1
00:00:00,000 --> 00:00:02,120
Welcome to another deep dive.

2
00:00:02,120 --> 00:00:05,360
Today, we're going deep into AI,

3
00:00:05,360 --> 00:00:08,800
specifically what a Chinese startup called DeepSeek is doing.

4
00:00:08,800 --> 00:00:11,480
Yeah, the article you sent me about them claiming

5
00:00:11,480 --> 00:00:14,160
they can go up against Google and chat GPT.

6
00:00:14,160 --> 00:00:15,000
Yeah, DeepSeek.

7
00:00:15,000 --> 00:00:17,120
But for way less money.

8
00:00:17,120 --> 00:00:19,780
It's a big deal, a real turning point, I think.

9
00:00:19,780 --> 00:00:21,240
It's not even just about the money, though.

10
00:00:21,240 --> 00:00:24,560
It's about the whole AI industry could change because of this.

11
00:00:24,560 --> 00:00:25,760
Yeah, no, totally.

12
00:00:25,760 --> 00:00:29,240
Like, okay, so DeepSeek is saying their AI can do

13
00:00:29,240 --> 00:00:32,280
what open AI does, the people who made chat GPT.

14
00:00:32,280 --> 00:00:36,200
But they say they can make it and run it for 96% less.

15
00:00:36,200 --> 00:00:37,280
That's insane.

16
00:00:37,280 --> 00:00:38,280
Yeah, it's huge.

17
00:00:38,280 --> 00:00:41,360
It's not just like they found a better deal on parts or something.

18
00:00:41,360 --> 00:00:43,680
It's like you said, a whole different way

19
00:00:43,680 --> 00:00:44,960
of thinking about AI development.

20
00:00:44,960 --> 00:00:47,320
The article said they spent something like $6 million

21
00:00:47,320 --> 00:00:48,280
training their model.

22
00:00:48,280 --> 00:00:51,480
Open AI is spending, what, like $3 billion a year?

23
00:00:51,480 --> 00:00:52,840
That's crazy.

24
00:00:52,840 --> 00:00:54,240
How is that even possible?

25
00:00:54,240 --> 00:00:56,480
Well, that's the million dollar question, right?

26
00:00:56,480 --> 00:00:59,880
But the article suggests they might be using this technique

27
00:00:59,880 --> 00:01:01,880
called distillation.

28
00:01:01,880 --> 00:01:03,960
Think of it like this.

29
00:01:03,960 --> 00:01:07,680
You know how sometimes you can recreate a dish

30
00:01:07,680 --> 00:01:09,160
just by tasting it?

31
00:01:09,160 --> 00:01:11,240
You figure out the ingredients, the method,

32
00:01:11,240 --> 00:01:13,080
without having the actual recipe.

33
00:01:13,080 --> 00:01:13,920
Okay, yeah, I eat that.

34
00:01:13,920 --> 00:01:16,040
So distillation is kind of like that for AI.

35
00:01:16,040 --> 00:01:17,880
They're analyzing existing models,

36
00:01:17,880 --> 00:01:20,000
like figuring out how they work.

37
00:01:20,000 --> 00:01:23,080
Then they use that knowledge to train their own model,

38
00:01:23,080 --> 00:01:24,840
but in a much more efficient way.

39
00:01:24,840 --> 00:01:27,400
So they're basically like reverse engineering

40
00:01:27,400 --> 00:01:29,080
the best AI out there.

41
00:01:29,080 --> 00:01:29,920
Pretty much, yeah.

42
00:01:29,920 --> 00:01:33,520
There's even speculation that they might be building off

43
00:01:33,520 --> 00:01:36,560
open AI as GPT-4, you know, the latest one,

44
00:01:36,560 --> 00:01:38,640
and doing it without permission, which, well, you know.

45
00:01:38,640 --> 00:01:40,120
That's pretty bold.

46
00:01:40,120 --> 00:01:41,320
So Deep Sigs basically saying,

47
00:01:41,320 --> 00:01:44,480
we can make what you made, but for way cheaper.

48
00:01:44,480 --> 00:01:45,320
Yeah, exactly.

49
00:01:45,320 --> 00:01:46,960
And that's what has everyone so spooked.

50
00:01:46,960 --> 00:01:48,640
This isn't just some small improvement.

51
00:01:48,640 --> 00:01:51,600
This is like shaking the foundations of Silicon Valley.

