608
00:37:49,000 --> 00:37:51,700
to be added to the market and a
lot of very successful startups

609
00:37:51,700 --> 00:37:54,500
going to be born.
I'll tell you one pet project,

610
00:37:54,500 --> 00:37:57,300
I'm excited for it, I'm excited
for AI.

611
00:37:57,600 --> 00:38:01,900
That scrapes, every piece of
text from a variety of let's

612
00:38:01,900 --> 00:38:04,700
call them gurus like mentors
like Steve Jobs, Etc.

613
00:38:05,500 --> 00:38:10,600
And then Crates on AI Persona so
that you can have a conversation

614
00:38:10,900 --> 00:38:12,900
of asking Steve Jobs.
Hey, what would you do?

615
00:38:12,900 --> 00:38:15,000
And it takes every transcript,
every conversation from that,

616
00:38:15,000 --> 00:38:18,700
perfect guy, every interview and
you can have Steve Jobs or

617
00:38:18,800 --> 00:38:22,700
anyone as your business Mentor.
That's an example of an AI that

618
00:38:22,800 --> 00:38:27,400
becomes fine-tuned and narrow
down with with the right data

619
00:38:27,400 --> 00:38:30,900
inputted.
So those are our two, there's a

620
00:38:30,900 --> 00:38:32,800
few others but those are two
fundamental lenses.

621
00:38:32,800 --> 00:38:36,500
I like to put on and advise
people to put on when thinking

622
00:38:36,700 --> 00:38:40,100
Of new companies and new
Innovations, still to be

623
00:38:40,100 --> 00:38:44,100
created.
But guys are we going to see?

624
00:38:44,300 --> 00:38:48,600
Like only like, for example,
historically, it seems like we

625
00:38:48,600 --> 00:38:51,000
are tied, 2-2, big names all the
time.

626
00:38:51,200 --> 00:38:55,800
So it was Apple Microsoft.
I don't know.

627
00:38:55,800 --> 00:39:00,000
Like we have all these two big
names.

628
00:39:00,000 --> 00:39:03,900
So here, now we have now
Microsoft in an open-air.

629
00:39:03,900 --> 00:39:09,000
I backed by yeah verses verse.
This is Google, are we going to

630
00:39:09,000 --> 00:39:12,200
see like this consolidation?
And the reason I'm asking you

631
00:39:12,200 --> 00:39:17,300
because nowadays, like even you
startups are coming up and there

632
00:39:17,300 --> 00:39:19,300
is the famous meme on the
internet.

633
00:39:19,300 --> 00:39:23,200
When they you know, when they
remove the the cover it's like

634
00:39:23,300 --> 00:39:26,900
running on top of open a iapi.
Yeah.

635
00:39:26,900 --> 00:39:30,900
Or maybe Google still didn't
release an API for board but I'm

636
00:39:30,900 --> 00:39:32,400
sure they're gonna do that very
soon.

637
00:39:32,400 --> 00:39:34,500
Yeah.
And the reason why I'm asking

638
00:39:34,500 --> 00:39:39,000
you like don't you think That
actually these two companies

639
00:39:39,000 --> 00:39:46,000
will have huge impact and
control over any startup that

640
00:39:46,000 --> 00:39:50,600
will come in the future like and
how we can tackle that actually

641
00:39:51,800 --> 00:39:54,600
that's a great point.
And you know, I think that there

642
00:39:54,600 --> 00:39:59,600
is a tendency for a lot of
companies become a lot of

643
00:39:59,600 --> 00:40:03,100
Industries become duopolies, you
know, PlayStation vs Xbox.

644
00:40:03,800 --> 00:40:10,700
So on, so forth and It's the AI
Wars are really shaping up for

645
00:40:10,700 --> 00:40:15,100
Google and Microsoft to be re
going head-to-head.

646
00:40:15,400 --> 00:40:19,100
Apple is because they like to be
very insular are not using the

647
00:40:19,100 --> 00:40:21,200
word AI in any of their press
releases.

648
00:40:21,200 --> 00:40:24,100
They're using words, machine
learning Transformers Etc.

