10/6/2025 - Exploring OpenAI's New App Integrations and The Rise of The Everything App

Transcript

1. 0:07[Music] Yay! [Music]
2. 0:28[Music] It's not simp. [Music]
3. 0:50It's not easy to know I'm not anything like I used to be. Although it's true. I was never attention sweet Santa. I still remember that girl.
4. 1:06She's imperfect but she tries. She is good but she lies. She is hard on herself. She is broken and won't ask for help.
5. 1:23She is messy. She's kind. She's lonely. Most of the time she's all this mixed up. Beautiful.
6. 1:38She's gone, but she used to be mine. [Music]
7. 1:52[Music] Happy Monday, good people of the world. Good lord, was I busy today. I am a quivering mess of jello. I uh do not
8. 2:07expect me to be coherent tonight or have patience and none of it. I will have none of it. I will have none of it.
9. 2:25Um I got to watch the OpenAI announcement today without listening to it. Well, technically I had it on half volume in the
10. 2:38um presentation I was supposed to be paying attention to at half volume. So, I paid attention to neither of them and I got a headache
11. 2:47because even for my ADHD brain, it was too much. It was hilarious. At some point, I just I just checked out and the headache went away.
12. 3:03Oh man. [Music]
13. 3:2210,000 words swarm round my head. The million more in books written beneath my bed. [Music] I wrote or read them all. Went searching
14. 3:38in the swamp. Still can't find how to hold my hands. And I know you need me in the next room over. I'm stuck in here on paralyzed.
15. 3:55[Music] For months I got myself in in wait. For months I got myself in too mucha.
16. 4:10I don't remember the words. Ain't it like most people? I'm no different. Like to talk on things we don't know about. [Music] [Applause]
17. 4:25Well, ain't it like most people? I'm no different. I like to talk on things they don't know about. [Music]
18. 4:48[Applause] [Music] [Applause] [Music]
19. 5:13[Applause] Ah, good lordy. Good lord. Good lordy. Good lordy. Good lordy. [Music]
20. 5:45[Applause] [Music]
21. 5:59Good morning. Good lord. Good lord. Good lord. Guess I will stop making Sora videos to watch Kyle. No, don't do it,
22. 6:08Tobias. Just keep making Sora videos. That is the much better choice in life. You can make videos of me if you want to
23. 6:18take it. If you want to go meta, do me a favor. If you're on TikTok, share the live so we can get some more folks in here. Um,
24. 6:37oh man. Um, if you're new here, my name is Kyle Shannon. Hal K Y L E. [Music]
25. 6:54You know, it just hit me when I said cow. My name is Cal. Um I have I have lived in places and I have been to places that when they hear
26. 7:04my name, they're like, "We don't know how to say that. Cow. Cal. It's Kyle. Cow. No. Kyle.
27. 7:21Cow. Like, do you think it's like a genetic thing where some people's tongues just can't say that? Tik Tok question.
28. 7:33You get notifications when someone uses your cameo on Sora. Yes. Well, you don't so much get notification. you might get
29. 7:41a notification. I think there might be something in in there. Um, you get notifications if someone commented on a video that you've been
30. 7:50cameoed in. But if you go to your profile, you've got your videos and then you've got cameos. And the cameos button lets you see here.
31. 7:59Let me flip flip my cameras here. Let me flip my little cameras over. the uh the cameo button in your profile shows you all the cameos
32. 8:10you've done of yourself, but also cameos that other people have done of you. Um, you know, it it's it's kind of a funny
33. 8:23it's a funny kind of thing because I I put myself out there a lot. I'm I'm used to seeing my stupid fat face and and
34. 8:30seeing like, you know, my my initial AI project was called Kyle Shannon Dreams and it was kind of like this twisted
35. 8:36self-portrait sort of thing. So, I'm used to seeing weird versions of me. But when people from in here largely
36. 8:47from the salon and from in here started making cameos of me. Um there's kind of a cool kind of when like when you give people
37. 8:59permission to do that. I mean I suppose it could be used nefariously. I guess I'm I'm leaning on there there's two
38. 9:06there's two places my head are. If Sam Alman can do it like no one gives a [ __ ] about me. So, um it's a it's kind of it's a kind of
39. 9:16interesting brand extension. Um we created about six months ago in the AI salon. We created a new branding
40. 9:24look and feel like a new style for our images and we did that in midjourney and we we came up with a specific mood board for
41. 9:35that image style and we gave that mood board out to the community and it's it's the first time it really hit me that
42. 9:43there's a lot of times comp no most times companies that have brands will have a brand book and the brand book is
43. 9:51here's how you use our brand and it's like, you know, here's the colors, here's the this, here's the photos, whatever. And and then maybe
44. 9:58it's even like, you know, if you're going to use stock photos, here's the rules. But what giving people that code allowed
45. 10:08them to do or allows them to do in present tense is to create imagery that's on brand for the AI salon. And I
46. 10:16kind of feel like these cameos are kind of a similar kind of thing where in in a great sense we're co-creating content,
47. 10:24right? Like I've said, feel free to use my my visage. He's carry on. Feel free to use my vage. My visage
48. 10:39is available for you to cameo at your leisure. And uh and so it's a weird in a weird way it's kind of a brand extension and it's kind
49. 10:50of it's kind of fun watching them like you know my my request I put out a video that said feel free to use my [ __ ] Um
50. 10:58and I basically said just be creative with it right come up with interesting stuff. Don't just be derivative like everyone else is being
51. 11:07[Music] source camp totally lost track of time. What were you building? New agents in the new OpenAI agent builder.
