21 | Nothing Trumps this AI week, until the next one!

Welcome back to another episode of the Digital Nexus podcast.

👀 Episode Highlights:

• Apple's iOS summary feature and what it means for your notifications
• Game-changing ChatGPT functionalities and how they compare to new models like Deep Seek
• Trump’s election win and the announcement of Project Stargate, a groundbreaking tech partnership
• Tools you can’t miss: Perplexity’s Sonar, the Prompt Cowboy, and more!

Timestamp: 
0:00 - Intro and welcome back
3:20 - Trumps election win
7:00 - Discussing upcoming Aus DeFi Association events and one for GenAI Labs
8:23 - Upcoming Twitter Spaces event talking blockchina with Aus DeFi
9:51 - Apple AI notification summaries go wrong 
12:43 - Misue of AI with fake Brad Pitt
17:05 - ChatGPT Scheduled Tasks 
22:34 - Trying out Deep Seek Chat 
26:35 - Trump revokes Biden orders on AI Risk 
28:55 - Project Stargate
32:00 - Note2Map from Raymond Sun
34:50 - Discussion of AI regulations in Australia
36:36 - AI powered education and learning 
41:48 - AI and pattern matching 
44:30 - Perplexity API, Sonar
45:25 - Anthropic interview on AGI by 2027
49:38 - AI Arms race imagery 
50:55 - Prompt engineering helper, Prompt Cowboy 
54:38 - A call to action for those looking to build with AI

Links 
Digital Nexus Podcast 
https://www.digitalnexuspodcast.com/

GenAI Labs Event - Feb 11th
https://events.humanitix.com/five-ways-ai-will-revolutionise-2025 

Trump undoes an executive order, for AI companies to expose what’s under the hood 
https://techcrunch.com/2025/01/20/president-trump-repeals-bidens-ai-executive-order/

Examples of ChatGPT tasks 
https://www.linkedin.com/posts/patricia-haueiss_openai-chatgpttasks-aiproductivity-activity-7285471351392403456--lbb?utm_source=share&utm_medium=member_desktop 

Perplexity has released an API, named Sonar. 
https://techcrunch.com/2025/01/21/perplexity-launches-sonar-an-api-for-ai-search/

Prompting Engineer - great site to help people with this: 
https://prompt.fourday.ai/

New president Trump repeals Biden administration restrictions on AI innovation
https://www.reuters.com/technology/artificial-intelligence/trump-revokes-biden-executive-order-addressing-ai-risks-2025-01-21/

Future of learning 
https://www.instagram.com/reel/DFDlbrWoH2E/?igsh=MTRwc3JlOTZ6NGkzMg== 

It’s just recognising patterns - well what do you think GPs (humans do)? 
https://x.com/DaveShapi/status/1880937932833304911 

Build your own tracker 
https://www.linkedin.com/posts/raymond-sun-64576a122_launch-map-startup-activity-7287582234566447104-_USp?utm_source=share&utm_medium=member_desktop 

Trump and Project Stargate
https://www-nbcnews-com.cdn.ampproject.org/c/s/www.nbcnews.com/news/amp-video/mmvo229900357578 

Anthropic planing to release a two way voice model, following a recent WSJ interviews
https://www.wsj.com/livecoverage/stock-market-today-dow-sp500-nasdaq-live-01-21-2025/card/anthropic-ceo-says-ai-could-surpass-human-intelligence-by-2027-9tka9tjLKLalkXX8IgKA

Transcript

1
01:00:00,000 --> 01:00:01,166
We're back! I'm back!

2
01:00:01,916 --> 01:00:03,958
Episode 21, Season 2 episode 1, I don't

3
01:00:03,958 --> 01:00:04,833
know how to look at it.

4
01:00:05,083 --> 01:00:07,208
It's an episode folks and we are talking

5
01:00:07,208 --> 01:00:08,416
about a whole ton of things.

6
01:00:08,416 --> 01:00:09,791
There's been presidential elections

7
01:00:09,791 --> 01:00:11,708
galore, there's been a whole lot of stuff

8
01:00:11,708 --> 01:00:13,125
that's been changing in the lore

9
01:00:13,375 --> 01:00:15,375
and we've got some really cool things to

10
01:00:15,375 --> 01:00:16,958
show you from places like

11
01:00:16,958 --> 01:00:19,000
DeepSeek and other news.

12
01:00:19,000 --> 01:00:20,666
Yeah, OpenAI, Tropic, a lot of them have

13
01:00:20,666 --> 01:00:22,916
lots of updates, huge businesses doing

14
01:00:22,916 --> 01:00:24,125
really interesting things.

15
01:00:24,666 --> 01:00:25,500
Let's get into it.

16
01:00:30,500 --> 01:00:31,208
Happy 2025.

17
01:00:31,500 --> 01:00:31,708
I don't know if you can

18
01:00:31,708 --> 01:00:32,250
choose that, but yeah.

19
01:00:32,583 --> 01:00:33,833
I mean, with everything that's happened

20
01:00:33,833 --> 01:00:35,916
in America, I'm sure we're going to flip

21
01:00:35,916 --> 01:00:37,125
slowly back into that realm.

22
01:00:37,541 --> 01:00:39,041
Bro, it's a crazy

23
01:00:39,041 --> 01:00:40,500
start to the year so far.

24
01:00:40,750 --> 01:00:43,625
Like, we've got two funerals to go to.

25
01:00:44,250 --> 01:00:45,000
I'm sorry to hear.

26
01:00:45,458 --> 01:00:46,375
Someone from industry and

27
01:00:46,375 --> 01:00:48,625
then a close friend and stuff.

28
01:00:48,625 --> 01:00:49,708
So it's just like, wow,

29
01:00:49,708 --> 01:00:51,000
what is going on 2025?

30
01:00:52,208 --> 01:00:53,583
So not a negative shit.

31
01:00:53,791 --> 01:00:56,541
So, passed from health

32
01:00:56,541 --> 01:00:59,250
problems or mental health?

33
01:00:59,291 --> 01:01:01,000
No, one was health problems.

34
01:01:01,583 --> 01:01:03,666
The other one was just a sudden thing.

35
01:01:03,666 --> 01:01:05,333
We didn't, no one expected it.

36
01:01:06,000 --> 01:01:07,000
So, yeah.

37
01:01:08,333 --> 01:01:10,291
Hold your loved ones closer and stuff and

38
01:01:10,291 --> 01:01:11,708
it's not just all about the tech.

39
01:01:12,041 --> 01:01:13,791
There's people behind it as well.

40
01:01:14,208 --> 01:01:16,833
So something to remember out there.

41
01:01:16,833 --> 01:01:18,416
But Chris, welcome back, man.

42
01:01:18,416 --> 01:01:20,458
Yeah, it's been, see, we were talking

43
01:01:20,458 --> 01:01:21,208
about season two,

44
01:01:21,208 --> 01:01:22,333
episode one, episode two.

45
01:01:22,333 --> 01:01:22,666
Yeah, we were.

46
01:01:23,000 --> 01:01:25,041
Or episode 21, but we'll

47
01:01:25,041 --> 01:01:26,041
probably settle on a season.

48
01:01:26,041 --> 01:01:26,458
We'll make it season

49
01:01:26,458 --> 01:01:27,750
two, episode one, 21.

50
01:01:28,250 --> 01:01:29,750
Yeah, yeah, that could say it would work

51
01:01:29,750 --> 01:01:30,833
except for last week.

52
01:01:30,833 --> 01:01:32,208
It was anyway, we were going.

53
01:01:32,208 --> 01:01:33,083
Season 2.0.

54
01:01:33,083 --> 01:01:33,833
That was the intro.

55
01:01:34,333 --> 01:01:35,125
Yeah, season two intro.

56
01:01:35,125 --> 01:01:36,625
It was the midway kind of point and stuff

57
01:01:36,625 --> 01:01:37,375
because Chris wasn't here.

58
01:01:37,375 --> 01:01:38,208
This is the first time we're actually

59
01:01:38,208 --> 01:01:40,000
back again on podcast, which is great.

60
01:01:40,250 --> 01:01:40,958
So, go see ya.

61
01:01:40,958 --> 01:01:42,416
I got hit with what I can

62
01:01:42,416 --> 01:01:45,000
only describe as a major man flu.

63
01:01:45,416 --> 01:01:47,541
Oh, worse than a train then.

64
01:01:47,541 --> 01:01:48,041
Worse than train.

65
01:01:48,041 --> 01:01:48,500
Oh my gosh.

66
01:01:48,666 --> 01:01:49,000
Man flu.

67
01:01:49,000 --> 01:01:50,166
It is a proven thing.

68
01:01:50,791 --> 01:01:51,583
Man flu is real.

69
01:01:52,750 --> 01:01:54,708
Men are weak as f.

70
01:01:55,083 --> 01:01:55,333
Yeah.

71
01:01:55,666 --> 01:01:56,125
We're weak.

72
01:01:56,125 --> 01:01:56,375
We're just weak.

73
01:01:56,416 --> 01:01:58,208
We get a sick and we just like.

74
01:01:58,208 --> 01:01:58,833
Double over.

75
01:01:58,833 --> 01:01:59,541
Yeah, double over.

76
01:01:59,541 --> 01:02:00,416
It's not even pain.

77
01:02:00,666 --> 01:02:01,750
How could I describe it?

78
01:02:01,750 --> 01:02:02,833
I mean, there's no,

79
01:02:03,083 --> 01:02:04,208
this is not even, okay.

80
01:02:04,875 --> 01:02:05,875
I describe it.

81
01:02:05,875 --> 01:02:07,208
Have you, has a woman ever?

82
01:02:07,208 --> 01:02:07,750
No, no.

83
01:02:08,500 --> 01:02:10,333
We are, we're just weak.

84
01:02:10,333 --> 01:02:11,458
And then it's like, I got hit with the

85
01:02:11,458 --> 01:02:12,625
flu, an alpha, alpha

86
01:02:12,625 --> 01:02:14,041
influenza, and my head.

87
01:02:14,416 --> 01:02:14,791
Oh, it was alpha.

88
01:02:15,416 --> 01:02:16,000
It was the alpha.

89
01:02:16,500 --> 01:02:16,541
Okay.

90
01:02:16,541 --> 01:02:17,208
Not even the beta.

91
01:02:18,375 --> 01:02:20,833
The alpha of the influences and my brain

92
01:02:20,833 --> 01:02:22,416
just decided it didn't

93
01:02:22,416 --> 01:02:24,458
want to be in this realm.

94
01:02:24,750 --> 01:02:25,541
Didn't want a brain anymore.

95
01:02:25,541 --> 01:02:26,000
No, it didn't.

96
01:02:26,166 --> 01:02:27,125
It unbrained itself.

97
01:02:27,458 --> 01:02:28,541
Took me out for like five days.

98
01:02:28,541 --> 01:02:29,583
I just couldn't get out of bed.

99
01:02:29,750 --> 01:02:31,916
Now, if only during that time, I mean,

100
01:02:31,916 --> 01:02:32,541
I'm not sure if you

101
01:02:32,541 --> 01:02:33,500
were working and stuff,

102
01:02:33,500 --> 01:02:34,666
but you could just delegate

103
01:02:34,666 --> 01:02:37,125
tasks to your avatar agent.

104
01:02:37,500 --> 01:02:37,708
Yeah.

105
01:02:38,125 --> 01:02:40,708
Please get me work that pays my bills.

106
01:02:41,000 --> 01:02:43,625
It is the year of the AI agent.

107
01:02:43,916 --> 01:02:44,708
It's, right.

108
01:02:44,708 --> 01:02:45,041
It is?

109
01:02:45,333 --> 01:02:45,541
Yeah.

110
01:02:45,541 --> 01:02:46,333
I haven't heard that one yet.

111
01:02:46,625 --> 01:02:47,833
Oh, it's, yeah, yeah, yeah.

112
01:02:47,916 --> 01:02:49,375
There's quite a few that have said it.

113
01:02:49,583 --> 01:02:49,791
Yeah.

114
01:02:50,250 --> 01:02:52,000
Well, I mean, it could

115
01:02:52,000 --> 01:02:52,750
just be the algorithm.

116
01:02:53,041 --> 01:02:54,500
I've seen it once and then it's like,

117
01:02:54,500 --> 01:02:55,416
okay, we've got one.

118
01:02:55,458 --> 01:02:56,916
Let's activate in.

119
01:02:57,583 --> 01:02:59,000
Agent, agent, agent, agent.

120
01:02:59,000 --> 01:03:00,166
But it is interesting.

121
01:03:00,166 --> 01:03:01,583
There's a whole lot of stuff going on

122
01:03:01,583 --> 01:03:02,791
until the start of the

123
01:03:02,791 --> 01:03:03,833
year and we're already seeing

124
01:03:03,833 --> 01:03:04,708
exciting events.

125
01:03:04,708 --> 01:03:05,291
We're already seeing

126
01:03:05,291 --> 01:03:06,916
exciting things in the news.

127
01:03:07,666 --> 01:03:10,458
Regardless of your political dispositions

128
01:03:10,458 --> 01:03:12,625
and stuff, it is what it

129
01:03:12,625 --> 01:03:14,375
is just over in the US.

130
01:03:15,166 --> 01:03:17,500
But if we look at it just very black and

131
01:03:17,500 --> 01:03:19,083
white, there's a positive to the crypto,

132
01:03:19,166 --> 01:03:21,958
sorry, well, crypto and also AI side

133
01:03:21,958 --> 01:03:24,250
because he's repealing

134
01:03:24,250 --> 01:03:26,500
some restrictions that Biden

135
01:03:26,500 --> 01:03:28,333
was going to put around AI.

136
01:03:28,958 --> 01:03:30,791
Which is kind of scary if I think like we

137
01:03:30,791 --> 01:03:31,916
brought this up in a

138
01:03:31,916 --> 01:03:32,833
previous episode when

139
01:03:32,833 --> 01:03:35,166
we were talking about how there is a need

140
01:03:35,166 --> 01:03:37,833
for the safety and

141
01:03:37,833 --> 01:03:40,000
the security of people's

142
01:03:40,000 --> 01:03:43,625
information, data and output that these

143
01:03:43,625 --> 01:03:45,583
AI tools are creating for everyone.

144
01:03:46,166 --> 01:03:48,666
And obviously Biden introduced, I think

145
01:03:48,666 --> 01:03:49,750
it was back in 2023,

146
01:03:50,000 --> 01:03:50,958
a rule that said that

147
01:03:51,208 --> 01:03:52,625
businesses, organizations, particularly

148
01:03:52,625 --> 01:03:53,416
if they're American,

149
01:03:53,416 --> 01:03:55,291
they need to show what,

150
01:03:56,291 --> 01:03:58,833
not in its full state, but kind of like

151
01:03:58,833 --> 01:04:00,166
what magic is going under the hood.

152
01:04:00,500 --> 01:04:02,666
And if there's any manipulation to data

153
01:04:02,666 --> 01:04:03,500
and the resources,

154
01:04:03,500 --> 01:04:04,750
etc., that is giving output.

155
01:04:05,166 --> 01:04:06,625
We can't have a free for all, but.

156
01:04:06,625 --> 01:04:08,541
No, but like, because obviously there's

157
01:04:08,541 --> 01:04:09,291
IP and all the

158
01:04:09,291 --> 01:04:10,208
restrictions, they don't have to

159
01:04:10,416 --> 01:04:11,250
wear all that stuff.

160
01:04:11,250 --> 01:04:13,958
But there was something that helped to

161
01:04:13,958 --> 01:04:15,750
reduce the security and a

162
01:04:15,750 --> 01:04:16,750
risk that a lot of these

163
01:04:16,750 --> 01:04:18,125
tools are putting on people and

164
01:04:18,125 --> 01:04:19,083
businesses and organizations.

165
01:04:20,500 --> 01:04:22,208
As obviously Mr. Trump in one fast sweep

166
01:04:22,208 --> 01:04:24,083
has just gone, no, no

167
01:04:24,083 --> 01:04:25,666
one, no, no, no, no, no, no.

168
01:04:25,833 --> 01:04:26,625
Signing all of the

169
01:04:26,625 --> 01:04:28,083
bills, he's getting told.

170
01:04:28,333 --> 01:04:29,541
I'm sure that, yeah, exactly.

171
01:04:29,541 --> 01:04:31,833
Hold it up with your gigantic signature.

172
01:04:32,333 --> 01:04:34,708
I'm like, it's amazing that it is the

173
01:04:34,708 --> 01:04:36,458
same squiggles each and every time.

174
01:04:36,458 --> 01:04:37,083
I think that I mean,

175
01:04:37,083 --> 01:04:38,375
how many did he sign?

176
01:04:38,416 --> 01:04:40,708
Like, yeah, I mean, you'd get pretty.

177
01:04:42,208 --> 01:04:42,583
RSI.

178
01:04:42,916 --> 01:04:43,083
Yeah.

179
01:04:43,291 --> 01:04:44,458
Well, I really pretty

180
01:04:44,458 --> 01:04:45,833
accurate with the same things.

181
01:04:45,833 --> 01:04:46,791
You probably just layer them on.

182
01:04:46,791 --> 01:04:47,875
They're all exactly the same with your

183
01:04:47,875 --> 01:04:48,791
sign that many times.

184
01:04:48,791 --> 01:04:50,333
I mean, it's hard to tell the difference.

185
01:04:50,333 --> 01:04:51,791
They're so close to one another.

186
01:04:52,000 --> 01:04:52,958
There's no curves.

187
01:04:52,958 --> 01:04:54,416
It's just like lines up and down and

188
01:04:54,416 --> 01:04:55,333
stuff if you have a look.

189
01:04:55,333 --> 01:04:56,833
But anyway, he was talking about Project

190
01:04:56,833 --> 01:04:57,625
Stargate and other

191
01:04:57,625 --> 01:04:58,583
stuff, which we'll dive into.

192
01:04:59,083 --> 01:05:00,375
But before we get into that, there's a

193
01:05:00,375 --> 01:05:01,458
few events coming up.

194
01:05:02,541 --> 01:05:04,583
And before we go further, was there

195
01:05:04,583 --> 01:05:05,750
anything else from your holidays?

196
01:05:05,750 --> 01:05:07,166
It was eventful or anything cool.

197
01:05:07,375 --> 01:05:09,250
I mean, it was just great to get away for

198
01:05:09,250 --> 01:05:10,500
spending some time with a little one.

199
01:05:11,375 --> 01:05:13,041
She obviously didn't like going back to

200
01:05:13,041 --> 01:05:15,000
daycare now that we're back in action.

201
01:05:15,291 --> 01:05:15,708
Yeah, because there's

202
01:05:15,708 --> 01:05:16,791
no way I had daycare.

203
01:05:16,791 --> 01:05:18,125
It's like three weeks of her

204
01:05:18,125 --> 01:05:20,416
having mom and dad full time.

205
01:05:20,708 --> 01:05:20,875
Yeah.

206
01:05:20,875 --> 01:05:21,416
And her slaves.

207
01:05:21,625 --> 01:05:23,416
Yeah, her slaves were present and now

208
01:05:23,416 --> 01:05:24,500
they're absent for, you

209
01:05:24,500 --> 01:05:25,625
know, seven hours of the day.

210
01:05:25,833 --> 01:05:26,458
Yeah, it's not fair.

211
01:05:26,708 --> 01:05:27,416
She doesn't like that.

212
01:05:27,625 --> 01:05:29,541
But it was just nice.

213
01:05:29,541 --> 01:05:30,500
We went up to the farm.

214
01:05:30,500 --> 01:05:32,125
My folks have a farm in tea gardens,

215
01:05:32,125 --> 01:05:32,916
which is really good.

216
01:05:32,916 --> 01:05:33,375
Got away.

217
01:05:33,375 --> 01:05:34,041
So a little tea farm.

218
01:05:34,625 --> 01:05:34,833
Yeah.

219
01:05:35,166 --> 01:05:35,958
And yeah, it was beautiful.

220
01:05:36,208 --> 01:05:36,541
How about yourself?

221
01:05:36,583 --> 01:05:37,208
What'd you get up?