649
00:40:24,300 --> 00:40:26,300
But they're avoiding that word
because they don't want to Hitch

650
00:40:26,300 --> 00:40:29,300
their brand to to that.
So, I think there's gonna be a

651
00:40:29,300 --> 00:40:33,200
lot of players that are pretty
big that are maybe not the

652
00:40:33,200 --> 00:40:38,300
center of attention in this, but
More importantly, I want to go

653
00:40:38,300 --> 00:40:40,800
back to what we're seeing in the
middle of today's conversation,

654
00:40:40,800 --> 00:40:44,100
which is the open source,
Community with the Innovation,

655
00:40:44,100 --> 00:40:49,300
from llama, being being leaked,
and the Innovations with

656
00:40:49,300 --> 00:40:57,600
chinchilla and low, low input
training, which is meaning that

657
00:40:57,800 --> 00:41:00,900
this technology becoming a lot
more democratized through

658
00:41:01,400 --> 00:41:04,100
through developers, building
building custom apps that are

659
00:41:04,100 --> 00:41:06,500
not tied at all to either of
these big players.

660
00:41:06,600 --> 00:41:09,800
And so I think that it's very
natural for a lot of Industries

661
00:41:09,800 --> 00:41:12,200
have duopolies and I think that
we will have something like

662
00:41:12,200 --> 00:41:15,500
that, but enough options that we
won't find ourselves in some

663
00:41:15,500 --> 00:41:20,000
sort of nightmare scenario where
a shadow Monopoly.

664
00:41:20,000 --> 00:41:25,800
Actually dominates too much.
I would have said that even

665
00:41:25,800 --> 00:41:30,000
three months ago until until the
realizations from that white

666
00:41:30,000 --> 00:41:31,600
paper.
Once again recommend you know we

667
00:41:31,600 --> 00:41:34,400
have no mote from from Google.
We're definitely recommend

668
00:41:34,400 --> 00:41:37,900
checking that out for more.
Evidence for that.

669
00:41:38,300 --> 00:41:40,400
Yeah.
And you're right.

670
00:41:40,400 --> 00:41:43,700
Like things are changing so fast
because what maybe I've said,

671
00:41:44,600 --> 00:41:48,400
even one month bag is it doesn't
apply anymore.

672
00:41:50,100 --> 00:41:53,100
You mentioned something about
coding and I'm interested like

673
00:41:53,400 --> 00:42:03,200
when we can expect a proper full
fledge code generated by because

674
00:42:03,200 --> 00:42:07,900
now you know the feedback so far
that it's Not the perfect code,

675
00:42:07,900 --> 00:42:11,600
it still need, you know, people
to go and check it back.

676
00:42:11,600 --> 00:42:14,100
Of course, always, this is why
we need, for example, quality

677
00:42:14,100 --> 00:42:17,000
engineers.
And we need like Q&A and all the

678
00:42:17,000 --> 00:42:20,200
stuff.
But like, really like, when you

679
00:42:20,200 --> 00:42:23,100
think day, I will be able to
generate proper code.

680
00:42:24,600 --> 00:42:27,400
Is it any?
Any, any near future?

681
00:42:28,600 --> 00:42:30,300
Yeah, that's, that's a fantastic
question.

682
00:42:32,400 --> 00:42:34,900
It's not going to be a single
moment, you know.

683
00:42:35,500 --> 00:42:39,100
I'm not sure if, if people think
of it this way.

684
00:42:39,100 --> 00:42:41,400
But I remember when Siri first
came out and everybody said was

685
00:42:41,400 --> 00:42:43,600
terrible, and they weren't
interests.

686
00:42:43,700 --> 00:42:45,600
And then five years later.
Siri, just works.

687
00:42:45,600 --> 00:42:47,200
And then people are like, well,
of course it works.

688
00:42:47,200 --> 00:42:49,700
It's been around for five years
and there was never a moment of

689
00:42:49,700 --> 00:42:52,500
like and you it's quote, unquote
arrived.