52
00:01:51,600 --> 00:01:52,440
I bet.

53
00:01:52,440 --> 00:01:55,640
And I'm sure that if you're a company like Nvidia, right?

54
00:01:55,640 --> 00:01:56,480
You need a chip maker.

55
00:01:56,480 --> 00:01:59,240
They make the chips that power a lot of this AI stuff.

56
00:01:59,240 --> 00:02:00,080
Right.

57
00:02:00,080 --> 00:02:03,200
Their stock took a massive hit when this news came out.

58
00:02:03,200 --> 00:02:05,760
People starting to rethink like the whole future

59
00:02:05,760 --> 00:02:07,400
of AI development because of this.

60
00:02:07,400 --> 00:02:08,800
Yeah, for sure.

61
00:02:08,800 --> 00:02:10,120
The whole landscape is shifting.

62
00:02:10,120 --> 00:02:10,960
Okay, hold on.

63
00:02:10,960 --> 00:02:12,800
The article mentioned this open source thing.

64
00:02:12,800 --> 00:02:14,480
What does that even mean for AI?

65
00:02:14,480 --> 00:02:16,360
Like break that down for me.

66
00:02:16,360 --> 00:02:18,560
Sure, it's important to understand that open source

67
00:02:18,560 --> 00:02:21,600
basically means the code behind the AI is free.

68
00:02:21,600 --> 00:02:23,800
Anyone can download it, use it, modify it.

69
00:02:23,800 --> 00:02:27,080
It's the opposite of how things have been done traditionally

70
00:02:27,080 --> 00:02:31,360
where companies guard their AI models like state secrets.

71
00:02:31,360 --> 00:02:32,920
That's like a recipe anyone can use.

72
00:02:32,920 --> 00:02:34,280
Yeah, exactly.

73
00:02:34,280 --> 00:02:37,760
Deep seek making their model open source is a huge shift.

74
00:02:37,760 --> 00:02:42,720
So okay, on one hand, you've got this rival rebrewing,

75
00:02:42,720 --> 00:02:44,640
especially between the US and China.

76
00:02:44,640 --> 00:02:46,920
The article even mentioned Trump calling this

77
00:02:46,920 --> 00:02:48,880
a wake up call for US tech,

78
00:02:48,880 --> 00:02:50,440
but then you've got this open source thing,

79
00:02:50,440 --> 00:02:53,160
which could level the playing field.

80
00:02:53,160 --> 00:02:56,160
It feels like two opposite things happening at once.

81
00:02:56,160 --> 00:02:58,160
It's a really interesting dynamic, right?

82
00:02:58,160 --> 00:02:59,840
I mean, on the one hand,

83
00:02:59,840 --> 00:03:04,160
the US is trying to control what tech gets exported to China.

84
00:03:04,160 --> 00:03:07,640
They want to slow them down basically.

85
00:03:07,640 --> 00:03:08,720
But on the other hand,

86
00:03:08,720 --> 00:03:11,080
some experts think these export controls

87
00:03:11,080 --> 00:03:13,080
might actually backfire.

88
00:03:13,080 --> 00:03:13,920
How so?

89
00:03:13,920 --> 00:03:16,200
Well, if you can't buy the ingredients you need,

90
00:03:16,200 --> 00:03:17,480
what do you do?

91
00:03:17,480 --> 00:03:19,080
You figure out how to make them yourself.

92
00:03:19,080 --> 00:03:20,360
Exactly.

93
00:03:20,360 --> 00:03:22,800
These restrictions could push deep seek

94
00:03:22,800 --> 00:03:26,160
and other Chinese companies to become even more innovative.

95
00:03:26,160 --> 00:03:27,760
They'll have to find their own solutions,

96
00:03:27,760 --> 00:03:29,240
their own ways of doing things.

97
00:03:29,240 --> 00:03:30,800
So instead of holding them back,

98
00:03:30,800 --> 00:03:32,160
it could push them forward.

99
00:03:32,160 --> 00:03:33,000
Wow.

100
00:03:33,000 --> 00:03:33,840
Wow.

101
00:03:33,840 --> 00:03:35,200
So where does that leave deep seek?

102
00:03:35,200 --> 00:03:37,640
Are they like the underdog here?

103
00:03:37,640 --> 00:03:39,760
Or are they the ones to beat now?

104
00:03:39,760 --> 00:03:40,600
That's the thing.