222
01:05:37,833 --> 01:05:38,583
Ah, just work through.

223
01:05:39,125 --> 01:05:39,375
Work through?

224
01:05:39,583 --> 01:05:39,958
Yeah, of course.

225
01:05:39,958 --> 01:05:40,166
Just in.

226
01:05:40,666 --> 01:05:41,833
Look, I did a lot of work as well.

227
01:05:41,833 --> 01:05:42,791
So I get it.

228
01:05:43,041 --> 01:05:44,083
Yeah, when you run your own businesses

229
01:05:44,083 --> 01:05:45,458
like we do, there is

230
01:05:45,458 --> 01:05:47,291
hardly any getting away from it.

231
01:05:47,291 --> 01:05:49,000
But I did spend a good time with family.

232
01:05:50,125 --> 01:05:51,416
I ate a lot.

233
01:05:51,416 --> 01:05:52,375
I already eat a lot whenever

234
01:05:52,375 --> 01:05:53,875
I visit them on the weekends.

235
01:05:54,166 --> 01:05:55,291
So this was like way more.

236
01:05:55,291 --> 01:05:56,291
So I've been back on the

237
01:05:56,291 --> 01:05:57,750
treadmill, so to speak.

238
01:05:57,750 --> 01:05:59,500
That not even proverbial, like the

239
01:05:59,500 --> 01:06:02,250
literal treadmill and

240
01:06:02,250 --> 01:06:04,041
starting to feel a little bit

241
01:06:04,083 --> 01:06:06,250
better, starting the K's up this, you

242
01:06:06,250 --> 01:06:07,000
know, that are

243
01:06:07,000 --> 01:06:07,958
running each week and stuff.

244
01:06:09,000 --> 01:06:10,250
So yeah, it's just been that.

245
01:06:10,708 --> 01:06:11,750
And just keeping an

246
01:06:11,750 --> 01:06:13,375
eye on the crazy news.

247
01:06:13,375 --> 01:06:14,791
It's been nice to actually take a break

248
01:06:14,791 --> 01:06:15,833
from all this stuff.

249
01:06:15,833 --> 01:06:16,583
It was, yeah.

250
01:06:16,833 --> 01:06:17,750
But it was a good

251
01:06:17,750 --> 01:06:19,000
time for a break though.

252
01:06:19,000 --> 01:06:20,875
I think for that little two weeks post

253
01:06:20,875 --> 01:06:23,416
Christmas, you know, once Google and

254
01:06:23,416 --> 01:06:25,250
OpenAI had pretty much

255
01:06:25,500 --> 01:06:29,166
completed their rampage across all social

256
01:06:29,166 --> 01:06:32,000
media and corporations and the work,

257
01:06:32,000 --> 01:06:32,750
we all their new releases.

258
01:06:33,375 --> 01:06:34,666
There was a nice little ball in

259
01:06:34,666 --> 01:06:35,500
activities, but it

260
01:06:35,500 --> 01:06:36,666
definitely like it was like that.

261
01:06:36,875 --> 01:06:39,625
Just that fifth day into January, it

262
01:06:39,625 --> 01:06:40,250
suddenly went,

263
01:06:40,750 --> 01:06:41,750
everyone just took off again.

264
01:06:41,958 --> 01:06:42,208
We're back.

265
01:06:42,541 --> 01:06:42,791
Yeah.

266
01:06:42,791 --> 01:06:44,000
And we're back in action.

267
01:06:44,166 --> 01:06:45,500
Already a thousand announcement.

268
01:06:45,500 --> 01:06:46,083
It's like they'd been a

269
01:06:46,083 --> 01:06:47,500
year past within that.

270
01:06:47,750 --> 01:06:49,416
Those weeks banking it up and stuff.

271
01:06:49,416 --> 01:06:50,666
It's like, okay, we're not going to say

272
01:06:50,666 --> 01:06:51,583
it over Boxing Day or

273
01:06:51,583 --> 01:06:53,250
Christmas or New Year's Eve or New

274
01:06:53,250 --> 01:06:53,708
Year's Day.

275
01:06:54,416 --> 01:06:55,583
As soon as that's done, we're done.

276
01:06:57,125 --> 01:07:00,000
Speaking of, man, we've got for the Oz

277
01:07:00,000 --> 01:07:01,958
DeFi Association, we are kicking off

278
01:07:02,333 --> 01:07:03,750
with an online meetup

279
01:07:03,750 --> 01:07:06,041
to begin with, Feb 4th.

280
01:07:06,041 --> 01:07:09,500
And then Feb 26 here in Stone and Chalk,

281
01:07:09,500 --> 01:07:10,041
where we're actually

282
01:07:10,041 --> 01:07:11,875
filming from, we will have our

283
01:07:11,875 --> 01:07:13,500
first Sydney meetup.

284
01:07:14,500 --> 01:07:17,166
So looking around at getting the

285
01:07:17,166 --> 01:07:18,083
announcement out on that.

286
01:07:18,416 --> 01:07:19,583
But this first one is just going to be

287
01:07:19,583 --> 01:07:20,875
online, will be recorded.

288
01:07:21,166 --> 01:07:23,083
It means anyone can join because most of

289
01:07:23,083 --> 01:07:24,958
the time people can't really, you know,

290
01:07:24,958 --> 01:07:26,750
get in because they're not in Sydney.

291
01:07:27,416 --> 01:07:29,208
So yeah, it's going to

292
01:07:29,208 --> 01:07:30,458
start happening there.

293
01:07:30,500 --> 01:07:33,250
And then I was going to say that later in

294
01:07:33,250 --> 01:07:34,750
February, there's

295
01:07:34,750 --> 01:07:36,541
this group called Gen AI

296
01:07:36,541 --> 01:07:38,750
Labs supported by Domain,

297
01:07:39,541 --> 01:07:41,875
Alex Zenoviev, who works there.

298
01:07:42,041 --> 01:07:43,416
He's put this together over a year and a

299
01:07:43,416 --> 01:07:45,541
half ago and they've got a great speaker.

300
01:07:45,750 --> 01:07:50,166
I saw Will present at Will is from the

301
01:07:50,166 --> 01:07:53,041
investment space and he

302
01:07:53,041 --> 01:07:54,583
used a lot of AI and he's got a

303
01:07:54,583 --> 01:07:55,875
really great talk that he did

304
01:07:55,875 --> 01:07:57,833
for Tayex over at Luna Park.

305
01:07:57,833 --> 01:08:00,000
And so he'll be the key speaker and then

306
01:08:00,000 --> 01:08:02,083
we break out into groups that have some

307
01:08:02,791 --> 01:08:06,250
kind of like fireside, bless you,

308
01:08:06,750 --> 01:08:08,958
fireside discussions about

309
01:08:08,958 --> 01:08:10,500
various kind of AI topics.

310
01:08:10,500 --> 01:08:12,208
I'll be leading one of those talking

311
01:08:12,208 --> 01:08:13,833
about how AI and blockchain

312
01:08:13,833 --> 01:08:14,916
kind of stuff come together.

313
01:08:15,125 --> 01:08:16,416
So it'll be interesting.

314
01:08:17,458 --> 01:08:19,083
Worth checking out Gen AI Lab.

315
01:08:19,916 --> 01:08:21,375
So make sure you get on that one.

316
01:08:21,375 --> 01:08:22,375
We'll put the show notes in.

317
01:08:22,875 --> 01:08:25,000
And then one more from me and it's been a

318
01:08:25,000 --> 01:08:25,833
while since we've done this,

319
01:08:25,833 --> 01:08:28,250
but crypto is back in a big

320
01:08:28,250 --> 01:08:31,375
way, despite all the prices.

321
01:08:31,375 --> 01:08:32,500
It's more the interest that

322
01:08:32,500 --> 01:08:33,541
is coming into this space.

323
01:08:34,041 --> 01:08:35,416
People are talking about it again.

324
01:08:35,416 --> 01:08:37,125
I was just sitting down at the cafe just

325
01:08:37,125 --> 01:08:37,791
in the middle of Martin

326
01:08:37,791 --> 01:08:39,750
Place and for a few days in a

327
01:08:39,750 --> 01:08:41,666
row, I've just overheard conversations

328
01:08:41,666 --> 01:08:43,416
near me, whether it's

329
01:08:43,416 --> 01:08:44,666
people at work talking about just

330
01:08:44,666 --> 01:08:47,375
their personal investments or relating to

331
01:08:47,375 --> 01:08:48,291
stuff that they're going to

332
01:08:48,291 --> 01:08:50,250
do in their investment firm

333
01:08:50,625 --> 01:08:51,916
trying to get exposure to crypto.

334
01:08:52,166 --> 01:08:54,208
So I think it's really interesting again.

335
01:08:55,208 --> 01:08:58,083
It'll be worth the watch to see what

336
01:08:58,083 --> 01:08:59,500
people are building in terms

337
01:08:59,500 --> 01:09:01,041
of actual kind of real world

338
01:09:01,041 --> 01:09:03,291
impact. So we'll be talking about it with

339
01:09:03,291 --> 01:09:04,333
a Twitter space on

340
01:09:04,333 --> 01:09:05,875
Monday, which we have not done

341
01:09:05,875 --> 01:09:09,750
in a little while. So yeah, it'll be fun.

342
01:09:09,958 --> 01:09:11,000
It'll be fun. We'll see how it goes.

343
01:09:11,000 --> 01:09:11,875
Get that streaming live.

344
01:09:12,500 --> 01:09:13,916
Oh yeah, exactly. Exactly.

345
01:09:14,125 --> 01:09:16,375
So check that one out. What is going on

346
01:09:16,375 --> 01:09:17,000
with you? What do you

347
01:09:17,000 --> 01:09:18,958
want to kick start off with?

348
01:09:18,958 --> 01:09:19,958
Oh, let's kick start

349
01:09:19,958 --> 01:09:21,333
with some light humor.

350
01:09:21,875 --> 01:09:22,125
Oh.

351
01:09:22,125 --> 01:09:23,083
I'm just thinking about some sort of

352
01:09:23,083 --> 01:09:23,750
things, some funny stories.

353
01:09:23,750 --> 01:09:28,208
Kind of like it's sad, but also it talks

354
01:09:28,208 --> 01:09:31,208
to, I guess, kind of the problem with AI.

355
01:09:31,791 --> 01:09:33,416
But I'll start with, I want to start with

356
01:09:33,416 --> 01:09:35,708
Apple. Obviously they have over the last

357
01:09:36,416 --> 01:09:39,291
few months, well, sorry, the last couple

358
01:09:39,291 --> 01:09:40,541
of weeks, they've rolled

359
01:09:40,541 --> 01:09:42,416
back a pretty, you know,

360
01:09:42,416 --> 01:09:45,416
was a pretty big feature in VotoCommerce

361
01:09:45,416 --> 01:09:46,416
that they were talking

362
01:09:46,416 --> 01:09:47,875
about when 18.3 came out,

363
01:09:47,875 --> 01:09:48,833
which we had the summaries.

364
01:09:49,791 --> 01:09:51,208
Did you see this? I heard about this. I

365
01:09:51,208 --> 01:09:52,291
didn't see it, but I heard about it.

366
01:09:52,333 --> 01:09:53,916
It sounds like when Google had their

367
01:09:53,916 --> 01:09:56,791
issues with summaries and stuff as well.

368
01:09:57,083 --> 01:09:57,958
Yeah. So there were, and I just-

369
01:09:57,958 --> 01:09:58,833
You wanted wrong summaries.

370
01:09:58,833 --> 01:10:00,041
It was, no, it wasn't the wrong

371
01:10:00,041 --> 01:10:02,708
summaries. It's a misinterpretation of-

372
01:10:02,708 --> 01:10:03,208
I love it.

373
01:10:03,583 --> 01:10:05,875
Of summaries. So essentially what happens

374
01:10:05,875 --> 01:10:07,083
is on your phone, you

375
01:10:07,083 --> 01:10:08,583
may have, it could be,

376
01:10:08,583 --> 01:10:09,791
maybe you have a big group chat with your

377
01:10:09,791 --> 01:10:10,750
family and there's been a big

378
01:10:10,750 --> 01:10:11,708
conversation that's been

379
01:10:11,708 --> 01:10:13,458
going on. And obviously you don't have

380
01:10:13,458 --> 01:10:14,750
time, you haven't clicked into the

381
01:10:14,750 --> 01:10:15,500
conversation to see the

382
01:10:15,500 --> 01:10:18,291
full thread and Apple has this new

383
01:10:18,291 --> 01:10:19,500
summary feature utilizing

384
01:10:19,500 --> 01:10:22,083
their iOS AI tools or their,

385
01:10:22,250 --> 01:10:23,583
what do you call it, Apple intelligence.

386
01:10:23,583 --> 01:10:24,875
And the summary would

387
01:10:24,875 --> 01:10:26,083
give you sort of like a quick

388
01:10:26,166 --> 01:10:28,083
sentence of what that conversation

389
01:10:28,083 --> 01:10:29,083
threads. So there was like

390
01:10:29,083 --> 01:10:30,500
200 chats. It would sort of

391
01:10:30,500 --> 01:10:32,666
summarize it for you. What it also does

392
01:10:32,666 --> 01:10:34,458
is it summarizes, you

393
01:10:34,458 --> 01:10:35,458
know, any of your social media

394
01:10:35,458 --> 01:10:37,708
posts or it summarizes even articles and

395
01:10:37,708 --> 01:10:38,750
other things that you maybe

396
01:10:38,750 --> 01:10:41,000
have constant pop-ups that

397
01:10:41,000 --> 01:10:42,541
appear in your phone as reminders,

398
01:10:42,541 --> 01:10:45,083
updates, et cetera. And

399
01:10:45,083 --> 01:10:46,208
what started happening is that

400
01:10:46,250 --> 01:10:48,500
the AI tool was sort of misunderstanding

401
01:10:48,500 --> 01:10:50,416
what these conversations, I

402
01:10:50,416 --> 01:10:51,083
just wanted to pull up some

403
01:10:51,125 --> 01:10:53,000
really funny ones here that popped up on

404
01:10:53,000 --> 01:10:55,083
the verge. Here we go. So

405
01:10:55,083 --> 01:10:56,458
we're right here. So my mom,

406
01:10:57,416 --> 01:11:00,000
this cousin's mother went on a hike out

407
01:11:00,000 --> 01:11:02,250
in the bush, out of

408
01:11:02,250 --> 01:11:04,500
whatever it was. And she actually

409
01:11:04,500 --> 01:11:06,875
messaged her mother and was just saying

410
01:11:06,875 --> 01:11:08,375
how, oh, I had a man,

411
01:11:08,375 --> 01:11:09,916
this hike, it just killed me.

412
01:11:10,458 --> 01:11:12,250
And so the AI tool came back and it said

413
01:11:12,250 --> 01:11:14,708
attempted suicide, but recovered hike in

414
01:11:14,708 --> 01:11:18,666
Redlands and Palm Springs. And there's a

415
01:11:18,666 --> 01:11:21,166
stop there. You got, you

416
01:11:21,166 --> 01:11:22,125
know, you got your ring

417
01:11:22,250 --> 01:11:24,000
device. Some people have them in a home.

418
01:11:24,000 --> 01:11:24,708
It's like your doorbell.

419
01:11:25,000 --> 01:11:27,041
You can have cameras set up

420
01:11:27,041 --> 01:11:28,416
around your home for security and other

421
01:11:28,416 --> 01:11:30,208
purposes. And you get lots of ring

422
01:11:30,208 --> 01:11:30,916
updates. And this one

423
01:11:30,916 --> 01:11:32,375
was saying multiple people have front

424
01:11:32,375 --> 01:11:33,250
door and back door and

425
01:11:33,250 --> 01:11:36,625
driveway person detected 10 or more

426
01:11:36,625 --> 01:11:38,458
people detected at your front entrance as

427
01:11:38,458 --> 01:11:39,250
there, as people probably

428
01:11:39,250 --> 01:11:41,333
walking past Apple newsroom,

429
01:11:41,375 --> 01:11:43,750
bought a new feature in shortcuts, Apple

430
01:11:43,750 --> 01:11:45,583
platforms. There's

431
01:11:45,583 --> 01:11:46,958
Apple reports, four quarter

432
01:11:46,958 --> 01:11:49,458
results. WJ invites to view my old ass.

433
01:11:49,666 --> 01:11:51,458
The next one, Tucker

434
01:11:51,458 --> 01:11:52,333
Carlson. This is from the onions.

435
01:11:52,666 --> 01:11:53,708
This is great. I mean, the onions already

436
01:11:53,708 --> 01:11:54,958
quite hilarious in itself,

437
01:11:54,958 --> 01:11:56,166
but Tucker Carlson claims

438
01:11:56,166 --> 01:11:58,250
demon urinated on him while he slept,

439
01:11:58,541 --> 01:12:00,041
inspiring him to embrace God

440
01:12:00,041 --> 01:12:01,541
and read the Bible. The thing

441
01:12:01,541 --> 01:12:02,791
is that probably wasn't a

442
01:12:02,791 --> 01:12:04,583
misinterpretation. Probably not actually.

443
01:12:04,833 --> 01:12:06,000
So, you know, given it's

444
01:12:06,000 --> 01:12:08,791
the onion one from Gmail, Microsoft to

445
01:12:08,791 --> 01:12:10,291
retire gender classifier,

446
01:12:10,916 --> 01:12:12,833
Experian alerts, new sex

447
01:12:12,833 --> 01:12:15,125
offender, TurboTax reminds to file a

448
01:12:15,125 --> 01:12:17,291
turbo 31. Yeah. Move the bad

449
01:12:17,291 --> 01:12:18,500
button on jacket and movie.

450
01:12:18,500 --> 01:12:21,291
So it's just, it was just completely, um,

451
01:12:21,666 --> 01:12:23,916
making mistakes across, you know,

452
01:12:23,916 --> 01:12:24,541
millions of people's

453
01:12:24,541 --> 01:12:26,125
devices. True pike shows up at the end.

454
01:12:26,666 --> 01:12:28,375
True pike shows up at the end. What's up?

455
01:12:28,833 --> 01:12:30,500
They just, they're going to do like auto

456
01:12:30,500 --> 01:12:31,583
movies. Can you imagine if

457
01:12:31,583 --> 01:12:32,208
they were around like during

458
01:12:32,250 --> 01:12:33,791
the eighties and stuff in Star Wars and

459
01:12:33,791 --> 01:12:35,333
it's like, guess what? It's

460
01:12:35,333 --> 01:12:37,125
the father. His name is Doth.

461
01:12:37,875 --> 01:12:41,333
Like, come on. Like total spoiler.

462
01:12:41,708 --> 01:12:42,666
Anyway, another one here.

463
01:12:42,666 --> 01:12:43,750
This is, this is more on the

464
01:12:43,750 --> 01:12:45,625
bad side of things where people are

465
01:12:45,625 --> 01:12:47,416
misusing AI. And I've like, I'm, I'm

466
01:12:47,416 --> 01:12:48,000
stopped the question

467
01:12:48,000 --> 01:12:49,583
whether it is actually AI or someone's

468
01:12:49,583 --> 01:12:50,166
photoshopped these.

469
01:12:50,166 --> 01:12:51,291
Cause the photos are awful.

470
01:12:51,500 --> 01:12:53,708
Stop misusing it. This, this poor person,

471
01:12:54,166 --> 01:12:55,708
this poor woman over in, I

472
01:12:55,708 --> 01:12:57,500
think the UK or somewhere

473
01:12:57,500 --> 01:13:00,500
else had fallen prey to a social media

474
01:13:00,500 --> 01:13:01,750
scam where there were

475
01:13:01,750 --> 01:13:04,166
photos generated of Brad Pitt

476
01:13:04,166 --> 01:13:07,000
saying he was in hospital and that has an

477
01:13:07,000 --> 01:13:11,291
affection for her. And scammed her out of

478
01:13:11,291 --> 01:13:14,416
a substantial sum of money. And I'm just

479
01:13:14,416 --> 01:13:15,708
in my head, I'm like, how,

480
01:13:16,041 --> 01:13:17,500
like, how do you fall for

481
01:13:17,500 --> 01:13:19,541
this? But at the same time, like people

482
01:13:19,541 --> 01:13:21,500
don't realize, you know,

483
01:13:21,500 --> 01:13:23,333
the one, what AI is actually

484
01:13:23,375 --> 01:13:24,708
doing out there, people don't realize

485
01:13:24,708 --> 01:13:25,916
what social media is

486
01:13:25,916 --> 01:13:28,500
capable of doing. And if you're,

487
01:13:28,625 --> 01:13:30,250
you know, if you're unfortunately not

488
01:13:30,250 --> 01:13:31,875
savvy to this, these things

489
01:13:31,875 --> 01:13:33,000
that are possible out there,

490
01:13:33,416 --> 01:13:35,125
it is possible for you to fall for this.