690
00:42:52,600 --> 00:42:58,900
And right now, the copilot tool
that Microsoft has for its a

691
00:42:58,900 --> 00:43:04,600
GitHub product already there.
Noted that I think more than 50%

692
00:43:04,600 --> 00:43:09,700
of all code committed in this
year has been written with

693
00:43:09,700 --> 00:43:14,400
copilot, which does not mean 50%
of developers are using.

694
00:43:14,400 --> 00:43:17,600
It means a 10%.
Developers are paying $8 a month

695
00:43:17,700 --> 00:43:21,000
are writing five times more code
than anyone else.

696
00:43:21,400 --> 00:43:25,200
Um, so so really what we're
looking at is still in the

697
00:43:25,200 --> 00:43:28,500
beginning stages of like we said
before, people being Amplified,

698
00:43:28,500 --> 00:43:32,400
not not a, I taking your job but
people being Amplified I'd

699
00:43:32,400 --> 00:43:35,400
becoming more productive.
And I think that that's kind of

700
00:43:35,400 --> 00:43:36,900
what I think.
That's really what's going to

701
00:43:36,900 --> 00:43:40,400
look like for for a decent
amount of time.

702
00:43:40,400 --> 00:43:44,000
And we're probably pretty far
away from the idea of you just

703
00:43:44,000 --> 00:43:45,800
typing in.
Hey, making a million-dollar,

704
00:43:46,000 --> 00:43:50,600
you know startup it's just not
going to happen that way at

705
00:43:50,600 --> 00:43:52,200
least while we're constrained by
physics.

706
00:43:52,700 --> 00:43:57,200
So, I think it's a long, a long
journey with a lot of little

707
00:43:57,200 --> 00:44:02,300
micro milestones and probably
was not gonna As an AI futurist.

708
00:44:02,300 --> 00:44:06,500
I'm not hitching.
My my planning around that

709
00:44:06,800 --> 00:44:11,200
anytime soon.
Okay, just to put context and

710
00:44:11,200 --> 00:44:16,600
maybe if there's anyone who
doesn't understand like how

711
00:44:16,600 --> 00:44:22,700
coding works and to your point
Garrick, so when when

712
00:44:23,300 --> 00:44:26,700
programming languages were
created, so they were just, you

713
00:44:26,700 --> 00:44:32,200
know, putting on top of some
Language just for the sake of a

714
00:44:32,200 --> 00:44:36,700
programmer to write a piece of
software in a human

715
00:44:36,700 --> 00:44:40,200
understandable format.
And then you know, with time the

716
00:44:40,200 --> 00:44:43,800
code, you know grew.
I mean it's start to have more

717
00:44:43,800 --> 00:44:46,500
libraries that people can reuse.
Why?

718
00:44:46,500 --> 00:44:49,200
I'm mentioning this, the reason
I'm mentioning this because this

719
00:44:49,200 --> 00:44:53,400
is remind us of how large
language models.

720
00:44:53,600 --> 00:44:56,700
If you are today you are not a
geeky guy, you are not a coder.

721
00:44:57,100 --> 00:45:03,200
So the way large language models
works is If they have a huge

722
00:45:03,200 --> 00:45:06,700
amount of data sitting
somewhere, and then when you go

723
00:45:06,700 --> 00:45:10,200
and put the question, like, even
if you are just saying, hey,

724
00:45:11,400 --> 00:45:13,900
give me a.
I don't know.

725
00:45:14,100 --> 00:45:18,200
Write me an essay about
Construction business.

726
00:45:18,200 --> 00:45:21,500
So it goes first, you know, what
information, same as you used to

727
00:45:21,500 --> 00:45:25,400
do in the past, I used to go to
books and then go to articles

728
00:45:25,400 --> 00:45:29,200
and then try to rename rewrite
it in your own way.

729
00:45:29,300 --> 00:45:33,300
So that's what AI does now,
connecting back to to code.

730
00:45:33,300 --> 00:45:37,300
So, actually, we have a library
of code that a, I can understand

731
00:45:37,300 --> 00:45:40,300
what it does.
So if I go and today say, to

732
00:45:40,300 --> 00:45:44,700
chat GPT, I want a function that
written in Python.