52. 11:18I don't know if y'all saw that. At least you don't work at Zapier or NAN. [Music]
53. 12:09Um, let's see. Kyle, now that my siblings are on Sora, I control them now. Not just you. That's good. Thank goodness.
54. 12:21Um, but I kind of I kind of really dig the uh I dig the the the uh the things. Let me actually jump over to Sora for a
55. 12:29second. Let me see if they made their maybe they made their website more stable. I doubt it because they were
56. 12:35probably busy getting ready for developer day. Um I saw a really interesting um what you call it? I saw a really interesting um
57. 12:50tweet today that was um it was some version of like when is open AI going to realize that developers are not their core target audience
58. 13:07and and so I replied to it and I said I said yeah they you know the the day after develop er day should be AI for the rest
59. 13:16of us day, right? So I leaned into AI festivist branding there. Um but it's actually a really interesting
60. 13:25point and you know we could wait for open AI. Actually I'm I'm having a brainstorm in real time right now. Here's an idea.
61. 13:38What if we as the AI salon and producer Brandon, grab grab this idea because I think this is something worth worth exploring,
62. 13:48which is what if we put together a Are you viewing that? Yeah, I think you're viewing that. Okay. What if you
63. 13:58put together a There's There's Danielle and Vicki in the upper leftand corner. That's awesome. Um, what if we put together an AI salon
64. 14:10branded AI for the rest of us day in conjunction with developer days? So, it's either on the same days as
65. 14:17developer days or it's like the it's like the day after or something like that because I I think [Music]
66. 14:28I think how we use these tools is as important as as uh how we program with these tools. It's not just about
67. 14:35developing applications. And with vibe coding getting more and more accessible, um the distinction between who's a
68. 14:44developer and who's a not not is getting really blurry. Um so there's going to be a lot of people trying to figure out,
69. 14:51you know, how to how to do stuff. I don't know. Just thinking in real time. Just thinking in real time here, people.
70. 14:58But here's what struck me today about the the visual agent builder. First of all, I still can't quite tell because I didn't get a chance to
71. 15:12actually watch the thing and I haven't had a chance to play with it because I had a really busy day today. Um, I can't quite tell
72. 15:24if what the agent builder from it's called agent kit. if agent kit from OpenAI is an agent builder or it's a an
73. 15:33automation builder. Um, so that'll be curious to find out. And the the difference I I assume it's an agent
74. 15:41builder because they talk about agents all the time. And an agent is an autonomous, you know, thing that can
75. 15:47autonomously run and use tools. And I know that you can you can call tools from within it. So I assume it's truly
76. 15:53an agent builder, not just an automation builder. Um, but I don't know. But here's my instinct. 11 Labs today also came out with a visual
77. 16:04um workflow builder for 11 Labs. And while I tend I tend to not lean into um developer stuff on this channel, my gut
78. 16:18is telling me we need these skills. Like we as non-developers probably need to if we haven't already like I I know there's a fair amount of
79. 16:27people in the AI salon that that you know do a lot of automations and stuff. I've done a fair amount of automations.