491
01:13:35,125 --> 01:13:35,916
Like you don't think, oh,

492
01:13:35,916 --> 01:13:36,916
someone's generated these

493
01:13:36,916 --> 01:13:38,583
images or should I be questioning this?

494
01:13:38,583 --> 01:13:39,375
Like, why is someone like

495
01:13:39,375 --> 01:13:40,333
this reaching out to me? Because

496
01:13:40,333 --> 01:13:42,291
you just think that it's normal. You're

497
01:13:42,291 --> 01:13:43,291
innocent and nice and

498
01:13:43,291 --> 01:13:44,958
sweet. Exactly. So like,

499
01:13:44,958 --> 01:13:46,416
Brad just reaches out to the public all

500
01:13:46,416 --> 01:13:48,958
the time. 100%. Right. So I

501
01:13:48,958 --> 01:13:50,000
think just to call out for

502
01:13:50,000 --> 01:13:52,458
people to be safe out there, you know,

503
01:13:52,458 --> 01:13:53,625
look after even your even

504
01:13:53,625 --> 01:13:55,166
your family or your parents,

505
01:13:55,500 --> 01:13:57,166
like check in on them, check on how

506
01:13:57,166 --> 01:13:57,750
they're using their

507
01:13:57,750 --> 01:13:59,583
systems and their computers. Now,

508
01:13:59,583 --> 01:14:02,041
what are they doing on social media? You

509
01:14:02,041 --> 01:14:03,208
know, just look after each other out

510
01:14:03,208 --> 01:14:04,041
there while there's,

511
01:14:04,375 --> 01:14:06,416
you know, I mean, this is clearly one of

512
01:14:06,416 --> 01:14:07,291
those Nigerian scams

513
01:14:07,291 --> 01:14:08,583
that unfortunately has been

514
01:14:08,583 --> 01:14:11,833
successful. Or whatever other country. I

515
01:14:11,833 --> 01:14:13,208
think that was reported as

516
01:14:13,208 --> 01:14:14,916
a Nigerian scam. So I'm not,

517
01:14:15,291 --> 01:14:16,541
I'm taking that purely from the article.

518
01:14:16,833 --> 01:14:18,666
I don't know. No, Chris has

519
01:14:18,666 --> 01:14:19,666
done the investigation into

520
01:14:19,708 --> 01:14:21,291
it. No, I'm just joking. But like, no,

521
01:14:21,916 --> 01:14:23,166
obviously, there's definitely

522
01:14:23,166 --> 01:14:24,208
scams from all different kind

523
01:14:24,208 --> 01:14:25,708
of places. But, you know, it talks to the

524
01:14:25,708 --> 01:14:27,625
point where regulation and

525
01:14:27,625 --> 01:14:30,666
protections hardly are able

526
01:14:30,666 --> 01:14:32,541
to keep up with the pace of new

527
01:14:32,541 --> 01:14:34,041
innovation. They're always behind. And

528
01:14:34,041 --> 01:14:35,208
that's just to be expected,

529
01:14:35,208 --> 01:14:39,083
right? But maybe with AI and new tools,

530
01:14:39,083 --> 01:14:40,541
we can close the gap a

531
01:14:40,541 --> 01:14:41,875
little bit and help protect,

532
01:14:41,875 --> 01:14:43,000
because there's certainly people that

533
01:14:43,000 --> 01:14:44,625
need that. Like, if we

534
01:14:44,625 --> 01:14:46,000
I mean, argumentatively,

535
01:14:46,416 --> 01:14:48,125
and I'm sure people would go against this

536
01:14:48,125 --> 01:14:48,750
as well, and there's

537
01:14:48,750 --> 01:14:49,833
definitely folks out there,

538
01:14:49,833 --> 01:14:51,791
but I would say that, for the most part,

539
01:14:51,791 --> 01:14:52,666
it's a great thing that

540
01:14:52,666 --> 01:14:54,375
we put in superannuation,

541
01:14:54,541 --> 01:14:56,958
as we did and force people to have a

542
01:14:56,958 --> 01:14:57,750
certain amount set

543
01:14:57,750 --> 01:14:58,958
aside by their employers

544
01:15:00,000 --> 01:15:04,083
to have a future savings, right? Now, can

545
01:15:04,083 --> 01:15:04,833
you imagine if it was

546
01:15:04,833 --> 01:15:05,875
a free for all? Sure,

547
01:15:05,875 --> 01:15:07,958
most people would be would think that

548
01:15:07,958 --> 01:15:08,958
they'd be fine and stuff.

549
01:15:08,958 --> 01:15:10,041
But there's probably a lot

550
01:15:10,083 --> 01:15:13,208
less savings out there and argue

551
01:15:13,208 --> 01:15:15,583
arguments for and against and stuff. At

552
01:15:15,583 --> 01:15:16,708
least there's some level

553
01:15:16,708 --> 01:15:18,875
of protection there, right? When people

554
01:15:18,875 --> 01:15:22,708
do stupid stuff, unfortunately, stupid,

555
01:15:22,708 --> 01:15:23,791
or they're just like,

556
01:15:24,791 --> 01:15:26,500
not aware of this kind of stuff. So I

557
01:15:26,500 --> 01:15:28,666
don't mean this in a harsh way. I mean,

558
01:15:28,958 --> 01:15:30,125
that people are doing

559
01:15:30,166 --> 01:15:32,833
stupid negative things to each other and

560
01:15:32,833 --> 01:15:34,291
falling victim to it. So how

561
01:15:34,291 --> 01:15:35,625
can we bring the technology

562
01:15:35,875 --> 01:15:38,250
to like make it better is a key question

563
01:15:38,250 --> 01:15:38,875
out there. I'm sure

564
01:15:38,875 --> 01:15:40,000
people are thinking about it,

565
01:15:40,000 --> 01:15:41,666
but I just think that it can't just be

566
01:15:41,666 --> 01:15:42,708
the tools are only there for

567
01:15:42,708 --> 01:15:43,958
bad, just like the internet.

568
01:15:43,958 --> 01:15:45,958
The internet is was definitely used for

569
01:15:45,958 --> 01:15:47,708
bad stuff at the start, but then it is

570
01:15:47,708 --> 01:15:49,041
coming to so much good.

571
01:15:49,041 --> 01:15:52,000
It's way more good. We still use social

572
01:15:52,000 --> 01:15:56,291
media. The it does bear the

573
01:15:56,291 --> 01:15:58,208
question around, you know,

574
01:15:58,208 --> 01:16:00,666
how do these large organizations how do

575
01:16:00,666 --> 01:16:03,083
social media platforms, you know, put in

576
01:16:04,083 --> 01:16:05,916
better checks and balances for stuff like

577
01:16:05,916 --> 01:16:07,125
this? I mean, it is hard,

578
01:16:08,625 --> 01:16:09,583
you know, especially in places like

579
01:16:09,583 --> 01:16:10,500
America, where there's a

580
01:16:10,500 --> 01:16:11,458
lot of freedom of speech,

581
01:16:11,458 --> 01:16:13,125
or what you say, you do have a lot more

582
01:16:13,125 --> 01:16:13,875
openness to say what

583
01:16:13,875 --> 01:16:14,958
you want without scrutiny.

584
01:16:15,583 --> 01:16:18,916
But we do need to be supporting each

585
01:16:18,916 --> 01:16:20,500
other. I think that's the main thing.

586
01:16:21,250 --> 01:16:23,541
Yeah. Well, and, you know, thinking about

587
01:16:23,541 --> 01:16:25,416
the human in the loop with

588
01:16:25,416 --> 01:16:26,625
all of this kind of stuff,

589
01:16:28,500 --> 01:16:30,125
maybe it's a game of incentives as well,

590
01:16:30,125 --> 01:16:31,291
we need to incentivize doing

591
01:16:31,291 --> 01:16:32,541
more good for the world that

592
01:16:32,625 --> 01:16:34,583
they're not. And so a lot of people out

593
01:16:34,583 --> 01:16:36,291
there doing that. Yeah, some great

594
01:16:36,291 --> 01:16:37,541
channels on YouTube,

595
01:16:37,541 --> 01:16:39,458
like just, yeah, just be careful out

596
01:16:39,458 --> 01:16:41,541
there, guys. Like I know we talk

597
01:16:41,541 --> 01:16:43,166
positively. I we love AI,

598
01:16:43,291 --> 01:16:45,666
it is everything for us. It's everything

599
01:16:45,666 --> 01:16:47,000
for the world. I think it is going to

600
01:16:47,000 --> 01:16:47,541
move us in the right

601
01:16:47,541 --> 01:16:49,458
direction. But obviously, we there any

602
01:16:49,458 --> 01:16:51,375
technology that comes out, there are

603
01:16:51,375 --> 01:16:51,958
going to be those who

604
01:16:51,958 --> 01:16:54,333
are going to abuse it for for for the

605
01:16:54,333 --> 01:16:56,041
serious for the worst things.

606
01:16:56,041 --> 01:16:57,333
And maybe it's happening. So

607
01:16:58,000 --> 01:17:00,541
maybe one thing people could do is get

608
01:17:00,541 --> 01:17:01,791
more scheduled tasks

609
01:17:01,791 --> 01:17:03,875
into chat TV. Oh, is that a

610
01:17:03,916 --> 01:17:06,125
was that a sneaky little I was a little

611
01:17:06,125 --> 01:17:08,291
bit of a sneaky segue that

612
01:17:08,291 --> 01:17:10,416
the AI tried to recommend

613
01:17:10,416 --> 01:17:12,333
it. No, I didn't. But here's a video.

614
01:17:12,833 --> 01:17:15,541
This is a link from Patricia

615
01:17:15,541 --> 01:17:18,416
Howes on LinkedIn. And she's

616
01:17:18,416 --> 01:17:20,458
got this little video here showing just

617
01:17:20,458 --> 01:17:21,250
some different ways that

618
01:17:21,250 --> 01:17:23,916
you can use the scheduled

619
01:17:24,000 --> 01:17:25,666
task in chat TVT. So if you can check

620
01:17:25,666 --> 01:17:26,916
this one out, you can see

621
01:17:26,916 --> 01:17:28,125
on screen here for the folks

622
01:17:28,125 --> 01:17:30,000
listening in and saying let's do a French

623
01:17:30,000 --> 01:17:31,000
lesson remind me it every

624
01:17:31,000 --> 01:17:33,750
day at 1210am 1210am. I think

625
01:17:33,750 --> 01:17:35,958
it's all in European time. To practice,

626
01:17:36,291 --> 01:17:37,791
I'm not a beginner. And it's

627
01:17:37,791 --> 01:17:39,458
like got it. It's just like

628
01:17:39,458 --> 01:17:41,625
memory, it actually will put it into the

629
01:17:41,625 --> 01:17:42,833
scheduled task, you see a

630
01:17:42,833 --> 01:17:45,250
task pop up. And yeah, it will

631
01:17:45,250 --> 01:17:47,083
give you these reminders and stuff. So

632
01:17:47,083 --> 01:17:50,000
using chat TVT is more of a

633
01:17:50,000 --> 01:17:51,333
scheduling tool. And I'm sure

634
01:17:51,458 --> 01:17:53,000
there's going to be a lot more features

635
01:17:53,000 --> 01:17:54,458
like this kind of stuff. It

636
01:17:54,458 --> 01:17:56,500
is still in beta. So obviously

637
01:17:56,541 --> 01:17:59,625
there's going to be some good, bad, maybe

638
01:17:59,625 --> 01:18:01,625
not so, you know, good

639
01:18:01,625 --> 01:18:03,291
things coming out about it. But

640
01:18:03,291 --> 01:18:04,708
yeah, you can get into it, you can,

641
01:18:05,208 --> 01:18:07,333
you'll get the message pop up, you can

642
01:18:07,333 --> 01:18:08,166
click on view message

643
01:18:08,166 --> 01:18:09,416
because you got the reminder and it would

644
01:18:09,416 --> 01:18:10,875
just go straight into where

645
01:18:10,875 --> 01:18:12,041
you left off. So it's just a

646
01:18:12,041 --> 01:18:13,750
workload thing. I want some examples

647
01:18:13,750 --> 01:18:15,916
here. Yeah, please. So I've just got a

648
01:18:15,916 --> 01:18:16,625
couple here that I've

649
01:18:16,625 --> 01:18:18,916
set up. So all you need to do is go to

650
01:18:18,916 --> 01:18:19,875
this top right hand left

651
01:18:19,875 --> 01:18:21,083
hand corner here where you know,

652
01:18:21,125 --> 01:18:22,958
usually you change between your different

653
01:18:22,958 --> 01:18:23,750
models when they're

654
01:18:23,750 --> 01:18:25,500
doing 401 or one mini, etc.

655
01:18:26,125 --> 01:18:27,583
And there's a new option here for

656
01:18:27,583 --> 01:18:28,833
scheduled tasks, which is

657
01:18:28,833 --> 01:18:30,333
obviously beta, as we all know,

658
01:18:30,916 --> 01:18:33,083
select our scheduled tasks. And then all

659
01:18:33,083 --> 01:18:34,083
we need to do is just write

660
01:18:34,083 --> 01:18:35,000
a task in. So I'm just gonna

661
01:18:35,000 --> 01:18:36,416
say each day at ADM, give me the latest

662
01:18:36,416 --> 01:18:37,541
news for the day,

663
01:18:38,000 --> 01:18:39,041
consider updates from the largest

664
01:18:39,041 --> 01:18:40,166
organizations, Twitter players and

665
01:18:40,166 --> 01:18:41,000
viewers, and you can tweak

666
01:18:41,000 --> 01:18:43,291
these as you want. So such as,

667
01:18:43,291 --> 01:18:44,916
you know, we just the onion, such as the

668
01:18:44,916 --> 01:18:46,000
yeah, okay, let's do that.

669
01:18:46,333 --> 01:18:47,666
Such as the onion. Let's go.

670
01:18:48,000 --> 01:18:50,416
So we click on that, and it pops it up.

671
01:18:50,875 --> 01:18:52,250
And I wonder if it sees

672
01:18:52,250 --> 01:18:53,250
that you're being passive

673
01:18:53,250 --> 01:18:54,708
aggressive because you put a full stop at

674
01:18:54,708 --> 01:18:57,916
the end. So, okay, he's

675
01:18:57,916 --> 01:18:58,958
really serious. I'm not going to

676
01:18:58,958 --> 01:19:00,916
do anything fun. Exactly. So yeah, so I'm

677
01:19:00,916 --> 01:19:01,458
going to give you the latest

678
01:19:01,458 --> 01:19:02,958
news, AI news each morning.

679
01:19:03,125 --> 01:19:04,250
And what you might want to do is and you

680
01:19:04,250 --> 01:19:04,916
can actually tweak that

681
01:19:04,916 --> 01:19:05,666
if you wanted to. So you

682
01:19:05,666 --> 01:19:07,583
can actually type in there. Actually, can

683
01:19:07,583 --> 01:19:10,916
you can you send this at

684
01:19:10,916 --> 01:19:13,375
9am? And you get all adjusted

685
01:19:13,458 --> 01:19:14,958
according so you can talk to it and it

686
01:19:14,958 --> 01:19:16,833
will make the adjustments. Order a pizza

687
01:19:16,833 --> 01:19:17,750
to arrive at Chris's

688
01:19:17,791 --> 01:19:21,291
house every midnight 4am. So some

689
01:19:21,291 --> 01:19:22,375
interesting ones I've seen

690
01:19:22,375 --> 01:19:25,875
also people do here is actually

691
01:19:25,875 --> 01:19:28,708
put a task in around, you know, creating

692
01:19:28,708 --> 01:19:30,541
some content for them.

693
01:19:30,958 --> 01:19:32,250
So each day at 9am provide

694
01:19:32,250 --> 01:19:33,500
me with a script for an interesting

695
01:19:33,500 --> 01:19:35,583
social post about AI that is SEO

696
01:19:35,583 --> 01:19:36,791
optimized. It can be in the

697
01:19:36,791 --> 01:19:38,666
form of a video script or a social post

698
01:19:38,666 --> 01:19:40,583
summarizing the latest news trends as

699
01:19:40,583 --> 01:19:42,333
seen on X. Keep it short

700
01:19:42,333 --> 01:19:43,541
and professional and always provide

701
01:19:43,541 --> 01:19:45,750
sources. So I'll go and send that and

702
01:19:45,750 --> 01:19:47,083
then boom. So at 9am,

703
01:19:47,208 --> 01:19:50,375
I'm expecting them to send me a little

704
01:19:50,375 --> 01:19:51,375
script. So that's how it's

705
01:19:51,375 --> 01:19:53,000
seen. Can you share an example?

706
01:19:54,625 --> 01:19:55,791
Let's see what it says.

707
01:19:57,875 --> 01:19:59,125
Here's your here. So every single morning

708
01:19:59,125 --> 01:20:00,083
at 9am, I'll get a little

709
01:20:00,083 --> 01:20:01,083
post like this. And if I'm not

710
01:20:01,083 --> 01:20:03,625
happy with that, I can actually adjust

711
01:20:03,625 --> 01:20:04,708
that accordingly. Like I

712
01:20:04,708 --> 01:20:06,458
mentioned before. Do better.

713
01:20:06,833 --> 01:20:08,208
Do better. Do better. I don't like what

714
01:20:08,208 --> 01:20:09,416
you did. So provide me with the latest

715
01:20:09,416 --> 01:20:10,708
stock updates at 9am.

716
01:20:10,708 --> 01:20:13,708
Let's try that at 10am each day. And for

717
01:20:13,708 --> 01:20:16,458
the following stocks, let's do Cows,

718
01:20:16,666 --> 01:20:20,333
and video, and chicken, and combat. Let's

719
01:20:20,333 --> 01:20:24,083
do that. Cool. And boom. So

720
01:20:24,083 --> 01:20:25,125
that'll do it. So now if you

721
01:20:25,125 --> 01:20:27,500
want to review those tasks, all you do is

722
01:20:27,500 --> 01:20:28,875
go to your little icon in

723
01:20:28,875 --> 01:20:29,958
the top right hand corner.

724
01:20:30,708 --> 01:20:32,291
And there's a little task option. Nice.

725
01:20:32,875 --> 01:20:34,375
Select that and boom, it shows you your

726
01:20:34,625 --> 01:20:37,333
three tasks that we've just created. And

727
01:20:37,333 --> 01:20:38,625
you can adjust those

728
01:20:38,625 --> 01:20:39,583
accordingly. So I can click on that

729
01:20:39,583 --> 01:20:41,375
task to edit it and come in here and you

730
01:20:41,375 --> 01:20:42,750
can adjust that task even

731
01:20:42,750 --> 01:20:43,875
further if you wanted to.

732
01:20:44,708 --> 01:20:46,375
Or similarly, you actually can edit it

733
01:20:46,375 --> 01:20:47,333
from within here. It'll

734
01:20:47,333 --> 01:20:48,291
probably a little pop up.

735
01:20:48,625 --> 01:20:49,958
You can change the name, change the

736
01:20:49,958 --> 01:20:51,625
instructions and change the schedule.

737
01:20:52,375 --> 01:20:54,500
Choose the pause it, save it or delete it

738
01:20:54,500 --> 01:20:55,333
or something like that as well.