733
00:45:44,700 --> 00:45:48,700
I need to specify, of course the
language, I Python code that

734
00:45:48,700 --> 00:45:52,700
goes.
And for example, check a website

735
00:45:52,700 --> 00:45:58,200
if it's up or down, check how
much is the page load speed and

736
00:45:58,600 --> 00:46:04,100
put that into a specific format
output to a text file?

737
00:46:04,100 --> 00:46:06,900
For example, just because we are
thinking non-ui now.

738
00:46:07,400 --> 00:46:10,200
So, what?
What, what, what opening our HR

739
00:46:10,200 --> 00:46:15,300
G PT will do to go and find,
okay, which Functions within the

740
00:46:15,300 --> 00:46:18,500
programming language, with
python can do that and then to

741
00:46:18,500 --> 00:46:20,500
write it for you.
Maybe it's not the ultimate Way

742
00:46:20,500 --> 00:46:23,700
what Garrick mentioned is that
it requires that because they

743
00:46:23,700 --> 00:46:26,900
need to train, the more data you
have and the more you use it and

744
00:46:26,900 --> 00:46:29,800
the more feedback you gave it
becomes better because it's what

745
00:46:29,800 --> 00:46:33,500
they call a supervised model.
So basically you need to give to

746
00:46:33,500 --> 00:46:37,600
keep giving feedback to it.
So it can you know, interact

747
00:46:37,600 --> 00:46:41,800
better with you and just for the
sake of the audience here

748
00:46:41,800 --> 00:46:45,700
because I've been using Using
chat, you Petey's maybe one week

749
00:46:45,700 --> 00:46:48,900
after it went out.
The thing is that because they

750
00:46:48,900 --> 00:46:51,900
can collect data and we know
that they can collect data about

751
00:46:51,900 --> 00:46:54,900
all our interactions.
And I can tell you even with the

752
00:46:54,900 --> 00:46:59,800
same problem that I used to have
three months back or four months

753
00:46:59,800 --> 00:47:03,900
back, if I use the same ones,
now I got much better answer

754
00:47:03,900 --> 00:47:08,800
because they know better about
me and what are my needs.

755
00:47:10,100 --> 00:47:14,700
So, yeah, like it's amazing.
I want to share one.

756
00:47:15,200 --> 00:47:17,800
I want to share one one thought,
just on that.

757
00:47:17,800 --> 00:47:21,800
Point of it learning.
We're going to people talk about

758
00:47:21,800 --> 00:47:24,100
prompt engineering, becoming a
new job description.

759
00:47:24,100 --> 00:47:26,100
I think we're going to have a
new job description that future

760
00:47:26,100 --> 00:47:30,500
as well, which is, which is it?
Large language model

761
00:47:30,800 --> 00:47:36,400
optimization, which is the same
as SEO and ASO with Jess is

762
00:47:36,400 --> 00:47:40,200
going to be.
What if you you ask, hey, what's

763
00:47:40,200 --> 00:47:42,400
a pizza place word?
Give me a list of ten pizza

764
00:47:42,400 --> 00:47:45,700
places.
And you prompt it so often to

765
00:47:45,700 --> 00:47:49,600
include Mom and Pop shop in
Atlanta that all of a sudden Mom

766
00:47:49,600 --> 00:47:52,900
and Pop Shop, Atlanta shows up
in more people's queries

767
00:47:53,100 --> 00:47:55,300
already.
I have some friends who are

768
00:47:56,300 --> 00:48:00,400
working on this to influence,
other people's answers to

769
00:48:00,400 --> 00:48:05,000
questions, like how they train,
they are training the model.

770
00:48:05,000 --> 00:48:06,500
Now of course that's going to
come an arms race.

771
00:48:06,500 --> 00:48:10,500
You know, SEO never SEO is still
a work in progress, It's a

772
00:48:10,500 --> 00:48:12,500
constant dynamic system, it
doesn't, it doesn't really hit

773
00:48:12,500 --> 00:48:13,800
equilibrium.
So Beak.