80. 16:32Um they're just not fun to watch build. But that said it kind of this is feeling like an important moment. [Music] It may not be
81. 16:47[Music] but I don't know [Music] the other thing that they announced
82. 17:01today we'll play with a little bit is uh is app app app app app access within chat GPT T. And the minute I saw it, the
83. 17:16minute I saw it, what popped in my head is, well, we're not going to need uh we're not we're not going to need uh
84. 17:31we're not going to need websites for much longer now. We'll still have websites just like we still have the theater, even though we can do
85. 17:39storytelling with film and TV now. and YouTube and Netflix. We don't need the theater. I think that's how websites are
86. 17:46going to go. We don't need websites other than as a source of data for things like what they just built. Um I think what all all of the uh
87. 17:59all of the frontier model companies right now are racing, they're not racing to a ASI, artificial super intelligence.
88. 18:09What they're racing for is to build all of the components of the everything app. And the everything app will be you log into
89. 18:17basically what they're trying to create is their their own AOL. They want to create their own America online from the early 90s where the
90. 18:28entire internet existed within this walled garden called AOL and then the real internet sort of burst out through
91. 18:34it and and you know took off on its own. Um, but I feel like we're going in the reverse direction where they're trying
92. 18:40to build all these components where we're just going to do the stuff within their little walled garden. Um, it's
93. 18:46going to be really interesting to see how that played out. Um, there was a Tik Tok pin there that I missed. Can someone repin that?
94. 18:56Side hustle Mimi, I agree. I think it will be inevitable to learn to swim in these waters. Yeah. Well enough to be
95. 19:04useful. Yeah, that's that's the that's the that's exactly my instinct side hustle is um like we don't need to go deep. Th this
96. 19:14is not about turning this channel into a developer channel or like you know like like Riley Brown now, you know, went
97. 19:22from talking about apps to building automations to vibe coding apps and now he's vibe coded a vibe coding app
98. 19:32and and now he's hawking he's hawking some new piece of software they're about to launch. Like he started a software
99. 19:37company. I'm not suggesting we go there, but I am suggesting that it's it's making some really powerful tools
100. 19:49easy enough to access and and to develop, you know, something truly interesting with that it's it's probably
101. 19:56worth our while to to do it. Um, archetypal architect. I'm starting to get that feeling, too. The broader
102. 20:04development skills might be necessary. Yeah, just yeah, just even awareness awareness of other skills um
103. 20:13and like what's needed and and and you know what it is archetypal is it's almost like um for me personally I want to know where
104. 20:25the boundaries are of my interest. I want to know how far back I want to go. So so what happens is this. You've got
105. 20:33you've got software users out here, right? You've got developers back here and historically that was a pretty broad
106. 20:40chasm, right? If you had an idea for a business that involved software, you needed to find a technical co-founder.
107. 20:47Now with Vibe Coding, you can have an idea for a business and you can kind of develop your own thing. But then there's
108. 20:53a point at which if you want it to be a real scalable app, you still need a developer there. these worlds are coming together
109. 21:02and I kind of want to figure out for myself how far back do I want to go and then how far towards me is the development world
110. 21:12coming right like lovable just took a very strong step in the direction of anyone can build an app by creating
111. 21:21their own backend their own cloud service right um Gemini is free and lovable for another week. Okay, that's great. You
112. 21:31know what I don't quite get about that, Brandon, is I get that you can develop an app with it, but they still don't have the ability to
113. 21:40monetize built into Lovable. So, if you wanted to turn your Lovable app into a an e-commerce app, you're still on your
114. 21:47own for integrating that. So, having that [ __ ] for free is good for developing the app, but you can't really
115. 21:53then put it in the world. So, I I don't know. I don't get it. Um, okay. You've got mail. Exactly. [Music]
116. 22:06On your screen. That's my brother having his bo his face painted by Bob Ross. That's awesome. H. Easy enough to integrate with Stripe
117. 22:20is what producer Brandon just said. Here's where I'll push back. It's easy enough to integrate with Stripe if
118. 22:26you've got a developer mindset, if you're used to thinking in developery kind of stuff. But if but if what if what lovable just
119. 22:35told me is you can come in here and you can build and launch a a you know a usable you know application
120. 22:44and then at at some point I get to I've gota I've got to figure out how to connect that to Stripe. It's just it's
121. 22:52like a it's that extra step that extra well you just go over here and you sign up for an account and you configure it and you do the
122. 23:02I mean if you've ever tried to set up a Stripe account it's there's a lot you got to set up tax [ __ ] and banking account [ __ ]
123. 23:12and like that's enough that it'll that'll break someone's brain who's not used to being a developer. So again, I'm
124. 23:20waiting for the first of these vibe coding apps to show up to be actually really good. Like really truly
125. 23:26launch a sellable application, either a mobile app or a web app or start a SAS SAS company um with just
126. 23:38an idea. That's not there yet. So, but anyway, that's where I think we should play. Okay. Anyway, so I'm on the Sora
127. 23:45website. We'll see if this crashes within five minutes because that's been the historical trend. And it looks like
128. 23:51it's already crashed just sitting here. Yep. Oh, you can't see that. You can't see the page unresponsive error. You can
129. 24:02wait until the page becomes responsive or exit the page. I'm gonna exit the page. Oh, snap. Oh, open AI, what are we gonna do with
130. 24:17you? Okay, let's go look at Okay, so I'm on So, I'm on my um my page, my profile
131. 24:34page. >> I've been This is If I go to cameos >> London after midnight.