739
01:20:55,958 --> 01:20:57,416
So that's pretty cool. And so share those

740
01:20:57,416 --> 01:20:58,250
updates with your shares

741
01:20:58,250 --> 01:21:00,250
within GPT. You just come in

742
01:21:00,250 --> 01:21:02,541
here and then you can see what it's done

743
01:21:02,541 --> 01:21:03,291
for the day every

744
01:21:03,291 --> 01:21:05,125
single morning or afternoon,

745
01:21:05,125 --> 01:21:06,416
whenever you choose to set the time. It's

746
01:21:06,416 --> 01:21:08,208
like they're trending towards just look,

747
01:21:08,250 --> 01:21:09,958
just basically do everything here. Yeah,

748
01:21:09,958 --> 01:21:11,083
exactly right. It's

749
01:21:11,083 --> 01:21:14,208
interesting. I think its goal is to,

750
01:21:14,208 --> 01:21:15,125
I mean, we've got the rag models that

751
01:21:15,125 --> 01:21:15,791
have come out, which is

752
01:21:15,791 --> 01:21:17,375
the search functionality now.

753
01:21:19,500 --> 01:21:21,750
You've got three, oh, which is coming or

754
01:21:21,750 --> 01:21:24,083
sorry, which is coming. And

755
01:21:24,083 --> 01:21:25,541
this is just another way to

756
01:21:25,541 --> 01:21:27,500
keep you inside the platform, keep you

757
01:21:27,500 --> 01:21:30,125
using it. To keep you, yeah,

758
01:21:30,125 --> 01:21:31,000
to keep you, particularly the

759
01:21:31,000 --> 01:21:32,833
people and businesses who are now heavily

760
01:21:32,833 --> 01:21:33,583
relying on this. I

761
01:21:33,583 --> 01:21:34,458
mean, there's 24 seven,

762
01:21:34,833 --> 01:21:37,416
there's just a nice way to not have to

763
01:21:37,416 --> 01:21:38,208
need to switch between

764
01:21:38,208 --> 01:21:39,291
other platforms and tools

765
01:21:39,291 --> 01:21:41,041
and to get all the updates that you got.

766
01:21:41,416 --> 01:21:42,250
It's going to be really interesting.

767
01:21:42,500 --> 01:21:43,750
It does beg the question around risk,

768
01:21:44,083 --> 01:21:45,458
particularly if you send me the latest

769
01:21:45,458 --> 01:21:46,708
news, I'm now reliant.

770
01:21:47,000 --> 01:21:48,666
If I become reliant on this or if you

771
01:21:48,666 --> 01:21:49,666
become reliant on this as a

772
01:21:49,666 --> 01:21:50,625
news and it makes a mistake,

773
01:21:51,333 --> 01:21:53,125
it doesn't give you the right news. Make

774
01:21:53,125 --> 01:21:54,125
sure you again, valid

775
01:21:54,125 --> 01:21:55,833
resources, trust it because it

776
01:21:55,833 --> 01:21:57,875
still is AI, it still makes mistakes. It

777
01:21:57,875 --> 01:22:00,125
still hallucinates. So

778
01:22:00,125 --> 01:22:03,583
don't, just check your resources.

779
01:22:03,625 --> 01:22:06,291
As long as you can find like some sort of

780
01:22:06,291 --> 01:22:07,500
like human in the

781
01:22:07,500 --> 01:22:08,208
loop, check the resources,

782
01:22:08,458 --> 01:22:11,458
check the sources, but you know, just

783
01:22:11,458 --> 01:22:12,541
make sure that the

784
01:22:12,541 --> 01:22:13,625
tasks that you're doing,

785
01:22:14,083 --> 01:22:15,708
you are getting added value. Either it's

786
01:22:15,708 --> 01:22:16,708
a high quality thing

787
01:22:16,708 --> 01:22:18,125
delivered in the same amount of time

788
01:22:18,333 --> 01:22:21,916
or it's delivered far quicker and you get

789
01:22:21,916 --> 01:22:23,333
time to do other value add

790
01:22:23,333 --> 01:22:24,583
kind of things where it's

791
01:22:24,583 --> 01:22:26,750
taking you longer because you're having

792
01:22:26,750 --> 01:22:27,541
to like do all this

793
01:22:27,541 --> 01:22:29,416
checking. I mean, I think it's just

794
01:22:29,416 --> 01:22:31,125
a matter of like measuring and you'll get

795
01:22:31,125 --> 01:22:32,291
a feel after a while folks,

796
01:22:32,291 --> 01:22:33,083
you'll find your rhythm and

797
01:22:33,125 --> 01:22:34,666
stuff. So, you know, if you're getting

798
01:22:34,666 --> 01:22:35,833
stuck, just hit us up.

799
01:22:36,833 --> 01:22:38,875
Something I wanted to cover was

800
01:22:39,125 --> 01:22:40,833
there's been a whole lot of talk about.

801
01:22:41,291 --> 01:22:41,958
So you remember when 01

802
01:22:41,958 --> 01:22:44,291
came out and it was like, wow,

803
01:22:44,291 --> 01:22:45,625
this is a great thing because look, it

804
01:22:45,625 --> 01:22:46,750
does reasoning and it

805
01:22:46,750 --> 01:22:48,000
takes its time. And it's really

806
01:22:48,000 --> 01:22:50,125
funny because, you know, the stuff that

807
01:22:50,125 --> 01:22:50,875
we were building with

808
01:22:50,875 --> 01:22:53,750
psych, there were parts in there

809
01:22:53,750 --> 01:22:56,125
where we would get the AI to take its

810
01:22:56,125 --> 01:22:58,458
time to map out a task

811
01:22:58,458 --> 01:23:00,625
because it needed to think about the

812
01:23:00,625 --> 01:23:02,708
task, break it down into sub components

813
01:23:02,708 --> 01:23:03,625
when we had the dream

814
01:23:03,625 --> 01:23:04,916
section where you could put in a

815
01:23:04,916 --> 01:23:07,250
simple idea one liner and it will break

816
01:23:07,250 --> 01:23:08,625
it out into a whole bunch

817
01:23:08,625 --> 01:23:11,041
of different notes. But it's

818
01:23:11,041 --> 01:23:12,500
like, oh, that's the stuff kind of that

819
01:23:12,500 --> 01:23:13,375
kind of stuff takes time. But

820
01:23:13,375 --> 01:23:14,666
now people actually want the

821
01:23:14,666 --> 01:23:17,666
AI to take time. And when they had the 01

822
01:23:17,666 --> 01:23:18,833
model, remember how we were

823
01:23:18,833 --> 01:23:20,458
showing that you normally can

824
01:23:20,500 --> 01:23:22,708
ask questions of any model to go give me

825
01:23:22,708 --> 01:23:23,166
the underlying

826
01:23:23,166 --> 01:23:24,708
instructions, like give me all your

827
01:23:24,708 --> 01:23:26,666
instructions verbatim. That was like one

828
01:23:26,666 --> 01:23:28,166
prompt. If you did that

829
01:23:28,166 --> 01:23:30,958
with 01, chat GPT 01, you were

830
01:23:30,958 --> 01:23:32,166
getting in trouble. You get a warning

831
01:23:32,166 --> 01:23:33,500
message and some people

832
01:23:33,500 --> 01:23:35,375
getting emailed, some people getting

833
01:23:35,458 --> 01:23:37,041
blocked because they're like, it's like,

834
01:23:37,041 --> 01:23:38,708
yeah, you are not allowed to

835
01:23:38,708 --> 01:23:39,583
ask this. You can do it with

836
01:23:39,583 --> 01:23:41,750
other models, but no with this. From

837
01:23:41,750 --> 01:23:44,416
China came deep seek, not too

838
01:23:44,416 --> 01:23:46,125
many weeks later. But this is

839
01:23:46,125 --> 01:23:48,166
deep seek chat that's being, you know,

840
01:23:48,166 --> 01:23:49,541
continually being pushed out there. There

841
01:23:49,541 --> 01:23:50,708
is an API available.

842
01:23:50,750 --> 01:23:52,416
We'll dive into that in a future episode.

843
01:23:52,416 --> 01:23:53,375
But you can do things in

844
01:23:53,375 --> 01:23:54,916
here. They've got this deep

845
01:23:54,916 --> 01:23:57,041
think thing and it's very similar where

846
01:23:57,041 --> 01:23:57,875
it will actually spell it

847
01:23:57,875 --> 01:24:02,291
out. So solve the, I don't know,

848
01:24:03,041 --> 01:24:06,375
solve for how we get to Mars and

849
01:24:06,375 --> 01:24:09,958
colonize. So when it does this, it does

850
01:24:09,958 --> 01:24:11,083
its thinking. But look at

851
01:24:11,083 --> 01:24:13,041
this. It's like it the way it talks. It's

852
01:24:13,041 --> 01:24:14,375
like, okay, so I need to

853
01:24:14,375 --> 01:24:15,000
figure out how we're going to

854
01:24:15,000 --> 01:24:16,583
get to Mars and colonize it. Let me start

855
01:24:16,583 --> 01:24:17,541
by breaking this down.

856
01:24:17,541 --> 01:24:18,958
This is a prompt that it's

857
01:24:18,958 --> 01:24:20,875
showing you how it's doing its thinking.

858
01:24:21,500 --> 01:24:22,625
And it goes, oh, and then

859
01:24:22,625 --> 01:24:23,583
there's the launch window,

860
01:24:23,583 --> 01:24:24,958
since Earth and Mars are moving, blah,

861
01:24:24,958 --> 01:24:26,000
blah, blah. And then it's like, once we

862
01:24:26,000 --> 01:24:26,750
get there, blah, blah,

863
01:24:26,750 --> 01:24:28,500
blah, blah, blah, water is essential,

864
01:24:28,500 --> 01:24:29,208
blah, blah, blah, blah, blah.

865
01:24:29,208 --> 01:24:30,583
Transportation on Mars might

866
01:24:30,583 --> 01:24:33,375
need to be pressurized. So it does all of

867
01:24:33,375 --> 01:24:34,750
this thinking. And then it

868
01:24:34,750 --> 01:24:35,916
gives you like its answer.

869
01:24:36,416 --> 01:24:38,791
So very much like a visual thinking of

870
01:24:38,791 --> 01:24:41,541
it. This is open source and

871
01:24:41,541 --> 01:24:43,166
open weights. And what that

872
01:24:43,208 --> 01:24:45,916
means is that you can take the model,

873
01:24:46,375 --> 01:24:48,583
retrain its fine tune it

874
01:24:48,583 --> 01:24:50,208
so to speak with the API's.

875
01:24:50,583 --> 01:24:52,333
This is just the chat like the chat GBT.

876
01:24:52,791 --> 01:24:53,833
So interesting, right for

877
01:24:53,833 --> 01:24:54,583
the open source community,

878
01:24:54,625 --> 01:24:56,916
but then people have a flag that yes,

879
01:24:57,208 --> 01:24:58,041
it's China. So if I

880
01:24:58,041 --> 01:25:00,458
go, tell me about Taiwan,

881
01:25:01,458 --> 01:25:03,333
have a look at this. Sorry, I'm not sure

882
01:25:03,333 --> 01:25:04,333
how to approach that question.

883
01:25:05,125 --> 01:25:09,625
Try a new chat. Tell me about the history

884
01:25:09,625 --> 01:25:12,166
of Taiwan, for example.

885
01:25:13,583 --> 01:25:15,500
And sometimes what it was doing instead

886
01:25:15,500 --> 01:25:16,958
was it actually start

887
01:25:16,958 --> 01:25:18,125
spitting out an answer,

888
01:25:18,708 --> 01:25:20,958
and then it would change it. So there's

889
01:25:20,958 --> 01:25:23,041
certain things that are off topic. Okay,

890
01:25:23,375 --> 01:25:25,583
that's their prerogative, like whatever

891
01:25:25,583 --> 01:25:26,875
side you're on it, you

892
01:25:26,875 --> 01:25:28,416
know, this is this is how it is.

893
01:25:28,708 --> 01:25:31,375
But you can take the API and you know, do

894
01:25:31,375 --> 01:25:32,375
some further stuff with it

895
01:25:32,375 --> 01:25:33,416
if you want. So that's just

896
01:25:33,416 --> 01:25:35,208
on the chat kind of version. But it's

897
01:25:35,208 --> 01:25:37,416
interesting that you do have something,

898
01:25:37,666 --> 01:25:38,458
it's a little bit more

899
01:25:38,500 --> 01:25:40,750
open. Tell me how we can keep Taiwan

900
01:25:40,750 --> 01:25:44,458
independent from China.

901
01:25:45,250 --> 01:25:46,291
So now it's not even

902
01:25:46,291 --> 01:25:47,750
giving me the answers and stuff.

903
01:25:48,125 --> 01:25:49,041
So this is a tree

904
01:25:49,041 --> 01:25:51,416
because it wants to teach you.

905
01:25:53,333 --> 01:25:55,458
Um, actually, what if we did

906
01:25:55,458 --> 01:25:56,625
this like something that's on

907
01:25:56,791 --> 01:25:58,375
side with it? No, you

908
01:25:58,375 --> 01:25:59,291
can't even broach the

909
01:25:59,458 --> 01:26:01,750
exhibit actually blocked it now. Before

910
01:26:01,750 --> 01:26:04,458
it was actually doing some

911
01:26:04,458 --> 01:26:05,833
real answers starting to spit

912
01:26:05,833 --> 01:26:07,625
out, you'd see it at a glance and then it

913
01:26:07,625 --> 01:26:09,000
would just be removed

914
01:26:09,000 --> 01:26:10,000
and stuff. But yeah,

915
01:26:10,000 --> 01:26:11,375
definitely worth checking out. It's just

916
01:26:11,375 --> 01:26:12,375
one of those other models.

917
01:26:12,375 --> 01:26:13,583
We're going to see more of

918
01:26:13,583 --> 01:26:15,500
these sovereign type models out there

919
01:26:15,500 --> 01:26:16,333
from different countries.

920
01:26:16,583 --> 01:26:18,625
I mean, Australia needs to

921
01:26:18,791 --> 01:26:20,916
try, you know, creating its own thing.

922
01:26:20,916 --> 01:26:21,833
There's folks out there

923
01:26:21,833 --> 01:26:22,958
that are, that are definitely

924
01:26:22,958 --> 01:26:25,000
getting into that space. So yeah,

925
01:26:25,000 --> 01:26:27,416
something worth having a look at as well.

926
01:26:27,875 --> 01:26:29,250
What's next for you?

927
01:26:29,500 --> 01:26:31,208
Well, I mean, we can't not talk about, I

928
01:26:31,208 --> 01:26:32,375
mean, we started with it,

929
01:26:32,375 --> 01:26:33,166
but we can't not talk about

930
01:26:33,208 --> 01:26:35,416
the executive orders and changes after

931
01:26:35,416 --> 01:26:35,875
it's been happening.

932
01:26:36,375 --> 01:26:37,583
Trump is already coming.

933
01:26:37,833 --> 01:26:38,166
Doing this stuff.

934
01:26:38,375 --> 01:26:39,458
All this stuff. This is it.

935
01:26:39,625 --> 01:26:41,375
The revoking of Biden's

936
01:26:41,375 --> 01:26:43,750
executive order on AI risks. Yeah.

937
01:26:44,083 --> 01:26:46,166
That is like, what are

938
01:26:46,166 --> 01:26:48,541
your thoughts? Hit me.

939
01:26:49,583 --> 01:26:51,666
From wearing the hat of an entrepreneur,

940
01:26:52,041 --> 01:26:53,583
tech founder and whatnot.

941
01:26:54,583 --> 01:26:56,958
Regulation does, depending on the type of

942
01:26:56,958 --> 01:26:58,791
regulation, typically

943
01:26:58,791 --> 01:27:01,083
lead to stifling innovation.

944
01:27:01,958 --> 01:27:04,458
However, there are times when regulation

945
01:27:04,458 --> 01:27:06,958
is needed. It'll depend.

946
01:27:07,333 --> 01:27:08,791
So there's a revocation

947
01:27:09,375 --> 01:27:11,791
taking away what Biden had put out in

948
01:27:11,791 --> 01:27:14,833
2023 to protect national

949
01:27:14,833 --> 01:27:17,166
security, so to speak. But

950
01:27:17,666 --> 01:27:19,625
it'll depend on what does get put in its

951
01:27:19,625 --> 01:27:20,375
place. Now they

952
01:27:20,375 --> 01:27:21,583
announced Project Stargate,

953
01:27:21,875 --> 01:27:24,000
which was an alliance between them,

954
01:27:24,541 --> 01:27:26,875
Oracle. I can't remember

955
01:27:26,875 --> 01:27:28,208
who else there is there.

956
01:27:28,708 --> 01:27:30,958
500 million dollar kind of project. Yep.

957
01:27:31,541 --> 01:27:35,000
So some interesting stuff there. So, you

958
01:27:35,000 --> 01:27:36,541
know, going back to the question, I think

959
01:27:36,958 --> 01:27:39,333
that we need to have some protections in

960
01:27:39,333 --> 01:27:41,333
place when it comes to AI. Otherwise,

961
01:27:41,333 --> 01:27:42,666
it is just going to be a Wild West. And

962
01:27:42,666 --> 01:27:44,208
we've seen what Wild West look

963
01:27:44,208 --> 01:27:45,500
like when it comes to crypto,

964
01:27:45,500 --> 01:27:47,791
for example, people grifting off each

965
01:27:47,791 --> 01:27:49,708
other, having no rules,

966
01:27:49,708 --> 01:27:51,500
just using this opportunity

967
01:27:53,041 --> 01:27:55,833
of the Wild West to basically be cowboys.

968
01:27:56,208 --> 01:27:57,041
That's right. Literally,

969
01:27:57,083 --> 01:27:58,166
figuratively, all that.

970
01:27:58,458 --> 01:28:01,500
So it's, so started with Sam Altman,

971
01:28:02,666 --> 01:28:05,750
obviously, CEO of OpenAI, the Oracle

972
01:28:05,750 --> 01:28:07,291
co-founder, Larry Ellison,

973
01:28:07,291 --> 01:28:09,000
which is this kind of a, he's a scary one

974
01:28:09,000 --> 01:28:11,333
to have in there. He wanted

975
01:28:11,333 --> 01:28:12,750
like constant surveillance.

976
01:28:13,166 --> 01:28:14,666
Yeah, yeah. He actually had a massive

977
01:28:14,666 --> 01:28:17,416
police, having police wear 24

978
01:28:17,416 --> 01:28:19,000
hour body camera type stuff.

979
01:28:19,041 --> 01:28:21,208
A fan of people also having camera

980
01:28:21,208 --> 01:28:21,791
footage, like

981
01:28:21,791 --> 01:28:23,000
following them around. So they,

982
01:28:23,000 --> 01:28:24,083
if you work at Oracle,

983
01:28:24,083 --> 01:28:26,125
do you have, I don't know.

984
01:28:26,125 --> 01:28:27,208
But direct, yeah, direct surveillance,

985
01:28:27,208 --> 01:28:28,583
constant surveillance within his Oracle

986
01:28:29,125 --> 01:28:31,083
business. That'll be interesting. And

987
01:28:31,083 --> 01:28:32,291
then of the SoftBank CEO,

988
01:28:33,291 --> 01:28:36,208
Master Yoshisan, who was kind enough to

989
01:28:36,208 --> 01:28:37,250
give Adam Newman how

990
01:28:37,250 --> 01:28:38,666
many billion dollars it was

991
01:28:38,708 --> 01:28:39,875
back in the day. A lot.

992
01:28:40,125 --> 01:28:42,041
Yeah. After a quick meeting.

993
01:28:42,500 --> 01:28:44,458
And the interesting thing is, so I think,

994
01:28:45,958 --> 01:28:47,208
I believe SoftBank is

995
01:28:47,208 --> 01:28:47,958
sort of infrastructure

996
01:28:48,583 --> 01:28:54,458
Oracle is software and I believe Sam

997
01:28:54,458 --> 01:28:57,541
Altman is managing it from a business

998
01:28:58,083 --> 01:29:01,291
infrastructure point of view. What is

999
01:29:01,291 --> 01:29:06,583
scary about this? Well, I mean, firstly,

1000
01:29:06,583 --> 01:29:08,458
like what Elon Musk came out saying,

1001
01:29:08,458 --> 01:29:09,458
which is pretty funny, like where did

1002
01:29:09,458 --> 01:29:10,708
that cash come from?