774
00:48:14,000 --> 00:48:18,100
So like it's just gonna be
another, another Vector of of

775
00:48:19,100 --> 00:48:23,000
games to play, but that is going
to become another.

776
00:48:23,000 --> 00:48:27,300
I believe industry on top of SEO
and on top of prompt engineering

777
00:48:27,300 --> 00:48:31,700
and and all the future holds,
yeah, on the prompt engine

778
00:48:31,700 --> 00:48:35,300
before I forget because you
talked about, you know, people

779
00:48:35,300 --> 00:48:37,800
doing prompts and trading
prompts.

780
00:48:38,100 --> 00:48:40,800
So actually this space guys, if
you thinking to keep this

781
00:48:40,800 --> 00:48:43,000
business model, I don't advise
you at all.

782
00:48:43,400 --> 00:48:46,400
Because and I mentioned this in
a previous episode couple of

783
00:48:46,400 --> 00:48:51,300
days back.
This is being sorted out by

784
00:48:51,400 --> 00:48:54,900
itself.
So you mentioned Garrick Jasper.

785
00:48:54,900 --> 00:48:57,600
I and just very I, they have a
function.

786
00:48:57,600 --> 00:49:00,700
So if you missed that episode,
I'm repeating it now and I'm not

787
00:49:00,700 --> 00:49:03,000
affiliated with them.
I just use them as well.

788
00:49:03,500 --> 00:49:07,900
So, just where they get a really
cool function where when you

789
00:49:07,900 --> 00:49:10,900
type The Prompt when you type,
what you want to do is it like a

790
00:49:10,900 --> 00:49:16,300
blog post or whatever.
Have a button now enhance it

791
00:49:16,800 --> 00:49:19,900
enhance.
So when you click on enhance

792
00:49:20,200 --> 00:49:21,300
I'll give you a very simple
example.

793
00:49:21,300 --> 00:49:25,200
If I say, I need the truth,
let's make it a bit more.

794
00:49:25,900 --> 00:49:32,500
I need, they link it in post
about sharing a new product with

795
00:49:32,500 --> 00:49:35,400
the market.
And when I write this phrase and

796
00:49:35,400 --> 00:49:39,800
I hit the enhance button, what
Jasper I would do, it will go

797
00:49:39,800 --> 00:49:43,000
change my prompt and to write
something like this.

798
00:49:43,400 --> 00:49:49,000
You are a marketing specialist
in social media and I want you

799
00:49:49,000 --> 00:49:52,300
to write, you know, like it make
it much longer the way that you

800
00:49:52,300 --> 00:49:54,400
should interact actually with
large language models.

801
00:49:54,800 --> 00:49:59,400
And when I saw this what?
Like, yeah, we don't think we

802
00:49:59,400 --> 00:50:01,800
know we don't need to enhance
the prompts anymore.

803
00:50:01,800 --> 00:50:03,800
I mean, there is a an AI that
can do this.

804
00:50:04,200 --> 00:50:08,400
So as we are coming with me,
yeah, I think, I think there's a

805
00:50:08,400 --> 00:50:11,300
lot of, a lot to be said there,
I know we're wrapping up but I

806
00:50:11,300 --> 00:50:15,100
think there's a prompt.
Engineering is definitely a lot

807
00:50:15,100 --> 00:50:19,000
of Transformations.
Fundamentally we're still

808
00:50:19,100 --> 00:50:21,700
constrained by physics and then
different skill sets.

809
00:50:22,000 --> 00:50:25,100
Put some people in better
situations and others so I think

810
00:50:25,100 --> 00:50:28,000
that I don't know if I would say
it's wasted effort but certainly

811
00:50:28,000 --> 00:50:30,100
is not going to be anything like
what it is right now.

812
00:50:30,100 --> 00:50:35,300
I completely agree with that.
100% one thing I want to ask you

813
00:50:35,300 --> 00:50:37,800
guys, let me maybe people now
who are listening to us.

814
00:50:37,800 --> 00:50:39,200
They're saying, hey, these two
guys.