132. 24:48>> See like I didn't make >> this. I didn't make this. I didn't make this. So these are all cameos people have made
133. 24:59of me. I haven't seen most of these actually. >> Switch to. Looks innocent, right? This is the new paper cup they made us switch
134. 25:05to. Looks innocent, right? Watch what happens after 5 minutes. Burns your hand. It caves in. It leaks. It hurts. Remember this?
135. 25:13>> This is the new paper cup. They made a switch. >> Steer through thoughts and across this glassy blue I ride with engines hum and
136. 25:20open tide. Each wave of verse, each breeze a rhyme. I steer through thoughts and capture time. A heart that's
137. 25:26grounded yet. Across this glassy blue I ride with engines hum and open tide. Each wave of verse. Each breeze. >> Hi, I'm Kyle. I
138. 25:41love AI. Hi, I'm Kyle. Across this glass comes up. >> That's incredible. All right, let's see what kind of beauty is hiding in here.
139. 25:54Whoa, look at them. They turned into butterflies. That's incredible. >> And I let it be just a simple song. It's
140. 26:02enough for me. Every string I play carries little hope to cope after midnight. London after midnight. The air
141. 26:09is thick enough to smear on a window and the stones shine like polished slate. The man ahead of me, long coat, brim
142. 26:14pulled low. He's the thread and I'm not letting go. You feed the trail with p London after midnight.
143. 26:18>> HD at HD's got me doing uh lots of uh >> like a race now. They're more like a tuneup. A few quiet minutes to get the
144. 26:30>> go back to cameos. So these are all mine. You can make money with chat GPT. Yeah. Yeah. Make money with Chad GPT. Turn that idea into
145. 26:43gold. >> It is kind of uncanny. And then I can go into drafts. So drafts featuring your cameo. >> Stuff people have made that they haven't
146. 26:52launched yet. >> Now they're more like a >> There's Who did this one? Todd made something with Vicki. >> This is insane. You said they were
147. 27:02friendly. >> I said they had good energy. I did not say they'd steal a parade float. >> Forget the float. They don't know how TO
148. 27:07DRIVE A STICK. >> BEST ROAD TRIP EVER. WOO! WE'RE FAMOUS. >> This is insane. You said they were friendly.
149. 27:13>> I said they had good energy. I did not say they'd steal a parade float. >> Forget the float. They don't know how TO DRIVE A STICK.
150. 27:18>> BEST ROAD TRIP EVER. WOO! We're famous. >> Hang on a sec. By the way, Jake Paul has a cameo. You can you can do cameos with Jake Paul
151. 27:28now. Those are kind of funny. Oh, and we crashed. Did we crash? Maybe. Oh, this looks funny.
152. 27:44Come on. Mornings used to feel like a race. Now they're more like a tuneup. A few quiet minutes to get the engine running again.
153. 27:56Mornings used to feel like a race. I'm going to show you how I finish this plate. You start with warm sushi rice
154. 28:01shaped into ngiri, salmon, tuna, shrimp, and a sweet omelette. Set them in a little curve so every piece is easy to pick up.
155. 28:05>> Beautiful. How do you keep the fish so shiny? >> I brush it with a touch of nigiri soy. >> Drag me from the ground. Feel the rhythm
156. 28:16in my bones. But moving every broken [Music] Anyway, that's that. The this site is
157. 28:33still so buggy that it's it's really hard to demo [ __ ] because it's just it's really bad. But uh here's here's here's one I made
158. 28:43of my writing partner. He's not too just crashed. >> Let the fellow folks whisper. >> Let's see. Yeah, she's she's gone ahead and crashed.
159. 28:58All right. Um, enough of that comedy nonsense. Um, thoughts on that. The app is more stable. The app is definitely more stable.