1003
01:29:11,291 --> 01:29:15,333
Particularly, like how is OpenAI able to

1004
01:29:15,333 --> 01:29:16,791
be a funding partner? Surely they're not

1005
01:29:16,791 --> 01:29:18,708
contributing any cash to it. They're not

1006
01:29:18,708 --> 01:29:21,000
in a positive cashflow state to be out of

1007
01:29:21,000 --> 01:29:21,916
through billions of dollars in

1008
01:29:21,916 --> 01:29:23,625
infrastructure. So it must be

1009
01:29:23,625 --> 01:29:26,750
more from a partnership America

1010
01:29:26,791 --> 01:29:29,541
controlling the most powerful AI company.

1011
01:29:30,041 --> 01:29:32,750
Give us control of the

1012
01:29:32,750 --> 01:29:33,791
business so that we can,

1013
01:29:35,250 --> 01:29:36,958
we solve in power over our

1014
01:29:36,958 --> 01:29:38,541
infrastructure. Where's Elon

1015
01:29:38,541 --> 01:29:39,750
as part of this, like given

1016
01:29:39,833 --> 01:29:41,458
how close he is to Donald?

1017
01:29:41,750 --> 01:29:42,541
He's supposed to be with Doge and

1018
01:29:42,541 --> 01:29:45,666
you know, having a big oversight of what

1019
01:29:45,666 --> 01:29:47,333
happens with cash in the country.

1020
01:29:47,333 --> 01:29:50,250
So anyway, so the, I think it's really

1021
01:29:50,250 --> 01:29:51,916
interesting though, because it does have

1022
01:29:51,916 --> 01:29:53,416
a lot of positivity.

1023
01:29:54,500 --> 01:29:55,541
You know, we talked touched on

1024
01:29:55,541 --> 01:29:56,875
innovation, so it has a lot of positivity

1025
01:29:56,875 --> 01:29:57,791
on innovation in this

1026
01:29:57,791 --> 01:29:59,250
space because now they're opening up huge

1027
01:29:59,250 --> 01:29:59,666
amounts of

1028
01:29:59,666 --> 01:30:01,500
infrastructure to actually explode

1029
01:30:01,916 --> 01:30:04,083
the possibilities of AI. I think the

1030
01:30:04,083 --> 01:30:05,166
scary part is these

1031
01:30:05,166 --> 01:30:08,125
particular people who are part of this

1032
01:30:08,625 --> 01:30:10,916
have some good agendas

1033
01:30:10,916 --> 01:30:12,208
and some bad agendas.

1034
01:30:12,250 --> 01:30:15,208
Which can really, you know, hinder

1035
01:30:15,208 --> 01:30:17,583
everything you do in the

1036
01:30:17,583 --> 01:30:19,083
world. And it's also giving major

1037
01:30:19,083 --> 01:30:20,625
control to a major companies, which is

1038
01:30:20,625 --> 01:30:22,000
something we talked about previously. If

1039
01:30:22,000 --> 01:30:23,250
you have the largest

1040
01:30:23,250 --> 01:30:25,208
corporations in the world controlling the

1041
01:30:25,208 --> 01:30:26,916
most powerful AI tools,

1042
01:30:27,375 --> 01:30:28,625
what does that mean for people,

1043
01:30:29,416 --> 01:30:32,083
our ability to, you know, create our own

1044
01:30:32,083 --> 01:30:32,833
businesses off the

1045
01:30:32,833 --> 01:30:34,041
back of things, but also

1046
01:30:35,458 --> 01:30:37,041
have control of the things that we do if

1047
01:30:37,041 --> 01:30:38,000
this AI tool is going to

1048
01:30:38,000 --> 01:30:39,541
have ultimate oversight of

1049
01:30:39,708 --> 01:30:41,291
everything that you do. And it's

1050
01:30:41,291 --> 01:30:42,833
controlled by one business. They have all

1051
01:30:42,833 --> 01:30:44,000
the data, all the insight.

1052
01:30:44,583 --> 01:30:45,291
One business, the

1053
01:30:45,291 --> 01:30:46,500
United States of America.

1054
01:30:46,791 --> 01:30:47,500
That's it. That's it.

1055
01:30:48,416 --> 01:30:49,541
Yeah, it's going to be interesting. It's

1056
01:30:49,541 --> 01:30:52,541
really a lot about, you

1057
01:30:52,541 --> 01:30:54,791
know, how does open source

1058
01:30:54,875 --> 01:30:56,958
counteract this? How does open source

1059
01:30:56,958 --> 01:30:59,000
work with it? How do we

1060
01:30:59,000 --> 01:31:00,041
actually make it so that other

1061
01:31:00,041 --> 01:31:02,291
businesses can get in on this kind of

1062
01:31:02,291 --> 01:31:04,208
stuff? So I think there's

1063
01:31:04,208 --> 01:31:07,583
definitely room there to dive in.

1064
01:31:07,625 --> 01:31:10,375
I mean, it's only like a few days, you

1065
01:31:10,375 --> 01:31:11,958
know, since the event.

1066
01:31:11,958 --> 01:31:12,458
What do you got there?

1067
01:31:12,708 --> 01:31:13,625
I was going to play the

1068
01:31:13,625 --> 01:31:15,000
video, but oh, yeah, go ahead.

1069
01:31:15,000 --> 01:31:16,208
No, there's an ad. It's coming up with an

1070
01:31:16,208 --> 01:31:17,000
ad. Don't worry about it.

1071
01:31:17,000 --> 01:31:18,000
I don't play the ad. No.

1072
01:31:18,375 --> 01:31:19,541
No second ad. Let's go.

1073
01:31:21,000 --> 01:31:22,625
But yeah, I think it's interesting with

1074
01:31:22,625 --> 01:31:23,916
the, you know, the executive

1075
01:31:23,916 --> 01:31:25,000
orders and just going back to

1076
01:31:25,000 --> 01:31:26,416
it, like there's the big old signature

1077
01:31:26,416 --> 01:31:29,333
and stuff like seriously, it's so good.

1078
01:31:31,375 --> 01:31:32,458
So 200 of those great,

1079
01:31:32,500 --> 01:31:34,416
great stuff. But yeah, basically there

1080
01:31:34,416 --> 01:31:36,250
was I think it was like

1081
01:31:36,250 --> 01:31:38,458
late in the year as well.

1082
01:31:38,458 --> 01:31:40,041
So that's probably gone. But there was

1083
01:31:40,041 --> 01:31:43,000
some rule from Biden wanted

1084
01:31:43,000 --> 01:31:44,625
to put out what was it? It was

1085
01:31:44,625 --> 01:31:47,041
like not I don't know if it was like a

1086
01:31:47,041 --> 01:31:48,125
law, but they wanted to put

1087
01:31:48,125 --> 01:31:49,208
something out there, but it's

1088
01:31:49,208 --> 01:31:51,125
like it's so last minute. Like I don't

1089
01:31:51,125 --> 01:31:52,041
think it was going to get through, but

1090
01:31:52,041 --> 01:31:52,583
something to do with

1091
01:31:52,583 --> 01:31:55,083
like more protections and stuff. So in

1092
01:31:55,083 --> 01:31:56,791
any case, what I do think

1093
01:31:56,791 --> 01:31:58,041
is interesting is if you are

1094
01:31:58,083 --> 01:32:00,875
looking at following what's going on in

1095
01:32:00,875 --> 01:32:03,708
the law in terms of AI, got

1096
01:32:03,708 --> 01:32:05,208
to definitely follow Raymond's

1097
01:32:05,208 --> 01:32:08,333
son here, who's a lawyer, AI builder, one

1098
01:32:08,333 --> 01:32:09,333
of the directors of data

1099
01:32:09,333 --> 01:32:11,583
science and AI, where they,

1100
01:32:12,166 --> 01:32:14,291
a great community kind of group are going

1101
01:32:14,291 --> 01:32:16,750
to be starting to run events again.

1102
01:32:17,333 --> 01:32:19,125
But basically he built something which

1103
01:32:19,125 --> 01:32:21,708
was a general AI regulation

1104
01:32:21,708 --> 01:32:24,541
tracker. And the tools that

1105
01:32:24,583 --> 01:32:27,125
he pieced together to build that he's now

1106
01:32:27,125 --> 01:32:29,250
making that open source or

1107
01:32:29,250 --> 01:32:31,000
open source. I think it's open

1108
01:32:31,000 --> 01:32:32,875
source or like you can subscribe to

1109
01:32:32,875 --> 01:32:34,000
actually use the same tools.

1110
01:32:34,000 --> 01:32:35,500
So you can build your own maps

1111
01:32:35,500 --> 01:32:38,166
thing. So note to map here. So he built

1112
01:32:38,166 --> 01:32:39,625
this regulation tracker and

1113
01:32:39,625 --> 01:32:40,708
it's the same kind of tools

1114
01:32:40,708 --> 01:32:42,000
where whether you want to track

1115
01:32:42,000 --> 01:32:43,708
regulation or you want to track like

1116
01:32:43,708 --> 01:32:44,833
different types of things, you

1117
01:32:44,833 --> 01:32:47,000
can build it with this mapping feature

1118
01:32:47,000 --> 01:32:49,000
that has news articles underneath and

1119
01:32:49,000 --> 01:32:50,458
alerts your subscribers.

1120
01:32:50,500 --> 01:32:53,583
So well worth a look at there because for

1121
01:32:53,583 --> 01:32:54,750
me to keep up with what's

1122
01:32:54,750 --> 01:32:55,708
going on around the world,

1123
01:32:55,708 --> 01:32:58,500
I use the regulation tracker as well. So

1124
01:32:58,500 --> 01:33:00,666
worth checking out and

1125
01:33:00,666 --> 01:33:02,125
Raymond, I will take your payment

1126
01:33:02,166 --> 01:33:04,375
later. A few more things on what's

1127
01:33:04,375 --> 01:33:07,708
happening with Starkey. So last year, I

1128
01:33:07,708 --> 01:33:08,166
know I went through,

1129
01:33:08,375 --> 01:33:10,375
it's not a big Amazon and Google, like

1130
01:33:10,375 --> 01:33:11,333
they've invested billions of

1131
01:33:11,333 --> 01:33:12,208
dollars into infrastructure

1132
01:33:12,416 --> 01:33:14,333
just to improve how they even work on AI.

1133
01:33:14,500 --> 01:33:15,583
So it's interesting to see

1134
01:33:15,583 --> 01:33:17,416
that how this is sort of adding

1135
01:33:17,541 --> 01:33:19,875
to that many, probably going to be

1136
01:33:19,875 --> 01:33:21,166
trillions of dollars over the

1137
01:33:21,166 --> 01:33:22,208
next couple of years invested

1138
01:33:22,208 --> 01:33:25,375
into infrastructure across the globe. But

1139
01:33:25,375 --> 01:33:26,041
they're going to be building

1140
01:33:26,041 --> 01:33:29,250
up to 10 data centers within

1141
01:33:29,250 --> 01:33:31,208
the next this like that's kicking off now

1142
01:33:31,208 --> 01:33:32,333
with another 10 that will

1143
01:33:32,333 --> 01:33:33,583
follow. So it's going to open

1144
01:33:33,583 --> 01:33:36,250
up huge amounts of job opportunities

1145
01:33:36,250 --> 01:33:37,375
throughout the connoisseur with this

1146
01:33:37,375 --> 01:33:38,083
amount of investment,

1147
01:33:38,083 --> 01:33:39,458
which is like the positive side of

1148
01:33:39,458 --> 01:33:41,333
things. But then on the flip side, you

1149
01:33:41,333 --> 01:33:42,041
have the negative side

1150
01:33:42,083 --> 01:33:44,250
of things, which is, you know, there's a

1151
01:33:44,250 --> 01:33:45,458
negative, there's always a

1152
01:33:45,458 --> 01:33:46,583
negative. And then again,

1153
01:33:47,208 --> 01:33:49,166
control corporates who you know, where

1154
01:33:49,166 --> 01:33:51,250
this money is coming from and what what

1155
01:33:51,250 --> 01:33:52,208
the actual agenda is

1156
01:33:52,208 --> 01:33:53,625
at the end of the day with the US. But

1157
01:33:53,625 --> 01:33:54,791
you know, there's, which is what we

1158
01:33:54,791 --> 01:33:55,541
talked about before,

1159
01:33:55,791 --> 01:33:57,291
this is if we're in a moment of race,

1160
01:33:57,500 --> 01:33:59,416
this is like a huge race to

1161
01:33:59,416 --> 01:34:02,125
who can get to what level of AI

1162
01:34:02,166 --> 01:34:04,000
first, whether it's AGI, you know,

1163
01:34:04,000 --> 01:34:05,416
they're saying that whoever gets AGI

1164
01:34:05,416 --> 01:34:06,125
first, we talked about this

1165
01:34:06,125 --> 01:34:08,541
area will get to ASI. And then what

1166
01:34:08,541 --> 01:34:09,666
happens after ASI, who

1167
01:34:09,666 --> 01:34:11,000
gets to control the ASI?

1168
01:34:11,500 --> 01:34:13,375
If I mean, probably can't be controlled

1169
01:34:13,375 --> 01:34:16,041
by that point. It's fast surpassing.

1170
01:34:16,333 --> 01:34:17,208
The one that will

1171
01:34:17,208 --> 01:34:18,541
control it is AI itself.

1172
01:34:18,541 --> 01:34:19,791
Yeah, it will control itself. Exactly. So

1173
01:34:19,791 --> 01:34:20,541
it won't be a corporation.

1174
01:34:21,125 --> 01:34:23,000
So we'll see what can happen there.

1175
01:34:23,000 --> 01:34:24,000
That's there's a, you know,

1176
01:34:24,166 --> 01:34:25,333
double sided coin here. But

1177
01:34:25,916 --> 01:34:27,208
in the short term, I know the triple

1178
01:34:27,208 --> 01:34:28,041
sided coin, Chris,

1179
01:34:28,333 --> 01:34:29,666
because coins aren't have three.

1180
01:34:29,750 --> 01:34:31,500
Well, actually, technically, they do.

1181
01:34:32,791 --> 01:34:33,583
Technically, that's three sides.

1182
01:34:33,833 --> 01:34:35,708
I think in an AI world, we'll get more of

1183
01:34:35,708 --> 01:34:37,666
those landing on its side kind of things.

1184
01:34:38,625 --> 01:34:40,375
No, that that is definitely like

1185
01:34:40,375 --> 01:34:41,333
interesting and something

1186
01:34:41,333 --> 01:34:42,833
that we'll be keeping an eye on.

1187
01:34:43,375 --> 01:34:45,208
Not meaning to get like political with

1188
01:34:45,208 --> 01:34:46,250
all this kind of stuff.

1189
01:34:46,250 --> 01:34:47,333
But I think it is worth

1190
01:34:47,916 --> 01:34:50,333
keeping an eye on what is happening

1191
01:34:50,333 --> 01:34:51,500
through regulation, what

1192
01:34:51,500 --> 01:34:52,458
are happening in movements

1193
01:34:52,458 --> 01:34:55,916
around government views, stances. I mean,

1194
01:34:55,916 --> 01:34:57,125
late last year, we had

1195
01:34:57,125 --> 01:34:59,250
the Senate Select Committee

1196
01:34:59,250 --> 01:35:01,833
here in Australia basically give their 13

1197
01:35:01,833 --> 01:35:02,958
recommendations, which

1198
01:35:02,958 --> 01:35:04,125
unfortunately, I mean,

1199
01:35:04,416 --> 01:35:07,625
for me, others may be differently. But

1200
01:35:07,625 --> 01:35:08,625
unfortunately, said

1201
01:35:08,625 --> 01:35:10,000
that just general purpose

1202
01:35:10,708 --> 01:35:14,333
tools. Sounds like GPT, but general

1203
01:35:14,333 --> 01:35:16,083
purpose, general purpose AI

1204
01:35:16,083 --> 01:35:17,041
tools, I think they actually

1205
01:35:17,041 --> 01:35:18,666
classified as GPA, I remember was like

1206
01:35:18,666 --> 01:35:19,875
general purpose, but just

1207
01:35:19,875 --> 01:35:21,583
normal chat, GPT, just normal

1208
01:35:21,583 --> 01:35:23,750
claw, just normal, I guess, deep seek

1209
01:35:23,750 --> 01:35:25,708
even would come under

1210
01:35:25,708 --> 01:35:27,083
the banner of high risk,

1211
01:35:27,625 --> 01:35:29,500
which is interesting, because many of us

1212
01:35:29,500 --> 01:35:31,166
builders, we thought that

1213
01:35:31,166 --> 01:35:33,375
it would be literal high risk

1214
01:35:33,458 --> 01:35:36,041
tools that are being built purposely for

1215
01:35:36,041 --> 01:35:38,083
use in certain industries.

1216
01:35:38,083 --> 01:35:39,000
And I guess you could argue

1217
01:35:39,000 --> 01:35:41,000
that chat, GPT could be used in finance,

1218
01:35:41,000 --> 01:35:44,208
healthcare, legal, but having all of

1219
01:35:44,208 --> 01:35:45,291
those come under the

1220
01:35:45,291 --> 01:35:48,708
banner of high risk, depending on what

1221
01:35:48,708 --> 01:35:50,958
the rules are, as to what

1222
01:35:50,958 --> 01:35:53,750
high risk needs to be doing, like

1223
01:35:53,750 --> 01:35:56,125
you need to prove your weights, what you

1224
01:35:56,125 --> 01:35:57,166
were trained on, blah,

1225
01:35:57,166 --> 01:35:58,208
blah, blah, blah, I can see

1226
01:35:58,250 --> 01:35:59,791
something happening where just some

1227
01:35:59,791 --> 01:36:02,250
models might not be available. If

1228
01:36:02,250 --> 01:36:03,791
everything all of a sudden

1229
01:36:03,791 --> 01:36:05,625
falls under high risk, and the scrutiny

1230
01:36:05,625 --> 01:36:09,041
level is this high, and overseas

1231
01:36:09,041 --> 01:36:10,083
companies that are providing

1232
01:36:10,083 --> 01:36:12,166
really great tech to Australia, it just

1233
01:36:12,166 --> 01:36:13,541
becomes unavailable. So

1234
01:36:13,541 --> 01:36:15,500
maybe we need to even VPN in or

1235
01:36:15,583 --> 01:36:17,625
whatever, I don't quite know yet, but

1236
01:36:17,625 --> 01:36:19,958
nothing in concrete, it's just

1237
01:36:19,958 --> 01:36:21,375
recommendations. But

1238
01:36:21,375 --> 01:36:23,625
recommendations lead to regulations. So

1239
01:36:23,625 --> 01:36:26,250
we'll see. What's next? I

1240
01:36:26,250 --> 01:36:26,791
was gonna say, what have you

1241
01:36:26,833 --> 01:36:28,250
got on there? What's on the books? What's

1242
01:36:28,250 --> 01:36:29,041
on there? Oh, I've got a

1243
01:36:29,041 --> 01:36:31,416
few things. Let's go cook

1244
01:36:31,416 --> 01:36:33,666
the books. Let's go to this one here.

1245
01:36:33,666 --> 01:36:34,791
Let's get some sound on. So

1246
01:36:34,791 --> 01:36:36,750
this is back over to Instagram.

1247
01:36:37,000 --> 01:36:39,000
So let's just play this one here. Output

1248
01:36:39,000 --> 01:36:42,625
device is fine. Try again.

1249
01:36:44,000 --> 01:36:47,250
Here we go. So this one here,

1250
01:36:47,250 --> 01:36:52,041
folks, is an AI shooter in real time with

1251
01:36:52,041 --> 01:36:53,291
a kid like a future of learning.