815
00:50:39,200 --> 00:50:41,700
What are they?
What they are talking about

816
00:50:41,700 --> 00:50:43,200
that.
We are interested in the eye.

817
00:50:43,300 --> 00:50:45,800
We want to get involved this in
this day.

818
00:50:45,800 --> 00:50:49,600
I think what a second from a
career perspective.

819
00:50:49,600 --> 00:50:52,400
Maybe they want to start a
company or maybe it's just a

820
00:50:52,400 --> 00:50:55,300
hobby.
So, what advice or resources?

821
00:50:55,300 --> 00:50:57,700
Would you recommend for these
individuals?

822
00:50:58,700 --> 00:51:03,800
Yeah, so the thing that I tell
everybody who is interested

823
00:51:03,800 --> 00:51:10,300
going to AI is to check out
playground dot, open a i.com,

824
00:51:10,300 --> 00:51:15,800
which is where they show off the
the Hidden variables about how

825
00:51:15,800 --> 00:51:18,900
their large language model
actually accepts inputs.

826
00:51:19,000 --> 00:51:22,200
So if you're just doing chat gbt
you're missing, big chunks of

827
00:51:22,200 --> 00:51:24,300
story.
So I would say it's just most

828
00:51:24,300 --> 00:51:27,200
people who want to get, maybe
not fully technical, maybe they

829
00:51:27,200 --> 00:51:30,000
don't want to code, but they do
want to understand the AI on a,

830
00:51:30,200 --> 00:51:33,000
on a more intimate level than
they would just by playing

831
00:51:33,000 --> 00:51:36,200
around with it to check out
playground dot, open Ai.

832
00:51:36,900 --> 00:51:40,100
And if you're interested in
starting a company, my company

833
00:51:40,100 --> 00:51:43,100
of a hollowed-out team is right
now, looking for partners.

834
00:51:43,300 --> 00:51:47,200
To start new Ventures.
Our mission is to be creating a,

835
00:51:47,200 --> 00:51:51,800
i based SAS companies to build
up and then exit from in two to

836
00:51:51,800 --> 00:51:53,700
three years.
So we're looking for partners

837
00:51:53,700 --> 00:51:58,600
and businesses, so anyone who
wants to can reach out to us,

838
00:51:58,900 --> 00:52:01,900
but if you're just starting as a
as a hobbyist, I definitely

839
00:52:01,900 --> 00:52:05,500
recommend reading more of the
open, a.i. documentation playing

840
00:52:05,500 --> 00:52:07,600
their playground great.
It's good.

841
00:52:07,600 --> 00:52:11,300
That you mentioned this because
this something I asked, so I

842
00:52:11,300 --> 00:52:13,100
gotta put, you know, the
website.

843
00:52:14,800 --> 00:52:17,600
Of the company in the episode
description and of course I will

844
00:52:17,600 --> 00:52:22,300
put also the profile access to
Garrick so you can get in touch

845
00:52:22,300 --> 00:52:25,200
with him Garrick.
Like this is my famous last

846
00:52:25,200 --> 00:52:28,700
question nowadays.
What is the question that you

847
00:52:28,700 --> 00:52:32,300
wished me to ask you and how you
would answer it?

848
00:52:33,100 --> 00:52:36,400
Oh that's a fantastic question
minute.

849
00:52:36,400 --> 00:52:38,100
I felt like you were a fantastic
interviewer.

850
00:52:38,100 --> 00:52:41,600
Let me just let me just go back
to you and Cam went this was a

851
00:52:41,600 --> 00:52:46,400
lot of fun and I I decided to go
a lot more technical than I

852
00:52:46,400 --> 00:52:51,500
normally do with the CTO show
and you were fantastic to talk

853
00:52:51,500 --> 00:52:54,000
with an exchange with so great.
Great job.