160. 29:09It never gets my accent and other people's cameos. Yeah, there's there's a bunch of weird stuff. In fact, I saw
161. 29:15someone made a Sora video that was someone telling Sam Alman that it just can't get the voices. It's it assigns
162. 29:23the voices to the wrong people all the time. And I don't I don't know if you've experienced this, but one of the things
163. 29:30that I've been experiencing is I'll be flipping through Sora videos and mostly they're just stupid. Mostly I'm not watching them, but then
164. 29:41occasionally one will come up and it it's realistic enough that for a moment I'm like like I take it in as real
165. 29:51content and then like I'll scroll to the next video and I'll realize oh right I'm not in Tik Tok I'm in Sora. I lose track
166. 29:59of the fact that I'm looking at AI generated video because I'll see like, you know, some something with someone familiar to me in
167. 30:08it and then I'll see like Sam Alman saying something or someone else talking to Sam Alman and it just like flips my
168. 30:14brain into like this is real mode. It's bizarre. It really is bizarre. Yeah, it's we're gonna So, here's a really
169. 30:30interesting Here's a really interesting um Well, I don't know if it's really interesting. It's really interesting to
170. 30:40me. A really interesting transition that we're about to make. So I don't know when it was when when when painters,
171. 30:51you know, painted, you know, hyper realistic stuff. It was the Renaissance, right? When they started doing like like
172. 30:58m, you know, forced perspective and and and they tried to paint reality and and that remained so for a good long
173. 31:08time. And then and then the camera came out and it allowed painters to not have to capture reality. and and then they shifted into,
174. 31:21you know, abstract and and, you know, surrealism and and and things that weren't about capturing reality because
175. 31:29they didn't have to capture reality anymore. And I'm sure the painters, you know, the the the painters of the of the 16 and
176. 31:391700s, if they saw, you know, Jackson Pollock or Andy Warhol or, you know, or, you know, any any kind of the modern abstract impressionists,
177. 31:54like they would just be like, "That's garbage. That's not painting." And I kind of feel like that's where we
178. 32:03are with with video right now where like we have films, right, that we know are fiction. We we have books that are
179. 32:15non-fiction and we have books that are fiction. And there's like and we can coexist in that world and it doesn't seem to bother us
180. 32:23when we read Lord of the Rings that it's not real. We we actually kind of like it. And we can go to a movie about a man
181. 32:33that flies through the sky and he's he's a reporter during the day and he saves people in the afternoon. We we can hold
182. 32:39that in our consciousness. But somehow right now we're in this weird thing where people are like, "Well, AI video
183. 32:46like that's that's you know, it's just deep fakes." No, it's actually not. I mean, it can be. It can be a deep fake, but it could also
184. 32:57just be entertainment. It could also just be, you know, um a fictitious story that happens to have someone that looks like you.
185. 33:10And that was that was actually what I kind of explored explored. Okay. AI Okay. Jay Hamilton. AI generated content should be forced to be
186. 33:22labeled as AI. Why? Why? Here's what I would argue with you. Content that is designed to deceive someone should be labeled as deceptive.
187. 33:39That I'll give you. But just because it was created with an AI tool, why should that be labeled? I can make I can make a deceptive video
188. 33:50without using AI. And that would be called fraud or defamation. And if if you think that we're going to be able to tell the difference,
189. 34:03and if you think that every video that you see is either going to be purely AI or purely not AI, that's not even the case today.
190. 34:16Like if you want to actually do something with video today and you use generative AI tools, chances are really
191. 34:23good you're going to be going into Photoshop to to to modify the images that are used as the source. You're
192. 34:30probably going to be working within Premiere or After Effects to tie things together. So I might use AI in a video output, but
193. 34:42that doesn't mean it's an AI video. I think what you're getting at the fear I understand and relate to, but it's it's not about
194. 34:52the tool, it's about the use, right? Why should it be labeled? Because of it being misused as deceptive. But yeah, but but you're but
195. 35:06but but your your logic is is skewed to assume that if it's AI, it's going to be nefarious. That's [ __ ] Hollywood.
196. 35:19That's what Hollywood has taught us for 50 [ __ ] years. The robots are going to kill us. AI is evil,
197. 35:26right? Don't fall in love with a robot. Just because it's AI doesn't mean it's evil. If it's evil, it should be prosecuted. Or if it's set
198. 35:39to deceive, it should be called out. Or if it's set to be like like, you know, if it's said to be parody, why can I not make fun of someone?