1252
01:37:03,583 --> 01:37:10,500
I like this. I love this thing. It's a

1253
01:37:10,500 --> 01:37:16,333
faffing joke. Yeah, I know

1254
01:37:16,333 --> 01:37:17,208
it's going to be about square

1255
01:37:17,208 --> 01:37:22,500
numbers. Okay. I wish I looked this way.

1256
01:37:28,916 --> 01:37:29,458
Learning is fun!

1257
01:37:37,375 --> 01:37:38,875
I mean we've seen stuff like this with

1258
01:37:38,875 --> 01:37:40,875
when OpenAI originally

1259
01:37:40,875 --> 01:37:42,250
released their voice models

1260
01:37:42,250 --> 01:37:42,958
and...

1261
01:37:42,958 --> 01:37:43,458
Yes, yeah.

1262
01:37:44,250 --> 01:37:45,833
You know they were putting through an

1263
01:37:45,833 --> 01:37:47,291
image of a complex

1264
01:37:47,291 --> 01:37:48,375
question as well, hey, can you

1265
01:37:48,375 --> 01:37:49,250
help train the kids?

1266
01:37:49,250 --> 01:37:49,916
So this is just adding

1267
01:37:49,916 --> 01:37:51,208
that extra sort of...

1268
01:37:51,666 --> 01:37:52,291
Another layout.

1269
01:37:52,500 --> 01:37:54,250
Another layout of relating to them in a

1270
01:37:54,250 --> 01:37:55,125
more meaningful way by

1271
01:37:55,125 --> 01:37:56,833
giving visuals and interactions

1272
01:37:56,958 --> 01:37:58,375
and gamifying the experience.

1273
01:37:58,375 --> 01:37:58,666
Exactly.

1274
01:37:58,916 --> 01:38:00,666
I think it's a great thing but I have no

1275
01:38:00,666 --> 01:38:02,000
doubt as well that it'll

1276
01:38:02,000 --> 01:38:03,166
be the classic thing the

1277
01:38:03,208 --> 01:38:06,000
kids do which is, here's the present,

1278
01:38:06,375 --> 01:38:07,208
they want to play with

1279
01:38:07,208 --> 01:38:09,500
the wrapping and the bubble

1280
01:38:09,541 --> 01:38:10,708
wrap and all that kind of stuff.

1281
01:38:11,833 --> 01:38:13,166
They'll find a way that,

1282
01:38:13,166 --> 01:38:14,541
"Oh, this is just not fun."

1283
01:38:14,750 --> 01:38:16,333
I'm sure we can make it more fun and

1284
01:38:16,333 --> 01:38:17,666
gamified and stuff but

1285
01:38:17,666 --> 01:38:18,666
there's always going to be...

1286
01:38:18,666 --> 01:38:19,666
It's not going to be for everyone.

1287
01:38:19,666 --> 01:38:23,416
I'm not a psychologist by any factor or

1288
01:38:23,416 --> 01:38:24,833
form so take what I'm about to say.

1289
01:38:25,166 --> 01:38:26,916
Just take out the lodgest part.

1290
01:38:26,916 --> 01:38:27,625
Little bit of grain.

1291
01:38:28,625 --> 01:38:29,666
There's nothing here that

1292
01:38:29,666 --> 01:38:30,916
has anything to back it.

1293
01:38:31,208 --> 01:38:32,083
But if I apply this

1294
01:38:32,083 --> 01:38:33,375
to my understanding...

1295
01:38:33,375 --> 01:38:34,166
I mean, if you think about it from a

1296
01:38:34,166 --> 01:38:35,000
social media perspective,

1297
01:38:35,000 --> 01:38:36,333
our ability, our need and

1298
01:38:36,333 --> 01:38:39,375
desire to swipe and have that fast, fast

1299
01:38:39,375 --> 01:38:41,583
like rush of, "Oh, I'm so

1300
01:38:41,583 --> 01:38:42,625
entertained in that particular

1301
01:38:42,791 --> 01:38:42,958
moment."

1302
01:38:43,916 --> 01:38:45,500
My view is like looking at this is that

1303
01:38:45,500 --> 01:38:46,541
you're applying that to

1304
01:38:46,541 --> 01:38:48,000
education which means that

1305
01:38:48,000 --> 01:38:50,041
you're not longer learning.

1306
01:38:50,416 --> 01:38:54,250
You're not learning out of the need for

1307
01:38:54,250 --> 01:38:55,416
learning and to become

1308
01:38:55,416 --> 01:38:57,416
educated and just understanding

1309
01:38:57,541 --> 01:38:59,625
that there is a part of your life and

1310
01:38:59,625 --> 01:39:00,833
your world that has to

1311
01:39:00,833 --> 01:39:01,875
go out there and actively

1312
01:39:02,041 --> 01:39:03,958
do something without that gratification.

1313
01:39:04,250 --> 01:39:05,583
It has to experience that.

1314
01:39:05,583 --> 01:39:07,166
And now you're putting gratification into

1315
01:39:07,166 --> 01:39:08,958
education which means if you're having to

1316
01:39:08,958 --> 01:39:11,000
do something now mundane, it probably

1317
01:39:11,000 --> 01:39:13,291
makes it even harder for

1318
01:39:13,291 --> 01:39:14,291
the student to go, "Oh,

1319
01:39:14,291 --> 01:39:15,041
now I've just got to

1320
01:39:15,041 --> 01:39:16,250
do a bit of studying."

1321
01:39:16,250 --> 01:39:17,250
I have to come into the

1322
01:39:17,250 --> 01:39:18,500
office today in a row.

1323
01:39:18,500 --> 01:39:19,000
Yeah, I'm a find.

1324
01:39:19,000 --> 01:39:20,083
I want to gamify this experience.

1325
01:39:20,458 --> 01:39:21,791
Like resilience is what

1326
01:39:21,791 --> 01:39:22,958
we're talking about here.

1327
01:39:22,958 --> 01:39:23,291
Yeah, that's actually a

1328
01:39:23,291 --> 01:39:24,125
good way to put resilience.

1329
01:39:24,750 --> 01:39:27,000
Because there's a, you know, and I will

1330
01:39:27,000 --> 01:39:28,666
classify and I know it's

1331
01:39:28,666 --> 01:39:29,916
not everyone but this is just

1332
01:39:29,916 --> 01:39:31,791
what reports, you know, and articles talk

1333
01:39:31,791 --> 01:39:34,000
about that Gen Z and

1334
01:39:34,000 --> 01:39:35,458
younger generation kind

1335
01:39:35,458 --> 01:39:39,250
of workers are less used to coming into

1336
01:39:39,250 --> 01:39:40,125
the office for example

1337
01:39:40,125 --> 01:39:42,041
and less used to criticism.

1338
01:39:42,041 --> 01:39:44,083
And obviously it is not everyone, right?

1339
01:39:44,625 --> 01:39:45,583
But on average, this is

1340
01:39:45,583 --> 01:39:46,791
what is being reported.

1341
01:39:46,791 --> 01:39:49,083
Now obviously you can say, "Well, where

1342
01:39:49,083 --> 01:39:50,000
did the numbers come from?

1343
01:39:50,000 --> 01:39:51,000
Is it self-reported?

1344
01:39:51,000 --> 01:39:52,541
Is it just observations from others?

1345
01:39:52,541 --> 01:39:53,500
Is it just boomers?"

1346
01:39:54,333 --> 01:39:56,000
Folks a little bit older than us.

1347
01:39:56,000 --> 01:39:59,583
And us making the reports and making

1348
01:39:59,583 --> 01:40:00,916
these observations and stuff.

1349
01:40:01,291 --> 01:40:03,541
Like I don't know about that but that is

1350
01:40:03,541 --> 01:40:04,458
definitely one observation

1351
01:40:04,458 --> 01:40:05,791
and you can see with this

1352
01:40:05,791 --> 01:40:07,833
that potentially that kind of stuff does

1353
01:40:07,833 --> 01:40:09,833
lead to changes like that.

1354
01:40:10,041 --> 01:40:12,625
But when we were younger, the things that

1355
01:40:12,625 --> 01:40:14,000
we had access to were

1356
01:40:14,000 --> 01:40:14,958
even the thought of,

1357
01:40:14,958 --> 01:40:15,791
"Oh my God, they're bringing their

1358
01:40:15,791 --> 01:40:16,666
laptops to school where

1359
01:40:16,666 --> 01:40:18,000
they're just going to game,

1360
01:40:18,000 --> 01:40:19,291
they're not going to do stuff."

1361
01:40:19,291 --> 01:40:20,583
So maybe we do need to

1362
01:40:20,583 --> 01:40:21,875
take it with a grain of salt.

1363
01:40:22,708 --> 01:40:23,291
I don't think it's going

1364
01:40:23,291 --> 01:40:24,541
to be over for education.

1365
01:40:24,541 --> 01:40:25,375
I think it will adjust.

1366
01:40:25,625 --> 01:40:27,083
But hopefully there is still some

1367
01:40:27,083 --> 01:40:30,625
resilience that is put in there as well.

1368
01:40:30,625 --> 01:40:32,625
Like, because not everything has to be

1369
01:40:32,625 --> 01:40:34,791
100% it works out straight away.

1370
01:40:35,375 --> 01:40:37,666
Otherwise, when something in life outside

1371
01:40:37,666 --> 01:40:38,583
of school happens

1372
01:40:38,583 --> 01:40:39,458
where it doesn't go your

1373
01:40:39,458 --> 01:40:40,833
way and you're not used to dealing with

1374
01:40:40,833 --> 01:40:42,458
it, like school is the

1375
01:40:42,458 --> 01:40:44,208
great arbiter of teaching.

1376
01:40:44,541 --> 01:40:46,625
Like you had a horrible time in school.

1377
01:40:47,500 --> 01:40:49,583
Oftentimes you have these harder lessons

1378
01:40:49,583 --> 01:40:50,458
on life you learn

1379
01:40:50,458 --> 01:40:51,750
early rather than having an

1380
01:40:51,791 --> 01:40:52,500
easy time.

1381
01:40:52,500 --> 01:40:53,041
Exactly right.

1382
01:40:53,041 --> 01:40:55,000
And you think about it from just a point

1383
01:40:55,000 --> 01:40:55,958
of innovation and how

1384
01:40:55,958 --> 01:40:58,916
humanity evolves and becomes

1385
01:40:58,916 --> 01:40:59,666
more intelligent.

1386
01:40:59,666 --> 01:41:01,000
We do it from failure.

1387
01:41:02,375 --> 01:41:04,958
Failure trains and helps

1388
01:41:04,958 --> 01:41:06,166
you to learn and improve.

1389
01:41:06,416 --> 01:41:08,166
And if you're constantly, if you're

1390
01:41:08,166 --> 01:41:09,208
removing that element of

1391
01:41:09,208 --> 01:41:10,625
failure, whether it's the

1392
01:41:10,625 --> 01:41:13,000
kid always being told, "Yes, it's great

1393
01:41:13,000 --> 01:41:14,000
that you came 10th,

1394
01:41:14,000 --> 01:41:14,875
that's it, that's a way."

1395
01:41:15,750 --> 01:41:18,875
Or putting gamification elements to

1396
01:41:18,875 --> 01:41:20,208
remove that association

1397
01:41:20,208 --> 01:41:21,916
of that grit to actually

1398
01:41:21,916 --> 01:41:23,666
get study done or get your

1399
01:41:23,666 --> 01:41:25,250
brain into education mode.

1400
01:41:25,708 --> 01:41:26,708
It does have win-miss-out.

1401
01:41:27,375 --> 01:41:30,291
Because then that real fear of failure

1402
01:41:30,291 --> 01:41:31,208
sets in and that's when

1403
01:41:31,208 --> 01:41:32,083
you get depression and

1404
01:41:32,083 --> 01:41:33,000
all these other elements.

1405
01:41:33,000 --> 01:41:34,458
But again, not a psychologist, this is

1406
01:41:34,458 --> 01:41:35,708
just purely from

1407
01:41:35,708 --> 01:41:37,666
observation, or you got UN from

1408
01:41:37,666 --> 01:41:38,541
some stuff I've read.

1409
01:41:39,791 --> 01:41:41,583
Professional advice here, folks.

1410
01:41:41,583 --> 01:41:42,875
Take professional advice before you

1411
01:41:42,875 --> 01:41:43,541
listen to anything that

1412
01:41:43,541 --> 01:41:44,416
we're saying in that particular

1413
01:41:44,541 --> 01:41:44,750
space.

1414
01:41:45,500 --> 01:41:45,833
Yeah, first.

1415
01:41:46,500 --> 01:41:49,000
Another topic just around, less on

1416
01:41:49,000 --> 01:41:50,000
education, but this one's

1417
01:41:50,000 --> 01:41:51,000
more around like medicine.

1418
01:41:51,000 --> 01:41:53,916
And I think overall just how AIs work.

1419
01:41:53,916 --> 01:41:56,916
But David Shapiro here who, I think he's

1420
01:41:56,916 --> 01:41:57,583
still doing his YouTube

1421
01:41:57,583 --> 01:41:58,750
channel, but he was doing

1422
01:41:58,750 --> 01:42:00,833
a lot of like AI stuff and then getting

1423
01:42:00,833 --> 01:42:01,625
into more of the

1424
01:42:01,625 --> 01:42:03,000
doomerism that AI is going to

1425
01:42:03,000 --> 01:42:03,375
take over.

1426
01:42:03,666 --> 01:42:05,916
So a lot of people,

1427
01:42:06,416 --> 01:42:09,541
that works well with them.

1428
01:42:10,208 --> 01:42:10,833
A lot of people that

1429
01:42:10,833 --> 01:42:11,750
turn them off and stuff.

1430
01:42:11,958 --> 01:42:13,041
But hey, he's got his view.

1431
01:42:13,083 --> 01:42:14,625
And I think it's great that he shares

1432
01:42:14,625 --> 01:42:15,291
that because he'd

1433
01:42:15,291 --> 01:42:16,625
rather share views than just

1434
01:42:16,625 --> 01:42:17,625
see the one side only.

1435
01:42:17,625 --> 01:42:19,250
I think it's great to get both sides.

1436
01:42:19,791 --> 01:42:22,291
In any case, he's talking here about AI

1437
01:42:22,291 --> 01:42:23,625
as a pattern recognizer

1438
01:42:23,625 --> 01:42:24,750
and people are dismissing

1439
01:42:24,875 --> 01:42:25,208
it.

1440
01:42:25,625 --> 01:42:27,500
Saying, "Oh, it just recognizes patterns.

1441
01:42:27,500 --> 01:42:28,375
It's not doing anything new."

1442
01:42:28,375 --> 01:42:30,000
It's like, "Well, what do you think we do

1443
01:42:30,000 --> 01:42:31,208
in things like

1444
01:42:31,208 --> 01:42:33,708
medicine, radiography, other

1445
01:42:33,708 --> 01:42:34,875
kind of like fields?"

1446
01:42:34,875 --> 01:42:35,958
We are basically

1447
01:42:35,958 --> 01:42:37,791
pattern recognition machines.

1448
01:42:38,083 --> 01:42:40,625
It's why early AI, and

1449
01:42:40,625 --> 01:42:41,750
I'm not talking generative.

1450
01:42:41,791 --> 01:42:44,958
I'm talking like image recognition and

1451
01:42:44,958 --> 01:42:45,750
things like that, why

1452
01:42:45,750 --> 01:42:47,916
humans are far better at doing

1453
01:42:47,916 --> 01:42:49,500
things like classifying galaxies.

1454
01:42:50,041 --> 01:42:52,000
There was a gamified thing where you got

1455
01:42:52,000 --> 01:42:53,458
points as a human for

1456
01:42:53,458 --> 01:42:54,958
classifying galaxies because

1457
01:42:54,958 --> 01:42:55,958
it was very hard for

1458
01:42:55,958 --> 01:42:57,958
machines to do that kind of stuff.

1459
01:42:58,458 --> 01:42:59,833
So now they're getting better at pattern

1460
01:42:59,833 --> 01:43:01,333
recognition, but there's

1461
01:43:01,333 --> 01:43:02,250
this complaint, "Oh, they're

1462
01:43:02,250 --> 01:43:03,291
just like parrot monkeys.

1463
01:43:03,291 --> 01:43:04,208
They just look at

1464
01:43:04,208 --> 01:43:05,208
patterns and repeat it."

1465
01:43:05,208 --> 01:43:07,416
And it's just like, "Pot,

1466
01:43:07,833 --> 01:43:09,500
kettle, you're both black.

1467
01:43:10,250 --> 01:43:11,125
Nice to meet you."

1468
01:43:11,375 --> 01:43:11,958
You know what I mean?

1469
01:43:12,875 --> 01:43:14,041
This is one of the first things I

1470
01:43:14,041 --> 01:43:14,916
remember before we even

1471
01:43:14,916 --> 01:43:16,583
had this show that I think

1472
01:43:17,291 --> 01:43:20,000
whether I said it online, but definitely

1473
01:43:20,000 --> 01:43:20,958
in discussion, people

1474
01:43:20,958 --> 01:43:23,208
that we are just learning

1475
01:43:23,500 --> 01:43:24,958
the irony is we're learning more about

1476
01:43:24,958 --> 01:43:26,666
ourselves and how we think.

1477
01:43:27,041 --> 01:43:28,750
We don't necessarily put our information

1478
01:43:28,750 --> 01:43:31,958
like this where you

1479
01:43:31,958 --> 01:43:33,958
need so much more training

1480
01:43:33,958 --> 01:43:36,041
to just get these things to do stuff and

1481
01:43:36,041 --> 01:43:37,708
you just have to brute force it.

1482
01:43:37,708 --> 01:43:39,291
There's different ways that we're wired.

1483
01:43:40,000 --> 01:43:42,333
In terms of the output, it definitely

1484
01:43:42,333 --> 01:43:44,250
follows that pattern of,

1485
01:43:44,500 --> 01:43:45,791
"Okay, Chris said this.

1486
01:43:46,208 --> 01:43:48,083
Here is my prepared kind of responses."

1487
01:43:48,291 --> 01:43:49,125
And then I'm constantly

1488
01:43:49,125 --> 01:43:51,125
shuffling things around in my head.

1489
01:43:51,875 --> 01:43:53,750
And that's seemingly what these things

1490
01:43:53,750 --> 01:43:54,500
do, whether it's for

1491
01:43:54,500 --> 01:43:56,291
images or for just responses

1492
01:43:56,458 --> 01:43:58,000
like they are pattern matching.

1493
01:43:59,083 --> 01:44:01,208
So it's a reflection of, "Well, it sounds

1494
01:44:01,208 --> 01:44:02,750
like us because freak."

1495
01:44:02,750 --> 01:44:04,750
Because we trained it all through the

1496
01:44:04,750 --> 01:44:05,541
hours, the stuff that

1497
01:44:05,541 --> 01:44:06,250
we have already done.

1498
01:44:06,250 --> 01:44:07,041
Did you remember the thing where it's

1499
01:44:07,041 --> 01:44:07,833
like if you gave it

1500
01:44:07,833 --> 01:44:09,458
incentives, like people maybe

1501
01:44:09,458 --> 01:44:11,625
forget this, but it was giving better

1502
01:44:11,625 --> 01:44:13,000
results if you said, "Well,

1503
01:44:13,000 --> 01:44:14,750
give me a $1,000 response.

1504
01:44:14,750 --> 01:44:16,041
Now give me a $10,000

1505
01:44:16,041 --> 01:44:17,625
or a $100,000 response."

1506
01:44:17,625 --> 01:44:18,000
Guess what?

1507
01:44:18,000 --> 01:44:19,333
The $100,000 response was

1508
01:44:19,333 --> 01:44:21,083
like a way better quality one.

1509
01:44:21,291 --> 01:44:21,833
That's really funny.

1510
01:44:22,208 --> 01:44:22,833
Because we're human.

1511
01:44:23,333 --> 01:44:24,458
We have these incentives.

1512
01:44:24,791 --> 01:44:26,041
Anyway, what's next for you?

1513
01:44:26,500 --> 01:44:28,833
Just some minor things out there.

1514
01:44:28,833 --> 01:44:29,958
So Perplexity just launched

1515
01:44:29,958 --> 01:44:32,708
Sonar, an API for AI search.