105
00:03:40,600 --> 00:03:42,040
It's really hard to say.

106
00:03:42,040 --> 00:03:43,200
They're kind of both, right?

107
00:03:43,200 --> 00:03:44,520
They're relatively new to this,

108
00:03:44,520 --> 00:03:46,600
but they're already challenging the giants.

109
00:03:46,600 --> 00:03:48,200
And this open source approach

110
00:03:48,200 --> 00:03:50,040
could be a real game changer.

111
00:03:50,040 --> 00:03:53,080
It might not just be about who has the best AI anymore.

112
00:03:53,080 --> 00:03:55,280
It could be about who makes it the most accessible.

113
00:03:55,280 --> 00:03:57,520
It's like the whole industry is on its toes right now.

114
00:03:57,520 --> 00:03:59,960
Yeah, everyone's watching to see what happens next.

115
00:03:59,960 --> 00:04:01,960
It's gonna be interesting, that's for sure.

116
00:04:01,960 --> 00:04:02,800
Okay, so let's go back

117
00:04:02,800 --> 00:04:05,080
to that open source thing for a sec.

118
00:04:05,080 --> 00:04:06,680
It sounds like a pretty big deal.

119
00:04:06,680 --> 00:04:08,800
Like how big of a change could this really be?

120
00:04:08,800 --> 00:04:10,880
It could be massive, honestly.

121
00:04:11,880 --> 00:04:13,160
Think of it this way.

122
00:04:13,160 --> 00:04:16,080
You have this amazing recipe,

123
00:04:16,080 --> 00:04:18,240
only a few chefs know it, right?

124
00:04:18,240 --> 00:04:19,480
Super secret.

125
00:04:19,480 --> 00:04:23,400
But then someone posts it online for free.

126
00:04:23,400 --> 00:04:25,120
Anyone can make it now.

127
00:04:25,120 --> 00:04:28,440
So you're saying more people having access to AI,

128
00:04:28,440 --> 00:04:30,160
like it could change things.

129
00:04:30,160 --> 00:04:32,960
Oh yeah, it's not just about making the same stuff cheaper.

130
00:04:32,960 --> 00:04:35,480
It's about who gets to create, you know.

131
00:04:35,480 --> 00:04:37,640
Imagine researchers, startups,

132
00:04:37,640 --> 00:04:40,040
people who couldn't afford this tech before.

133
00:04:40,040 --> 00:04:41,920
Suddenly they can do all sorts of new things.

134
00:04:41,920 --> 00:04:43,160
Okay, that makes sense.

135
00:04:43,160 --> 00:04:46,160
But the article kept mentioning this distillation thing.

136
00:04:46,160 --> 00:04:47,800
I'm still not totally clear on that.

137
00:04:47,800 --> 00:04:49,400
Like how does that actually work?

138
00:04:49,400 --> 00:04:51,040
Okay, so imagine you're trying to teach someone

139
00:04:51,040 --> 00:04:53,520
to paint like, I don't know, Picasso, right?

140
00:04:53,520 --> 00:04:55,520
You wouldn't just give them a brush and say go for it.

141
00:04:55,520 --> 00:04:57,760
You'd show them how Picasso mixed colors,

142
00:04:57,760 --> 00:05:00,160
the strokes he used, the whole technique.

143
00:05:00,160 --> 00:05:02,360
Distillation is kind of like that for AI.

144
00:05:02,360 --> 00:05:04,920
You take an existing model, a really good one.

145
00:05:04,920 --> 00:05:07,520
You analyze how it processes information,

146
00:05:07,520 --> 00:05:09,600
how it thinks, basically.

147
00:05:09,600 --> 00:05:13,480
Then you teach a new model to do the same thing,

148
00:05:13,480 --> 00:05:14,720
but faster, cheaper, you know?

149
00:05:14,720 --> 00:05:16,800
So DeepSeek is using this distillation

150
00:05:16,800 --> 00:05:21,160
to make a model that's as good as like GPT-4,

151
00:05:21,160 --> 00:05:22,680
but without spending billions.

152
00:05:22,680 --> 00:05:23,520
That's the idea.

153
00:05:23,520 --> 00:05:25,400
They're leveraging the work that's already been done,

154
00:05:25,400 --> 00:05:27,400
but finding shortcuts, basically.

155
00:05:27,400 --> 00:05:31,000
But isn't that like kind of ethically questionable?