854
00:52:54,200 --> 00:52:58,600
Thank you, my pleasure.
The question of what I wish you

855
00:52:58,600 --> 00:53:05,000
had asked I think we covered
most every I don't know where

856
00:53:05,000 --> 00:53:11,000
the question would have been.
I love talking with a future of

857
00:53:11,000 --> 00:53:15,700
a, I think there's a lot to be
learned from how it's from

858
00:53:15,700 --> 00:53:19,400
looking at darwinian Evolution
from looking at how biology has

859
00:53:19,700 --> 00:53:23,100
has been solving similar
problems over billions of years

860
00:53:23,300 --> 00:53:27,100
and how we're now solving
similar problems with AI with

861
00:53:27,100 --> 00:53:31,200
similar results.
So maybe, maybe if we do a

862
00:53:31,200 --> 00:53:34,800
follow-up episode, we'll talk.
More about that.

863
00:53:35,300 --> 00:53:36,900
But, no, this, this has been
fantastic.

864
00:53:36,900 --> 00:53:39,700
I met, and I think we pretty
much covered everything we

865
00:53:39,700 --> 00:53:43,000
wanted to okay, great.
That I'm really happy Garrick.

866
00:53:43,000 --> 00:53:47,100
Like, I'm really thrilled to
have you today on the show

867
00:53:47,300 --> 00:53:51,600
because I think you shed light
on a lot of aspects that people

868
00:53:52,100 --> 00:53:56,100
have thought, okay how the
future looks like and from

869
00:53:56,400 --> 00:53:59,500
healing from someone like you,
it would be really enlightening

870
00:53:59,500 --> 00:54:02,200
for the audience.
Thank you very much for being on

871
00:54:02,200 --> 00:54:03,100
the show today.
Day.

872
00:54:03,500 --> 00:54:08,100
And guys, like, as usual as I
and every episode, like, if you

873
00:54:08,100 --> 00:54:11,800
have any questions to Garrick,
or to me, you can reach out to

874
00:54:11,800 --> 00:54:14,400
us, for me.
You can reach out to me by

875
00:54:14,400 --> 00:54:17,400
email, LinkedIn, or Twitter,
where I'm the most active.

876
00:54:17,800 --> 00:54:21,200
If it's the first time you are
watching this on YouTube, you

877
00:54:21,200 --> 00:54:24,300
can subscribe to the channel.
We are always having, you know,

878
00:54:24,300 --> 00:54:28,400
Superior content, like, great
interviews with CEOs, subject

879
00:54:28,400 --> 00:54:31,400
matter experts and you are
hearing about all the new

880
00:54:31,400 --> 00:54:35,900
technology from AI.
To emerging Technologies and all

881
00:54:35,900 --> 00:54:38,200
the rest.
And if you are listening on your

882
00:54:38,200 --> 00:54:40,900
favorite protesting platform,
also don't forget to leave us a

883
00:54:40,908 --> 00:54:43,700
review and also subscribe to the
podcast.

884
00:54:43,800 --> 00:54:46,400
And if you are interested to be
like Garrick today to be a

885
00:54:46,400 --> 00:54:49,300
guest, I would love to discuss
this with you.

886
00:54:49,300 --> 00:54:52,700
Like this is how actually we are
growing the show more and more

887
00:54:53,000 --> 00:54:55,600
and I will be discussing that
with you, one to one.

888
00:54:55,600 --> 00:54:58,300
Of course we can choose the
topic the form at the time.

889
00:54:58,300 --> 00:55:00,200
The date can arrange for all of
that.

890
00:55:00,600 --> 00:55:05,600
And don't forget also to Leave
us a review with appreciate this

891
00:55:05,700 --> 00:55:07,800
and until we meet in next
episode.

892
00:55:07,800 --> 00:55:09,800
Thank you very much, take care.
Bye bye.

893
00:55:11,500 --> 00:55:12,300
Fantastic.
Thank you.

894
00:55:12,300 --> 00:55:15,500
Meant that subscribe button.
Share the show with your

895
00:55:15,500 --> 00:55:18,800
tech-savvy friends and fellow
entrepreneurs and leave us a

896
00:55:18,808 --> 00:55:21,200
review on your favorite podcast
app.

897
00:55:21,200 --> 00:55:23,900
Your support means the world to
us.