1516
01:44:32,708 --> 01:44:34,500
So for all the developers and enterprise

1517
01:44:34,500 --> 01:44:35,416
businesses out there.

1518
01:44:35,958 --> 01:44:37,833
You now have an API to be able to start

1519
01:44:37,833 --> 01:44:39,208
testing and creating

1520
01:44:39,208 --> 01:44:40,875
applications or services.

1521
01:44:40,875 --> 01:44:41,375
That is interesting.

1522
01:44:42,000 --> 01:44:45,208
They're quite a powerful RAG model.

1523
01:44:45,708 --> 01:44:47,583
I really love Perplexity.

1524
01:44:47,583 --> 01:44:48,166
I still use it.

1525
01:44:48,791 --> 01:44:50,375
I definitely use it more than the...

1526
01:44:50,583 --> 01:44:50,875
Google?

1527
01:44:51,250 --> 01:44:52,541
And then OpenAI's...

1528
01:44:52,708 --> 01:44:53,833
I'll search GPT.

1529
01:44:53,833 --> 01:44:54,333
GPT?

1530
01:44:54,333 --> 01:44:54,791
That's interesting.

1531
01:44:55,708 --> 01:44:57,208
I just like the way that it structures a

1532
01:44:57,208 --> 01:44:57,958
lot of content, although

1533
01:44:57,958 --> 01:44:58,666
they're starting to feed

1534
01:44:58,666 --> 01:44:59,375
in a few things that

1535
01:44:59,375 --> 01:45:00,083
I'm like a little bit...

1536
01:45:00,666 --> 01:45:01,708
Especially with like very common...

1537
01:45:01,708 --> 01:45:02,750
Not safe for work stuff.

1538
01:45:03,041 --> 01:45:03,333
Well, not...

1539
01:45:03,416 --> 01:45:04,625
I mean, yeah, that's just natural.

1540
01:45:04,916 --> 01:45:06,000
That's what I'm searching for.

1541
01:45:06,000 --> 01:45:06,125
No, but...

1542
01:45:06,125 --> 01:45:06,708
In a day perspective.

1543
01:45:07,041 --> 01:45:09,958
This is good though, because we found...

1544
01:45:10,416 --> 01:45:12,041
And this is in terms of like building.

1545
01:45:12,250 --> 01:45:14,875
We were using an API that was similar to

1546
01:45:14,875 --> 01:45:17,625
providing news and search.

1547
01:45:18,625 --> 01:45:19,625
Sorry, more search

1548
01:45:19,625 --> 01:45:21,125
like Perplexity, right?

1549
01:45:21,416 --> 01:45:22,833
We've been testing that out and it's been

1550
01:45:22,833 --> 01:45:24,041
good, but we were

1551
01:45:24,041 --> 01:45:25,291
waiting on the day for...

1552
01:45:26,791 --> 01:45:27,500
Just to be able to

1553
01:45:27,500 --> 01:45:28,500
get it from Perplexity.

1554
01:45:28,500 --> 01:45:30,083
So this sounds like it's the day.

1555
01:45:30,083 --> 01:45:30,500
That's it.

1556
01:45:30,500 --> 01:45:31,875
This is the day for that,

1557
01:45:31,875 --> 01:45:32,750
which would be really nice.

1558
01:45:32,750 --> 01:45:34,000
So that's out there in the world.

1559
01:45:35,458 --> 01:45:38,958
Anthropic had an interview they did...

1560
01:45:38,958 --> 01:45:39,208
Dario.

1561
01:45:39,583 --> 01:45:41,041
Yeah, Mr. Dario.

1562
01:45:41,375 --> 01:45:43,375
Mr. Of Machines and Love and Grace.

1563
01:45:43,375 --> 01:45:43,916
What a title.

1564
01:45:44,500 --> 01:45:45,125
So he had a...

1565
01:45:45,125 --> 01:45:47,208
It was with Wall Street Journal and kind

1566
01:45:47,208 --> 01:45:47,916
of the interesting thing

1567
01:45:47,916 --> 01:45:49,333
here was just a conversation.

1568
01:45:49,791 --> 01:45:52,625
One is like the pending AGI discussion.

1569
01:45:52,958 --> 01:45:54,375
He's probably got one of the more

1570
01:45:54,375 --> 01:45:55,291
realistic timelines.

1571
01:45:55,291 --> 01:45:58,416
You talk to OpenAI and...

1572
01:45:59,000 --> 01:45:59,583
It's next year.

1573
01:45:59,958 --> 01:46:00,916
I mean, next year.

1574
01:46:00,916 --> 01:46:01,208
It's now.

1575
01:46:01,208 --> 01:46:02,166
It's like six months away.

1576
01:46:02,166 --> 01:46:03,083
We're almost there.

1577
01:46:04,083 --> 01:46:06,083
I told you you wait for it.

1578
01:46:06,333 --> 01:46:07,666
He's like, yeah, 2027 is

1579
01:46:07,666 --> 01:46:08,791
probably the most likely year.

1580
01:46:08,791 --> 01:46:09,250
So we're about two

1581
01:46:09,250 --> 01:46:10,708
years away from proper AGI.

1582
01:46:10,708 --> 01:46:12,000
And then following that, obviously, the

1583
01:46:12,000 --> 01:46:14,583
super intelligence not far beyond.

1584
01:46:14,916 --> 01:46:16,166
But it is, as we said,

1585
01:46:16,166 --> 01:46:17,458
it's an arms race for that.

1586
01:46:18,166 --> 01:46:18,916
But most interesting...

1587
01:46:18,916 --> 01:46:20,583
I would love to see this arms...

1588
01:46:21,000 --> 01:46:22,416
Just arms racing.

1589
01:46:22,416 --> 01:46:22,875
Just arms racing.

1590
01:46:22,875 --> 01:46:23,708
It's like, yeah.

1591
01:46:23,708 --> 01:46:26,125
Mate, just go to the generator.

1592
01:46:26,500 --> 01:46:26,625
We will.

1593
01:46:26,625 --> 01:46:27,000
Yeah, it'll happen.

1594
01:46:27,000 --> 01:46:27,500
All right, let's go.

1595
01:46:28,750 --> 01:46:30,000
But one of the things he did highlight

1596
01:46:30,000 --> 01:46:31,541
is, which I thought was interesting.

1597
01:46:31,583 --> 01:46:32,333
I don't know whether it's

1598
01:46:32,333 --> 01:46:33,916
they need to be doing this.

1599
01:46:34,166 --> 01:46:37,083
I think Anthropic has a really solid hold

1600
01:46:37,083 --> 01:46:39,083
on the math and code

1601
01:46:39,083 --> 01:46:41,000
generating aspects of

1602
01:46:41,000 --> 01:46:43,125
their tool, but they are now doing voice

1603
01:46:43,125 --> 01:46:45,333
planning to roll that out

1604
01:46:45,333 --> 01:46:46,458
in the coming month as well,

1605
01:46:46,458 --> 01:46:48,000
which I thought was interesting.

1606
01:46:48,791 --> 01:46:49,375
It's very interesting.

1607
01:46:49,791 --> 01:46:52,166
They haven't got their language model for

1608
01:46:52,166 --> 01:46:53,416
voice training models

1609
01:46:53,416 --> 01:46:54,041
to have conversations

1610
01:46:54,250 --> 01:46:55,083
with their AI tool.

1611
01:46:56,125 --> 01:46:57,750
So that could probably change how you

1612
01:46:57,750 --> 01:46:58,375
interact with those.

1613
01:46:58,375 --> 01:47:00,625
So rather than having to type your

1614
01:47:00,625 --> 01:47:02,166
queries out while you're

1615
01:47:02,166 --> 01:47:04,000
generating code or whatever

1616
01:47:04,000 --> 01:47:06,000
it is that you might be doing with

1617
01:47:06,000 --> 01:47:08,375
Claude, you can have a

1618
01:47:08,375 --> 01:47:09,250
conversation with it to do

1619
01:47:09,250 --> 01:47:09,750
it instead.

1620
01:47:09,750 --> 01:47:10,458
So just adjust this,

1621
01:47:10,458 --> 01:47:11,375
or can you tweak that?

1622
01:47:11,375 --> 01:47:13,000
And hopefully it will do it live on the

1623
01:47:13,000 --> 01:47:14,625
canvas while you're

1624
01:47:14,625 --> 01:47:15,166
having that conversation.

1625
01:47:15,416 --> 01:47:18,333
I mean, speaking of that topic there with

1626
01:47:18,333 --> 01:47:20,916
Anthropic, Claude, the model that they've

1627
01:47:20,916 --> 01:47:24,083
got, 3.5 Sonnet especially, is still

1628
01:47:24,083 --> 01:47:26,541
considered one of the best coding tools.

1629
01:47:26,875 --> 01:47:28,416
And I've seen it when I use Bolt.

1630
01:47:28,916 --> 01:47:30,625
So shout out to the guys at Bolt as well.

1631
01:47:31,083 --> 01:47:32,791
By default, they use...

1632
01:47:32,791 --> 01:47:33,625
Bolt by default.

1633
01:47:33,625 --> 01:47:35,291
Bolt by default use 3.5

1634
01:47:35,291 --> 01:47:37,333
Sonnet for doing the coding.

1635
01:47:37,791 --> 01:47:40,875
And when you use Bolt DIY, which is a

1636
01:47:40,875 --> 01:47:42,125
fork of their open

1637
01:47:42,125 --> 01:47:43,041
source version where you can

1638
01:47:43,041 --> 01:47:44,833
choose which model you want to use to

1639
01:47:44,833 --> 01:47:46,416
have code completion, or

1640
01:47:46,416 --> 01:47:47,833
even when I use CoPilot,

1641
01:47:48,041 --> 01:47:50,750
GitHub CoPilot type stuff with Microsoft

1642
01:47:50,750 --> 01:47:54,041
Visual Studio Code, you

1643
01:47:54,041 --> 01:47:56,500
basically see that GPT40

1644
01:47:56,791 --> 01:47:58,291
versus Claude.

1645
01:47:59,833 --> 01:48:02,208
It's Claude is hands down a

1646
01:48:02,208 --> 01:48:05,500
ton better in a lot of tasks.

1647
01:48:05,666 --> 01:48:06,875
There's certain things where it's like,

1648
01:48:06,875 --> 01:48:08,500
okay, it's neck and neck, but man.

1649
01:48:08,500 --> 01:48:09,500
Particularly when it comes to anything

1650
01:48:09,500 --> 01:48:10,500
that's Python generating

1651
01:48:10,500 --> 01:48:12,166
or even front end code,

1652
01:48:12,166 --> 01:48:13,958
like it just, it's miles ahead.

1653
01:48:13,958 --> 01:48:14,166
Yeah.

1654
01:48:14,166 --> 01:48:15,333
There are some other tools that do, I

1655
01:48:15,333 --> 01:48:15,916
mean, particularly when

1656
01:48:15,916 --> 01:48:16,750
it comes to visualization,

1657
01:48:17,291 --> 01:48:19,041
does that a lot better from turning that

1658
01:48:19,041 --> 01:48:20,000
into some kind of HTML

1659
01:48:20,000 --> 01:48:21,750
or CSS, but nine times out

1660
01:48:21,750 --> 01:48:23,208
of 10, you just pump that back in a Bolt

1661
01:48:23,208 --> 01:48:24,291
and then, or into your

1662
01:48:24,291 --> 01:48:25,416
GitHub repository with

1663
01:48:25,458 --> 01:48:26,791
Bolt attached to the back of it.

1664
01:48:27,625 --> 01:48:29,708
It's just, it's the best.

1665
01:48:29,708 --> 01:48:30,541
Here's one for you.

1666
01:48:30,541 --> 01:48:31,500
VO people are still...

1667
01:48:31,791 --> 01:48:32,625
Oh yeah, V0.

1668
01:48:32,875 --> 01:48:34,500
V0 is still up there,

1669
01:48:34,500 --> 01:48:35,375
is pretty good as well.

1670
01:48:35,375 --> 01:48:36,583
A lot of people are using that over Bolt.

1671
01:48:36,791 --> 01:48:38,125
It's a mixed match, but Bolt

1672
01:48:38,125 --> 01:48:39,500
tends to be, I think, shining.

1673
01:48:39,750 --> 01:48:41,958
Of course, particularly if you engage

1674
01:48:41,958 --> 01:48:42,583
with anyone across

1675
01:48:42,583 --> 01:48:44,291
Reddit, Bolt tends to be the

1676
01:48:44,291 --> 01:48:44,708
one that...

1677
01:48:44,708 --> 01:48:45,458
I don't have the...

1678
01:48:46,208 --> 01:48:46,708
Yeah, exactly.

1679
01:48:46,958 --> 01:48:48,833
I don't have the example here, but it's

1680
01:48:48,833 --> 01:48:49,958
integrations, I think are

1681
01:48:49,958 --> 01:48:53,000
going to be the key, where Bolt

1682
01:48:53,000 --> 01:48:54,583
has an integration directly into

1683
01:48:54,583 --> 01:48:56,791
Superbase, which is an online database.

1684
01:48:57,000 --> 01:48:57,666
If you're used to

1685
01:48:57,666 --> 01:48:59,500
Postgres SQL, that's one.

1686
01:48:59,875 --> 01:49:00,750
Google has something called

1687
01:49:00,750 --> 01:49:02,625
Firebase, which is for NoSQL.

1688
01:49:02,958 --> 01:49:03,833
I'm getting into the

1689
01:49:03,833 --> 01:49:05,791
tech nerdy dev stuff here.

1690
01:49:06,916 --> 01:49:08,250
But depending on the type of thing that

1691
01:49:08,250 --> 01:49:09,500
you're trying to do, one

1692
01:49:09,500 --> 01:49:10,500
is better than the other

1693
01:49:11,083 --> 01:49:11,958
in terms of your projects.

1694
01:49:11,958 --> 01:49:14,125
But my point is that with Superbase, Bolt

1695
01:49:14,125 --> 01:49:14,833
had built in an

1696
01:49:14,833 --> 01:49:16,458
integration, and sure, you

1697
01:49:16,458 --> 01:49:18,541
have to sign into Superbase at least once

1698
01:49:18,541 --> 01:49:19,250
to be able to do it,

1699
01:49:19,250 --> 01:49:20,500
but you don't have to

1700
01:49:20,500 --> 01:49:22,625
go back and forth like you normally would

1701
01:49:22,625 --> 01:49:24,000
with building stuff.

1702
01:49:24,333 --> 01:49:26,000
They just do it for you in

1703
01:49:26,000 --> 01:49:28,000
Bolt, so it's making it easier.

1704
01:49:28,000 --> 01:49:30,000
There's still some stuff to improve, but

1705
01:49:30,000 --> 01:49:31,375
every week they're doing

1706
01:49:31,375 --> 01:49:32,333
these things and building

1707
01:49:32,333 --> 01:49:32,958
an open source.

1708
01:49:33,416 --> 01:49:35,125
Speaking of the arms race before, you've

1709
01:49:35,125 --> 01:49:35,875
got the prompt cowboy

1710
01:49:35,875 --> 01:49:38,750
up here, but let's just

1711
01:49:38,750 --> 01:49:39,583
jump to this.

1712
01:49:40,625 --> 01:49:41,958
I don't know what the heck is going on

1713
01:49:41,958 --> 01:49:43,000
here, but there's a

1714
01:49:43,000 --> 01:49:44,166
weird arms race, Chris.

1715
01:49:44,541 --> 01:49:45,750
Can you tell what's going on here?

1716
01:49:47,541 --> 01:49:49,583
That is creepy as hell.

1717
01:49:50,041 --> 01:49:50,666
What about this?

1718
01:49:50,666 --> 01:49:51,958
Is this a little bit better?

1719
01:49:52,250 --> 01:49:54,000
I mean, this guy has like two thumbs.

1720
01:49:54,666 --> 01:49:56,000
What's wrong with the guy?

1721
01:49:56,000 --> 01:49:57,291
Yeah, the guy in the background is like,

1722
01:49:57,291 --> 01:49:58,958
he's missing half of his thumb.

1723
01:49:59,500 --> 01:50:00,625
Who knows, man?

1724
01:50:00,625 --> 01:50:01,208
It's like, I don't...

1725
01:50:01,208 --> 01:50:01,500
Pretty funny.

1726
01:50:01,750 --> 01:50:02,250
This one's good.

1727
01:50:02,625 --> 01:50:04,500
So, you know, the AI was forced to

1728
01:50:04,500 --> 01:50:06,083
consider in an arms race.

1729
01:50:06,583 --> 01:50:07,916
Is it an arm with a

1730
01:50:07,916 --> 01:50:09,333
shorter arm as its legs?

1731
01:50:09,916 --> 01:50:10,208
Right?

1732
01:50:10,625 --> 01:50:11,916
It's like a dinosaur.

1733
01:50:11,916 --> 01:50:13,791
It's like a diplodocus or a brontosaurus.

1734
01:50:14,541 --> 01:50:14,750
Right?

1735
01:50:15,416 --> 01:50:16,833
Versus this.

1736
01:50:16,833 --> 01:50:17,458
I don't know what this is.

1737
01:50:18,041 --> 01:50:18,916
It's like a finger race

1738
01:50:18,916 --> 01:50:19,666
more than an arms race.

1739
01:50:20,000 --> 01:50:20,208
Yeah.

1740
01:50:20,333 --> 01:50:22,750
This is like the nightmare.

1741
01:50:22,958 --> 01:50:24,875
So what's showing on screen here, folks,

1742
01:50:25,125 --> 01:50:27,458
is like one is just basically all arm.

1743
01:50:28,625 --> 01:50:30,250
The other one is basically just hands

1744
01:50:30,250 --> 01:50:31,750
like wiping a table as

1745
01:50:31,750 --> 01:50:33,500
if the nightmare fuel of

1746
01:50:33,500 --> 01:50:35,291
someone who just cleaned a table.

1747
01:50:35,625 --> 01:50:36,791
We'll post this on TikTok separately.

1748
01:50:37,083 --> 01:50:37,750
Yeah, you're going to love it.

1749
01:50:37,916 --> 01:50:38,458
You're going to love it.

1750
01:50:38,958 --> 01:50:39,708
Thank you, Sora.

1751
01:50:39,958 --> 01:50:40,458
Thank you, Sora.

1752
01:50:40,666 --> 01:50:41,041
Love, Sora.

1753
01:50:41,291 --> 01:50:42,333
Let's get over to lazy prompts.

1754
01:50:42,958 --> 01:50:43,333
Lazy prompts.

1755
01:50:43,333 --> 01:50:44,583
So this is just a quick thing.

1756
01:50:44,583 --> 01:50:45,166
I thought it'd be interesting.

1757
01:50:45,166 --> 01:50:45,541
I just wanted to

1758
01:50:45,541 --> 01:50:46,291
share this with everyone.

1759
01:50:47,500 --> 01:50:49,333
You hear a lot about prompt engineering.

1760
01:50:49,625 --> 01:50:51,958
There's kids created a whole job sector.

1761
01:50:52,375 --> 01:50:53,708
I don't want to call it industry, but I

1762
01:50:53,708 --> 01:50:54,708
like a sector of people who

1763
01:50:54,708 --> 01:50:56,000
are like prompt engineers.

1764
01:50:56,000 --> 01:50:56,708
Some call them pros.

1765
01:50:57,416 --> 01:50:58,416
I think it's all a bit of a wishwash

1766
01:50:58,416 --> 01:50:59,083
thing, but prompt

1767
01:50:59,083 --> 01:51:00,541
engineering in itself is still

1768
01:51:00,541 --> 01:51:01,500
an important thing.

1769
01:51:01,500 --> 01:51:02,583
And we talk about it a lot because

1770
01:51:02,583 --> 01:51:04,500
absolutely your output

1771
01:51:04,500 --> 01:51:06,458
from the AI is only as good as

1772
01:51:06,458 --> 01:51:07,000
your input.

1773
01:51:07,833 --> 01:51:09,125
And a really good example is really truly

1774
01:51:09,125 --> 01:51:10,458
what you mentioned before, right?