156
00:05:31,000 --> 00:05:32,920
If they're building off someone else's AI,

157
00:05:32,920 --> 00:05:34,440
I mean, is that even allowed?

158
00:05:34,440 --> 00:05:35,520
That's a great question.

159
00:05:35,520 --> 00:05:37,680
It's a bit of a gray area, legally speaking.

160
00:05:37,680 --> 00:05:41,000
The whole debate around distillation is just starting, really.

161
00:05:41,000 --> 00:05:43,280
Some people say if the original AI was open source,

162
00:05:43,280 --> 00:05:46,440
then it's fair game, but others argue it's more complicated.

163
00:05:46,440 --> 00:05:49,320
Especially if DeepSeek is profiting from it.

164
00:05:49,320 --> 00:05:50,880
This whole thing is just moving so fast.

165
00:05:50,880 --> 00:05:52,080
I feel like it's hard to keep up.

166
00:05:52,080 --> 00:05:53,000
It really is.

167
00:05:53,000 --> 00:05:56,760
The pace of AI development is just insane right now.

168
00:05:56,760 --> 00:05:59,640
This DeepSeek situation is like a perfect example.

169
00:05:59,640 --> 00:06:01,760
It's forcing us to rethink everything.

170
00:06:01,760 --> 00:06:03,920
How we make AI, who gets to use it,

171
00:06:03,920 --> 00:06:05,880
what it even means for society.

172
00:06:05,880 --> 00:06:06,960
Right, right.

173
00:06:06,960 --> 00:06:09,760
Speaking of society, like the article mentioned

174
00:06:09,760 --> 00:06:13,360
how DeepSeek trains its models to avoid certain topics.

175
00:06:13,360 --> 00:06:16,960
That could actually reflect a very specific worldview.

176
00:06:16,960 --> 00:06:18,280
Like a Chinese one.

177
00:06:18,280 --> 00:06:19,680
That's a really important point.

178
00:06:19,680 --> 00:06:20,840
And it's easy to forget.

179
00:06:20,840 --> 00:06:25,680
We tend to think of AI as this neutral thing, objective.

180
00:06:25,680 --> 00:06:26,880
But it's made by people, right?

181
00:06:26,880 --> 00:06:28,120
And people have biases.

182
00:06:28,120 --> 00:06:30,080
So even if the code is open, the AI itself

183
00:06:30,080 --> 00:06:31,800
might not be totally open.

184
00:06:31,800 --> 00:06:32,640
Exactly.

185
00:06:32,640 --> 00:06:34,400
It's this weird paradox.

186
00:06:34,400 --> 00:06:36,920
Open sourcing AI could democratize things.

187
00:06:36,920 --> 00:06:38,000
That's true.

188
00:06:38,000 --> 00:06:41,280
But we can't forget about the potential for bias, control,

189
00:06:41,280 --> 00:06:42,120
all that.

190
00:06:42,120 --> 00:06:44,240
Even in systems that seem totally open.

191
00:06:44,240 --> 00:06:46,960
Man, it feels like we're just scratching the surface here.

192
00:06:46,960 --> 00:06:49,040
What do you think the next five years of AI

193
00:06:49,040 --> 00:06:50,120
are going to look like?

194
00:06:50,120 --> 00:06:51,640
It's hard to say for sure.

195
00:06:51,640 --> 00:06:53,720
But if I had to guess, I think we'll

196
00:06:53,720 --> 00:06:56,360
see even more crazy innovation.

197
00:06:56,360 --> 00:06:58,400
More things that nobody saw coming.

198
00:06:58,400 --> 00:07:01,320
More competition, especially between the US and China.

199
00:07:01,320 --> 00:07:03,520
Yeah, this whole AI race is just heating up.

200
00:07:03,520 --> 00:07:04,960
And the stakes are huge.

201
00:07:04,960 --> 00:07:06,680
It's not just about technology anymore.

202
00:07:06,680 --> 00:07:09,040
It's about power, global influence,

203
00:07:09,040 --> 00:07:11,200
the whole future of how we live and work.

204
00:07:11,200 --> 00:07:12,800
So what about this open source thing?

205
00:07:12,800 --> 00:07:15,000
Is DeepSeq just an outlier?

206
00:07:15,000 --> 00:07:17,320
Or do you think other companies will follow suit?

207
00:07:17,320 --> 00:07:19,440
That's the big question, right?