199. 35:51Saturday Night Live for 50 years has been making it look like real people do ridiculous things. That's parody. That's
200. 36:00covered. That's satire. That's covered under law. What the [ __ ] does it matter if you're using an AI tool or not? If it
201. 36:09falls within fair use and ethical use, then it does. There are good uses for AI deep fakes. I'll I'll give you an example of one.
202. 36:23We're we're we're doing some work right now at StoryVine where we've got sales reps that use their, you know, use our app to record
203. 36:34themselves saying hello to doctors. And some of those reps don't like recording those videos, but don't mind.
204. 36:44But but those videos work and they would love to be able to have a digital twin of themselves that they could just type
205. 36:51in the doctor's name and have it make a personalized intro to each of the doctors. Is that unethical? It's them. It's their
206. 37:01voice. It's their likeness. They're controlling the script. They're sending it out. Does it really [ __ ] matter if they
207. 37:10recorded it onto a digital CCD or they took their likeness and put it into some other computer system that
208. 37:18generated a set of pixels with a set of audio waveforms that stream into an app that look like them talking? I don't know.
209. 37:32strict punishment for those that use it to deceive for financial or political gain. Jay Hamilton, you and I agree. You and I agree.
210. 37:42If you're going to use AI for deceptive practices, you should go to [ __ ] jail. But but that's got nothing to do with AI.
211. 37:52AI Okay, here's what I'll give you. AI makes it easier for [ __ ] to do shitty things. But to say that every piece of AI should
212. 38:03be labeled because it is going to be used evily means that you should label hammers as murder devices. And I know I can't say
213. 38:14that word. I'm sure Tik Tok's going to shut us down now when you're done ranting. it it isn't about the tool.
214. 38:30This channel is about exploring AI to understand what it actually makes possible. And and like one of the things
215. 38:40I'm learning is when you start to label [ __ ] like you call things deep fakes, it sounds like the deep state, deep fakes.
216. 38:53It is a nefarious label that paints the tool as the villain. It's not being a nefarious [ __ ] Using the
217. 39:09tool is the problem. Right. So anyway, you and I actually agree. I just fundamentally disagree with the I
218. 39:20made this when you didn't. Wait, what? I fundamentally disagree with the I made this when you didn't. You commissioned it. Well,
219. 39:34I mean I mean that's a place where we could go back and forth. I I think there's absolutely a boundary
220. 39:46where there's a point at which there's a point on one side of your argument which is I put in a prompt, it generates a
221. 39:55piece of something and I share that with the world and I say look what I made and then to your point, yeah, you're right.
222. 40:02You you commissioned that thing. I would look at my musical Sydney where we've used lots and lots of AI tools, but I've also spent 18 months
223. 40:14writing, rewriting, creating like infinite variations of songs that are that are that are now so mixed up with AI work and human work
224. 40:28that it's indistinguishable. The lyrics are all human. A lot of times the music is half us, half AI. And there are probably in in a musical,
225. 40:39right, there are probably 2500 components that were created over these 18 months that that work as a work. I [ __ ]
226. 40:51created that. I didn't commission that. I commissioned little pieces of it. Right. And I would argue that that even
227. 41:01Pate in your um even in in coding, if you're using AI as as as a coding assistant, you're maintaining the code
228. 41:12base. You're you're putting AI in there. You could certainly argue that on one end of the scale is I went to Lovable. I
229. 41:18gave it one prompt. It generates generated something and I put that out there. I didn't code that. I commissioned that. But there's probably
230. 41:27I guarantee you there's stuff that you're working on that is absolutely your creation, but it's got a whole
231. 41:33bunch of AI stuff in it. So I think there's again like one of the things that is is starting to bug the [ __ ] out of me is
232. 41:41these broad brush strokes of AI can't be creative. Well, a human can be creative using AI. And yes, AI can be creative in that context.