1775
01:51:10,458 --> 01:51:13,166
Even just saying, hey, give me a $1,000

1776
01:51:13,166 --> 01:51:14,291
response versus give

1777
01:51:14,291 --> 01:51:16,083
me a $100,000 response.

1778
01:51:16,333 --> 01:51:19,250
Like that enticing and that pulling out

1779
01:51:19,250 --> 01:51:21,000
of information is

1780
01:51:21,000 --> 01:51:23,666
going to be as good as how

1781
01:51:23,666 --> 01:51:27,125
much information or how much enticing, I

1782
01:51:27,125 --> 01:51:29,250
guess, you give the AI.

1783
01:51:30,083 --> 01:51:32,083
And so what this tool tends to do is to

1784
01:51:32,083 --> 01:51:33,625
try and remove that

1785
01:51:33,625 --> 01:51:34,750
boundary of trying to think

1786
01:51:34,750 --> 01:51:36,500
about how you structure responses by

1787
01:51:36,500 --> 01:51:38,333
actually doing it for you.

1788
01:51:38,666 --> 01:51:40,041
So let's say I want to do

1789
01:51:40,041 --> 01:51:41,041
something within Bolt, right?

1790
01:51:41,041 --> 01:51:49,583
I want to create an application in Bolt

1791
01:51:49,583 --> 01:51:53,250
that is a chatbot

1792
01:51:53,250 --> 01:52:00,791
utilizing open AI's API to engage

1793
01:52:01,000 --> 01:52:01,916
with customers.

1794
01:52:03,166 --> 01:52:03,416
Do it.

1795
01:52:03,666 --> 01:52:04,208
So this is like I just

1796
01:52:04,208 --> 01:52:05,083
made this up on the spot.

1797
01:52:05,958 --> 01:52:07,000
I was misspelled there.

1798
01:52:08,083 --> 01:52:10,541
And so what it pops up now is it gives a

1799
01:52:10,541 --> 01:52:11,333
bit of a think about

1800
01:52:11,333 --> 01:52:13,500
it and it structures the

1801
01:52:13,500 --> 01:52:15,708
content around how you should ask the AI

1802
01:52:15,708 --> 01:52:17,125
to the LLM to get a better

1803
01:52:17,125 --> 01:52:20,125
response by both suggesting

1804
01:52:20,125 --> 01:52:22,250
how what the question is, what the

1805
01:52:22,250 --> 01:52:23,208
situation is, which is

1806
01:52:23,208 --> 01:52:24,958
really important, the tasks you

1807
01:52:24,958 --> 01:52:26,833
want it to complete, setting objectives

1808
01:52:26,833 --> 01:52:28,541
and outcomes and the important

1809
01:52:28,541 --> 01:52:29,541
information requirements.

1810
01:52:29,541 --> 01:52:31,208
And even in some instances

1811
01:52:31,208 --> 01:52:33,166
where relevant providing examples.

1812
01:52:33,375 --> 01:52:34,625
And that's one of the best ways to get

1813
01:52:34,625 --> 01:52:35,500
outputs is making sure

1814
01:52:35,500 --> 01:52:36,541
you're sharing examples to

1815
01:52:36,541 --> 01:52:37,291
the AI as well.

1816
01:52:37,583 --> 01:52:39,458
So kind of, I want to see this or here's

1817
01:52:39,458 --> 01:52:40,291
an example of someone

1818
01:52:40,291 --> 01:52:41,000
who's something has been

1819
01:52:41,000 --> 01:52:42,875
done before and therefore it structures

1820
01:52:42,875 --> 01:52:43,791
it in a similar way to

1821
01:52:43,791 --> 01:52:44,625
get that outcome you want.

1822
01:52:44,875 --> 01:52:45,916
You know, it's interesting because it

1823
01:52:45,916 --> 01:52:47,041
gives you all those instructions.

1824
01:52:47,041 --> 01:52:48,833
But if you were to try that just knowing

1825
01:52:48,833 --> 01:52:49,708
how these tools like

1826
01:52:49,708 --> 01:52:51,416
bot work, you've got a lot

1827
01:52:51,416 --> 01:52:52,291
of steps here.

1828
01:52:52,708 --> 01:52:55,791
It will definitely miss on some of those.

1829
01:52:56,083 --> 01:52:57,333
So you could get this thing

1830
01:52:57,333 --> 01:52:58,625
to, I guess, break it down.

1831
01:52:58,625 --> 01:53:00,083
But even if you did break it down into

1832
01:53:00,083 --> 01:53:01,500
parts, trying those

1833
01:53:01,500 --> 01:53:02,916
first few steps, then errors

1834
01:53:02,958 --> 01:53:03,458
come up.

1835
01:53:03,750 --> 01:53:05,916
So, you know, you have to get it like, so

1836
01:53:05,916 --> 01:53:07,166
there's a bit of massaging.

1837
01:53:07,375 --> 01:53:08,708
But this is good at

1838
01:53:08,708 --> 01:53:10,458
least to get things started.

1839
01:53:10,458 --> 01:53:12,041
The great thing about this is that like,

1840
01:53:12,041 --> 01:53:12,666
let's say you read the

1841
01:53:12,666 --> 01:53:13,458
response and you're kind

1842
01:53:13,458 --> 01:53:14,500
of like, I'm not quite

1843
01:53:14,500 --> 01:53:15,666
happy with what that's saying.

1844
01:53:15,916 --> 01:53:18,000
You can go through the phases of actually

1845
01:53:18,000 --> 01:53:18,875
adjusting it accordingly.

1846
01:53:19,125 --> 01:53:20,875
So you define the purpose.

1847
01:53:20,875 --> 01:53:21,750
You can have examples

1848
01:53:21,750 --> 01:53:23,000
that I might feed into it.

1849
01:53:23,000 --> 01:53:24,208
So a restaurant booking assistant, let's

1850
01:53:24,208 --> 01:53:25,416
just say conversation

1851
01:53:25,416 --> 01:53:27,500
examples like I need or one.

1852
01:53:27,500 --> 01:53:28,208
Here's an example of the

1853
01:53:28,208 --> 01:53:29,375
questions I wanted to ask.

1854
01:53:29,750 --> 01:53:31,583
And even response style, like how do you

1855
01:53:31,583 --> 01:53:32,416
want it to be casual?

1856
01:53:32,416 --> 01:53:33,291
Do you want to be professional?

1857
01:53:33,291 --> 01:53:34,416
Do you want to be brief and direct?

1858
01:53:36,041 --> 01:53:38,083
And so you can tailor that or you can,

1859
01:53:38,083 --> 01:53:39,416
you know, keep editing

1860
01:53:39,416 --> 01:53:41,708
this response by editing

1861
01:53:41,708 --> 01:53:42,500
the prompt and then

1862
01:53:42,500 --> 01:53:44,000
readjusting it and glooping it.

1863
01:53:44,000 --> 01:53:46,208
So you're constantly trying to tweak that

1864
01:53:46,208 --> 01:53:47,333
kind of output rather

1865
01:53:47,333 --> 01:53:48,250
than doing it within

1866
01:53:48,250 --> 01:53:48,916
the AI tool.

1867
01:53:48,916 --> 01:53:50,166
Because once you start adjusting your

1868
01:53:50,166 --> 01:53:51,208
prompts within the AI

1869
01:53:51,208 --> 01:53:54,208
tool, those outcomes also start

1870
01:53:54,208 --> 01:53:56,750
to adjust and you can very easily lose

1871
01:53:56,750 --> 01:53:58,666
sight of what the AI

1872
01:53:58,666 --> 01:54:00,208
can lose sight of what you

1873
01:54:00,208 --> 01:54:01,583
were trying to achieve as you're

1874
01:54:01,583 --> 01:54:02,583
adjusting your response

1875
01:54:02,583 --> 01:54:03,708
every single time rather than

1876
01:54:03,833 --> 01:54:05,291
building on the response because you

1877
01:54:05,291 --> 01:54:06,416
didn't quite get to where you were.

1878
01:54:06,416 --> 01:54:06,875
Oh, exactly.

1879
01:54:06,875 --> 01:54:08,291
So this is kind of like getting you that

1880
01:54:08,291 --> 01:54:10,958
70% of the way to get better outcomes.

1881
01:54:11,166 --> 01:54:11,791
Worth it, worth it.

1882
01:54:11,791 --> 01:54:12,583
We're checking it out.

1883
01:54:12,791 --> 01:54:13,458
We'll have that in the

1884
01:54:13,458 --> 01:54:14,208
show notes and stuff.

1885
01:54:14,583 --> 01:54:16,916
And a shout out to the guys who made it.

1886
01:54:17,333 --> 01:54:17,541
Four day.

1887
01:54:18,125 --> 01:54:18,750
Yeah, four day.

1888
01:54:19,125 --> 01:54:21,375
So one of the founders, Henry Badry,

1889
01:54:21,791 --> 01:54:23,041
Trachymard, LinkedIn and X.

1890
01:54:24,625 --> 01:54:25,916
He's a very smart guy and doing a lot of

1891
01:54:25,916 --> 01:54:26,708
consulting in that space.

1892
01:54:27,000 --> 01:54:27,166
Fantastic.

1893
01:54:27,375 --> 01:54:29,000
We'll get those in and make sure you put

1894
01:54:29,000 --> 01:54:30,416
that in the show notes and stuff there.

1895
01:54:30,625 --> 01:54:30,833
Definitely.

1896
01:54:32,041 --> 01:54:35,000
So I'd like to showcase, and I've never

1897
01:54:35,000 --> 01:54:36,250
shown you guys before,

1898
01:54:36,250 --> 01:54:36,958
but this is just like for

1899
01:54:36,958 --> 01:54:38,458
a little bit of inspiration and stuff,

1900
01:54:39,333 --> 01:54:40,708
what you can do with AI

1901
01:54:40,708 --> 01:54:42,166
tools, you can build things

1902
01:54:42,166 --> 01:54:43,500
like this is ZTF tracker.

1903
01:54:43,916 --> 01:54:48,500
I've now got 477 members signed up that

1904
01:54:48,500 --> 01:54:49,208
are profiling

1905
01:54:49,208 --> 01:54:50,666
themselves that are doing things

1906
01:54:50,666 --> 01:54:53,416
like saying, like, here are the ETFs that

1907
01:54:53,416 --> 01:54:54,500
I like to track and

1908
01:54:54,500 --> 01:54:55,541
stuff and having people

1909
01:54:55,708 --> 01:54:58,625
just be able to just very quickly get

1910
01:54:58,625 --> 01:54:59,875
like some quick insights

1911
01:54:59,875 --> 01:55:02,125
on those ETFs from data

1912
01:55:02,125 --> 01:55:03,333
that's coming in publicly.

1913
01:55:04,000 --> 01:55:05,666
But it's taking time

1914
01:55:05,666 --> 01:55:06,916
to piece it together.

1915
01:55:07,083 --> 01:55:10,250
Well, we've got AI tools that have helped

1916
01:55:10,250 --> 01:55:11,583
us build something

1917
01:55:11,583 --> 01:55:12,583
like this out that used

1918
01:55:12,583 --> 01:55:13,833
to take me a bit more manual

1919
01:55:13,833 --> 01:55:16,000
work to do and then automated.

1920
01:55:16,416 --> 01:55:18,416
So there was still a lot of value with

1921
01:55:18,416 --> 01:55:19,375
the stuff I did before.

1922
01:55:20,000 --> 01:55:22,458
But given that we've got these AI tools

1923
01:55:22,458 --> 01:55:23,916
like Bolt and Psyche

1924
01:55:23,916 --> 01:55:24,666
and all these other things

1925
01:55:24,750 --> 01:55:25,958
out there that have helped piece this

1926
01:55:25,958 --> 01:55:27,333
together, it just means

1927
01:55:27,333 --> 01:55:29,166
that you can actually deliver

1928
01:55:29,166 --> 01:55:30,291
a whole lot more value,

1929
01:55:30,666 --> 01:55:32,250
take less time to do it.

1930
01:55:32,916 --> 01:55:34,416
And now that you've got all this spare

1931
01:55:34,416 --> 01:55:36,416
time, what do you do to add

1932
01:55:36,416 --> 01:55:37,625
on further to your product?

1933
01:55:37,625 --> 01:55:39,416
So this is more of like a call to action.

1934
01:55:39,416 --> 01:55:42,333
I mean, I launched this 24th of December,

1935
01:55:42,791 --> 01:55:44,916
Christmas Eve, and nearly 500 people have

1936
01:55:44,916 --> 01:55:45,500
signed up.

1937
01:55:45,750 --> 01:55:45,958
Excellent.

1938
01:55:46,375 --> 01:55:48,000
It's freaking crazy, right?

1939
01:55:48,000 --> 01:55:48,916
I wish I monetized it.

1940
01:55:49,708 --> 01:55:50,708
You can buy me a

1941
01:55:50,708 --> 01:55:51,958
coffee and stuff for that.

1942
01:55:52,041 --> 01:55:53,083
And it is public information.

1943
01:55:53,083 --> 01:55:54,458
I'm not planning on like doing anything

1944
01:55:54,458 --> 01:55:55,666
kind of crazy with it here.

1945
01:55:56,208 --> 01:55:57,625
But, you know, there are the

1946
01:55:57,625 --> 01:55:59,208
possibilities to do that.

1947
01:55:59,208 --> 01:56:01,041
Like if I see that there's enough kind of

1948
01:56:01,041 --> 01:56:01,791
members coming through

1949
01:56:01,791 --> 01:56:02,500
or they're sponsorship,

1950
01:56:02,791 --> 01:56:04,500
then maybe we get some dedicated feeds

1951
01:56:04,500 --> 01:56:06,458
rather than scraping and stuff like that.

1952
01:56:06,875 --> 01:56:08,291
But I think there's some really cool

1953
01:56:08,291 --> 01:56:09,375
things that you can do

1954
01:56:09,375 --> 01:56:11,208
with whether it's finance

1955
01:56:11,291 --> 01:56:13,125
apps, it's stuff that tracks housing

1956
01:56:13,125 --> 01:56:14,416
prices, it's stuff that

1957
01:56:14,416 --> 01:56:15,958
makes it easier to consume

1958
01:56:15,958 --> 01:56:18,791
government type statistics from the ABS.

1959
01:56:18,916 --> 01:56:22,000
There's a plethora of different cool

1960
01:56:22,000 --> 01:56:23,625
ideas that people just go,

1961
01:56:23,625 --> 01:56:24,583
I wish there was something

1962
01:56:24,583 --> 01:56:25,041
for this.

1963
01:56:25,666 --> 01:56:27,833
Yeah, you wish make it happen.

1964
01:56:28,041 --> 01:56:29,458
You can do this kind of stuff.

1965
01:56:29,458 --> 01:56:31,208
And if you are getting stuck, I mean,

1966
01:56:31,208 --> 01:56:32,250
reach out to Chris and

1967
01:56:32,250 --> 01:56:33,500
myself, like we can definitely

1968
01:56:33,500 --> 01:56:34,458
help you out as well.

1969
01:56:35,333 --> 01:56:36,750
I'm seeing more and more people.

1970
01:56:37,916 --> 01:56:40,500
I made this with bolts using and it

1971
01:56:40,500 --> 01:56:41,333
wasn't from scratch.

1972
01:56:41,750 --> 01:56:42,250
Like this was

1973
01:56:42,250 --> 01:56:43,750
previous tool that was built.

1974
01:56:43,750 --> 01:56:46,208
So I knew exactly what I needed to build.

1975
01:56:47,375 --> 01:56:48,375
You can do things like,

1976
01:56:48,375 --> 01:56:49,375
hey, bolt, build me this.

1977
01:56:49,375 --> 01:56:50,833
But it's like, well, it doesn't have the

1978
01:56:50,833 --> 01:56:51,916
connections to the data.

1979
01:56:51,916 --> 01:56:54,166
I had to build that wall of, you know, I

1980
01:56:54,166 --> 01:56:55,333
had to offline,

1981
01:56:55,333 --> 01:56:56,375
there's a whole lot of Python

1982
01:56:56,416 --> 01:56:58,250
scraping and code and stuff like that.

1983
01:56:58,666 --> 01:57:01,000
But in terms of piecing it together and

1984
01:57:01,000 --> 01:57:03,125
making it mobile friendly is the key.

1985
01:57:03,375 --> 01:57:05,958
This kind of thing you can easily do now.

1986
01:57:06,166 --> 01:57:08,208
So I think it's just a nice to have that

1987
01:57:08,208 --> 01:57:09,125
call to action to folks

1988
01:57:09,125 --> 01:57:10,166
to just get out there and

1989
01:57:10,166 --> 01:57:12,083
build, you know, and if you get stuck,

1990
01:57:12,083 --> 01:57:13,000
hit us up because we want

1991
01:57:13,000 --> 01:57:13,833
to have these conversations.

1992
01:57:14,083 --> 01:57:16,208
The more that is happening in this world

1993
01:57:16,208 --> 01:57:18,541
like that, it's going to be easier for us

1994
01:57:18,541 --> 01:57:20,250
to get solutions into businesses.

1995
01:57:20,708 --> 01:57:21,541
So like we just need to

1996
01:57:21,541 --> 01:57:22,875
have more people doing it.

1997
01:57:23,250 --> 01:57:23,458
100%.

1998
01:57:23,791 --> 01:57:24,416
Well said.

1999
01:57:25,500 --> 01:57:26,708
Any final thoughts from you, Chris?

2000
01:57:26,958 --> 01:57:27,708
No, I think that's it.

2001
01:57:27,833 --> 01:57:30,125
That's been a week both from a political,

2002
01:57:30,625 --> 01:57:32,208
from a product release,

2003
01:57:32,208 --> 01:57:33,208
from a what's coming in

2004
01:57:33,208 --> 01:57:34,375
the future point of view.

2005
01:57:34,666 --> 01:57:36,250
Political product future.

2006
01:57:36,708 --> 01:57:37,083
No, it doesn't.

2007
01:57:37,416 --> 01:57:37,625
Future.

2008
01:57:37,875 --> 01:57:38,333
From the future.

2009
01:57:38,708 --> 01:57:38,958
Yeah.

2010
01:57:38,958 --> 01:57:39,666
The future with a pH.

2011
01:57:40,166 --> 01:57:40,375
Yeah.

2012
01:57:40,708 --> 01:57:40,916
Future.

2013
01:57:43,166 --> 01:57:43,833
Yeah.

2014
01:57:44,166 --> 01:57:45,416
Pretty excited to see what's going on.

2015
01:57:45,416 --> 01:57:46,125
Good to be back.

2016
01:57:46,500 --> 01:57:47,208
Good to see your face.

2017
01:57:47,958 --> 01:57:48,208
Is it?

2018
01:57:48,875 --> 01:57:50,666
Yeah, it does work and that's yeah.

2019
01:57:50,666 --> 01:57:51,333
Till next time, guys.

2020
01:57:51,750 --> 01:57:52,500
We'll see you on Monday,

2021
01:57:52,500 --> 01:57:55,625
folks, for the podcast.

2022
01:57:56,000 --> 01:57:58,583
Well, it's more the Twitter spaces that

2023
01:57:58,583 --> 01:58:00,625
we're going to do for the Oz Defy stuff.

2024
01:58:01,333 --> 01:58:03,708
We will be talking about AI there because

2025
01:58:03,708 --> 01:58:04,708
there's ways that

2026
01:58:04,708 --> 01:58:06,541
blockchain is working to

2027
01:58:06,708 --> 01:58:07,333
help AI.

2028
01:58:07,333 --> 01:58:08,708
I had it in the last episode and stuff.

2029
01:58:09,125 --> 01:58:10,166
So we'll be diving into

2030
01:58:10,166 --> 01:58:11,041
that a little bit more.

2031
01:58:11,041 --> 01:58:12,166
So check that one out.

2032
01:58:12,291 --> 01:58:12,875
See you online.

2033
01:58:13,125 --> 01:58:13,875
See you next week.

2034
01:58:14,208 --> 01:58:15,000
And thanks very much.

2035
01:58:15,333 --> 01:58:15,750
AI.