208
00:07:19,440 --> 00:07:21,640
I think DeepSeq's success has definitely

209
00:07:21,640 --> 00:07:22,840
gotten people's attention.

210
00:07:22,840 --> 00:07:24,040
It's like they're showing everyone

211
00:07:24,040 --> 00:07:25,800
a different way of doing things.

212
00:07:25,800 --> 00:07:29,480
And as AI gets more expensive to develop the old way,

213
00:07:29,480 --> 00:07:31,800
open source might start to look more and more appealing.

214
00:07:31,800 --> 00:07:34,000
Like a cheaper, faster way to compete.

215
00:07:34,000 --> 00:07:35,480
Yeah, but it's more than that.

216
00:07:35,480 --> 00:07:38,440
Open source AI represents a different philosophy.

217
00:07:38,440 --> 00:07:41,280
It's about collaboration, sharing knowledge,

218
00:07:41,280 --> 00:07:42,920
making things accessible.

219
00:07:42,920 --> 00:07:45,240
It's almost like a social movement in a way.

220
00:07:45,240 --> 00:07:46,760
Wow, that's really interesting.

221
00:07:46,760 --> 00:07:50,120
So this DeepSeq thing is about more than just one company, huh?

222
00:07:50,120 --> 00:07:50,800
Way more.

223
00:07:50,800 --> 00:07:52,800
It's a reflection of the bigger picture.

224
00:07:52,800 --> 00:07:55,840
Globalization, new superpowers emerging,

225
00:07:55,840 --> 00:07:58,280
the whole way technology is changing the world.

226
00:07:58,280 --> 00:08:00,080
AI is at the center of all that.

227
00:08:00,080 --> 00:08:03,080
And DeepSeq just threw a big wrench into the gears.

228
00:08:03,080 --> 00:08:04,520
So we've talked about a lot, huh?

229
00:08:04,520 --> 00:08:08,240
DeepSeq's tech, this whole US-China thing, open source.

230
00:08:08,240 --> 00:08:10,560
What's the thing that really stood out to you from all this?

231
00:08:10,560 --> 00:08:13,200
I keep thinking about who this could change things for,

232
00:08:13,200 --> 00:08:14,200
you know?

233
00:08:14,200 --> 00:08:18,080
Like if DeepSeq's model is as good as they say,

234
00:08:18,080 --> 00:08:21,520
and it's that much cheaper, it could be huge for all sorts

235
00:08:21,520 --> 00:08:24,000
of people, smaller companies, research teams,

236
00:08:24,000 --> 00:08:26,440
even just individuals, people who never

237
00:08:26,440 --> 00:08:28,960
had a chance to work with this kind of AI before,

238
00:08:28,960 --> 00:08:30,920
because it was just too expensive.

239
00:08:30,920 --> 00:08:33,160
So DeepSeq is saying, like, you don't

240
00:08:33,160 --> 00:08:35,560
need to be a Google to play in the AI game anymore.

241
00:08:35,560 --> 00:08:36,840
Yeah, exactly.

242
00:08:36,840 --> 00:08:39,000
And the article even said you can run their models

243
00:08:39,000 --> 00:08:40,640
on regular computers, basically.

244
00:08:40,640 --> 00:08:43,840
You don't need these massive super computers.

245
00:08:43,840 --> 00:08:45,520
It's really opening things up.

246
00:08:45,520 --> 00:08:49,560
But doesn't that also mean more people could misuse AI?

247
00:08:49,560 --> 00:08:51,960
If it's easier to get, it's easier to use for anything,

248
00:08:51,960 --> 00:08:52,320
right?

249
00:08:52,320 --> 00:08:53,640
Not all of it good.

250
00:08:53,640 --> 00:08:55,480
That's the other side of the coin, for sure.

251
00:08:55,480 --> 00:08:57,280
Like you give everyone a powerful tool.

252
00:08:57,280 --> 00:09:00,400
Some people will build amazing things, and others, as well.

253
00:09:00,400 --> 00:09:00,920
Not so much.

254
00:09:00,920 --> 00:09:03,320
That's where ethics, guidelines, regulations, all that.

255
00:09:03,320 --> 00:09:04,520
It becomes even more important.

256
00:09:04,520 --> 00:09:08,480
It reminds me of the early days of the internet.

257
00:09:08,480 --> 00:09:10,960
A lot of potential, but also kind of a wild west.