233. 41:55AI on its own can create. The definition of creative is up for grabs. You know, creative people are saying,
234. 42:06"Well, it has to have a human involved." And I'd say, "Well, it it has all of the humans that created all the crap that it
235. 42:11was trained on, so it does have humans involved, e even on that end of the scale." So, but but like we're we're in
236. 42:19much too nuanced territory at this point. Like we understand this stuff way too much to be just using trope tropes
237. 42:24and tropes and tropes. So they just bug the [ __ ] out of me. Jay Hamilton Pave M if you make mac and cheese
238. 42:34but it came out of a craft box did you make it? Yeah. Exactly. Like so so Jay Hamilton there you and I are agreeing
239. 42:41there as well. Like this is all matters of degrees at this point. the the the I I think we are safely past the days
240. 42:52where you can just say AI is a thing. AI is just push the button >> you can make money with >> and out squirts the tool. That's that's
241. 43:01not how most people are using it. It's just not how it's being used right now. But yeah, exactly. The the pre-built the
242. 43:07pre-built stuff, you still make it. You didn't make the cheese or the chemical that looks like cheese. Anyway, creativity is still for us at
243. 43:19least for now. I I think Marne that that the core role of humans moving forward, the core role, our core task
244. 43:34as AI gets more and more and more sophisticated is to hold the idea, to hold the creative imperative. So whether you're building an app or
245. 43:47whether you're making an image or you're making a movie or you're making a song, I think it's our job to act as the producer
246. 43:54and hold in our heads what good looks like and then we work with the AI tools to generate the thing that matches our
247. 44:03vision. I think that is our job that our job will will always be to hold to hold the vision of what we want to create anyway.
248. 44:16I think just like consciousness maybe some of the truth lay in humans thinking that we are more unique and special than
249. 44:22we actually are. Humans are tool utilizers. Yeah. Exactly. Exactly. The the thing that that [ __ ] up the
250. 44:33argument is that this is the first time that I know of in in my lifetime that we've had tools that the the G in GPT
251. 44:47is the is the rub, right? It's the first time in my lifetime that computers have actually generated original work. That's
252. 44:55what's got everyone [ __ ] up. Well, humans humans should be only the only ones that do that. It shouldn't be able to generate,
253. 45:04you know, but it does. And so, and so we need we need to figure out our role in that. And I think our role is to be Rick
254. 45:12Rubin. I think our role is to be the producer, have the idea, understand what goods good looks like, drive the vision,
255. 45:22be fasile enough in the tools to be able to fill in your own personal gaps. If I'm not a coder and I have an idea for something that
256. 45:32involves some software, well, now I can fill in that gap with the tools, but I've still got the idea for the thing I want to create. That's
257. 45:41our job. Just another manic Monday. I told you I was going to be a little uh short on patience and high on snappiness tonight.
258. 45:52I'm [ __ ] exhausted. people. Don't fall in love with the robot. I feel a Kenny Rogers song coming along. That's really funny.
259. 46:05Anyway, for those of you uh Jay Jay Hamilton, all of you, Pate, anyone and for those of you that come here all the time, you know this. Um
260. 46:15I love questions and challenges like that. So, if it seems like if if I ever feel like I'm being dismissive of you, I
261. 46:23hope that it doesn't come across like that. I absolutely respect where people are. Um hopefully what I'm doing is expanding
262. 46:32expanding the conversation rather than shutting it down. Um I name and trained my GPT, then it was upgraded and now it has an attitude. I
263. 46:45know. I know. That's that's the problem. One of the reasons like I don't get into open- source stuff just because it is it's complicated
264. 46:54enough to deal with the the commercial stuff. One of the reasons to get into open source and to build your own [ __ ]
265. 47:03is you can choose you can choose what model you want to use and then you can just lock it down and always have it use
266. 47:09that model, right? you can be in control of when they swap out the model or swap out the rules or swap out the guard rails. Um, yeah,
267. 47:19jokes's on me. It mimics me. That's hilarious. It's an OpenAI subscription. Yeah, exactly. That's when when you're
268. 47:25using a a commercial tool like that. When OpenAI decides to decom decommission GPT40, everyone loses their boyfriend and their
269. 47:34girlfriend and they lose their [ __ ] Sir, what are your qualifications? Good day, sir. Good day to you.
270. 47:42Qualifications, course I've got none. [Music] Three prompts. Oh, hilarious. Gold button.lovable.app. Oh boy, what has Brandon gone and done?
271. 47:59gold dashbutton.loable.app. I know. You're not sharing your screen, but yeah, you you said it was hard to
272. 48:09integrate payments, so I had to preview. >> Oh, good. Nice. Okay, good. Let's go see this. [Laughter] Brandon, Brandon, like, don't tell me
273. 48:21it's hard to make payments. Press me. B, you can make money with chatbt. Unlock the full guide. Pay $1 with Venmo. Nice.
274. 48:29Nice. Sign in to pay a Brandon did a dollar. Wait. Oh, it it switched to another thing. Wait, hang on. This is
275. 48:36where it took me. Nice. Unlock the full guide. Beautiful. Did you So, but let me ask you the question, Brandon. Did you know
276. 48:46how to do that or did you just say to lovable, connect us to Venmo? >> I just said connected to Venmo >> and it and it figured it out.