258
00:09:10,960 --> 00:09:12,160
Totally.

259
00:09:12,160 --> 00:09:14,640
And just like the internet, eventually,

260
00:09:14,640 --> 00:09:16,360
things settled down a bit.

261
00:09:16,360 --> 00:09:18,680
Rules came in, safeguards.

262
00:09:18,680 --> 00:09:20,400
I think the same thing will happen with AI.

263
00:09:20,400 --> 00:09:23,600
It'll take some time, probably some mistakes along the way.

264
00:09:23,600 --> 00:09:25,160
But we'll figure it out.

265
00:09:25,160 --> 00:09:26,560
So big picture.

266
00:09:26,560 --> 00:09:28,120
Where do you see all this going?

267
00:09:28,120 --> 00:09:31,000
Five years from now, what's AI going to look like?

268
00:09:31,000 --> 00:09:33,520
Man, predicting the future is always tricky,

269
00:09:33,520 --> 00:09:34,880
especially with AI.

270
00:09:34,880 --> 00:09:36,400
But if I had to guess, I think we're

271
00:09:36,400 --> 00:09:38,600
going to see a ton of new stuff, new ideas,

272
00:09:38,600 --> 00:09:41,040
new applications, stuff we haven't even thought of yet.

273
00:09:41,040 --> 00:09:43,880
Because more people will have the tools to create.

274
00:09:43,880 --> 00:09:45,400
That's pretty exciting, actually.

275
00:09:45,400 --> 00:09:48,560
What about for someone who's just starting to learn about AI?

276
00:09:48,560 --> 00:09:49,880
What would you tell them?

277
00:09:49,880 --> 00:09:51,240
Where should they start?

278
00:09:51,240 --> 00:09:52,480
Stay curious.

279
00:09:52,480 --> 00:09:53,440
That's the biggest thing.

280
00:09:53,440 --> 00:09:55,600
There are tons of resources out there.

281
00:09:55,600 --> 00:09:56,760
Try different things.

282
00:09:56,760 --> 00:09:57,480
Experiment.

283
00:09:57,480 --> 00:09:58,760
Ask questions.

284
00:09:58,760 --> 00:10:00,560
The AI world is changing all the time.

285
00:10:00,560 --> 00:10:01,920
So just keep learning.

286
00:10:01,920 --> 00:10:03,440
Great advice.

287
00:10:03,440 --> 00:10:05,880
This whole deep dive has been really eye opening, honestly.

288
00:10:05,880 --> 00:10:07,520
DeepSeek is just one example, but it

289
00:10:07,520 --> 00:10:09,920
feels like there's so much more happening out there.

290
00:10:09,920 --> 00:10:10,880
Oh, absolutely.

291
00:10:10,880 --> 00:10:12,200
It's just the beginning.

292
00:10:12,200 --> 00:10:13,840
And that's what's so cool about it, right?

293
00:10:13,840 --> 00:10:16,280
We're seeing this whole new technology unfold.

294
00:10:16,280 --> 00:10:17,840
And who knows what it'll lead to?

295
00:10:17,840 --> 00:10:19,520
Well, I think we've covered a lot of ground here.

296
00:10:19,520 --> 00:10:22,080
Any final thoughts for our listeners before we wrap up?

297
00:10:22,080 --> 00:10:23,400
Yeah, just this.

298
00:10:23,400 --> 00:10:24,960
Don't just watch this happen.

299
00:10:24,960 --> 00:10:26,240
Be a part of it.

300
00:10:26,240 --> 00:10:27,320
Learn about AI.

301
00:10:27,320 --> 00:10:29,080
Explore what you can do with it.

302
00:10:29,080 --> 00:10:31,480
And think about how it's all going to affect us.

303
00:10:31,480 --> 00:10:34,200
Because the future of AI is being written right now.

304
00:10:34,200 --> 00:10:35,680
And we all have a say in it.

305
00:10:35,680 --> 00:10:36,960
That's a great point.

306
00:10:36,960 --> 00:10:38,520
Well, thanks for all your insights today.

307
00:10:38,520 --> 00:10:40,680
This has been a fantastic conversation.

308
00:10:40,680 --> 00:10:42,120
And to everyone listening, thanks

309
00:10:42,120 --> 00:10:46,160
for joining us on this deep dive into the wild world of AI.

310
00:10:46,160 --> 00:11:02,200
Until next time, keep exploring and keep learning.