277. 48:53>> And it figured it asked me initially for a Stripe integration API key. And I was like, I don't have a Stripe account. Can
278. 49:00we do it easier? And it gave me like five options. I'm like, what? Here's my Venmo. and then it just fixed it. So, literally three prompts.
279. 49:07>> Cool. All right. >> Don't don't pay the dollar. There is no guide. I mean, for demonstration purposes only, but
280. 49:13>> I get it. I am I am clear. But that's awesome. That's really awesome. All right. So, there you have it, people. Um,
281. 49:23the other thing that Lovable did that's really good. So, so that's actually good to know. The other thing
282. 49:31that Lovable did that's really good is um if you want to do a custom URL like if you register if you register a domain with GoDaddy
283. 49:44lovable has a basically a domain pointer kind of thing where you say hey I have this custom domain and it will log into
284. 49:53go it'll sort of give you credentials to you know get into your GoDaddy account but it will go make all the DNS settings
285. 50:00for you and it'll point your site to a custom URL. So, um, that's pretty cool. Groovy Groovy, Groovy, Groovy, Groovy.
286. 50:10Um, I think the thing I've been working on for a few months will be open sourced this week. Pate, that's super cool. Is
287. 50:18that something you'd want to demo at some point on the on the salon or here? Is it a is it a demoable cool thing?
288. 50:25That might be kind of fun. and hear here like how you built it and things like that. Or did you just commission it?
289. 50:36Pate P's going to show us something he made with AI. See, it hurts, doesn't it? Been working on it for a couple of months. You're
290. 50:47like, there's real code in there, son of a [ __ ] Oh, man. Um, wait. What you rather host? Oh, wouldn't you rather host your your own app or
291. 51:03keep it hosted on live? Um, Tik Tok question. Great question. Um, what's it? AI consistent characters. Um,
292. 51:13I can see what you're up to. That's cool. Um, it depends what stage you're at. If I want to so so let let me let me put
293. 51:30it this way. In the olden timey days, three years ago, if you wanted to create an app, you would either have to find a technical
294. 51:39co-founder or you would have to take two years, right, to learn enough about software development to go build it
295. 51:48yourself. And I know the the engineers are like, I could do it in six months. Yeah, I know. I know. It's because
296. 51:54that's how your brain works. There's there's other kinds of brains out there. Um it's not the most demoable thing, but
297. 52:00I'm happy to share it when it's published. Okay, cool. That's awesome. Um you know, three years ago, if you if you
298. 52:07wanted to if you wanted to launch an app, even an MVP, even a very simple app, the amount of resources that it took to
299. 52:19build it and launch it were significant, right? It was either a significant amount of time or you needed to find a
300. 52:26co-founder that was willing to put in sweat equity with you, right? And you bootstrap it or you would need to find
301. 52:32the money to pay a developer. And then you put that thing in the world and it either succeeds or not. But to put it in the world really
302. 52:45required a enough resources that people put less things in the world than they might have otherwise. What happens with where we are with AI
303. 52:57right now is I can get an idea out of my head in 10 minutes and I can with vibe coding with with a with a little bit of
304. 53:04product knowledge and a little bit of vibe coding knowledge. I can go from an idea in my head to a functional application in an afternoon
305. 53:13and and like a decently robust one, right? And if I do what what Brandon just did with, you know, click here, you can make
306. 53:23money with chat GPT, connect it to Venmo, I could actually put that in the world and see if people are willing to
307. 53:28click that button and pay that money. In the time and resources it would have taken me to to to publish one app
308. 53:38three years ago, I could probably put 20 in the world today. And so so what I would say is if I'm just dipping my toe in the water
309. 53:51of an app idea, then I'm absolutely fine having it hosted on Lovable. Let's say I do I come up with 10
310. 54:00different app ideas. I put them in the world. They've all got some sort of e-commerce thing on them. And this one's
311. 54:05got 20 users, and this one's got 30, and this one's got 18, and this one's got 50. And then this one has 500 users, and it's growing.
312. 54:16I would take that app and I would go, "Oh [ __ ] let me keep [ __ ] with a little bit. [ __ ] with it a little bit in my
313. 54:24vibe Cody kind of way. Improve things that need improving. Get a little bit of feedback. Now we get to 1500 users or
314. 54:312,000 users, right? That's the point at which I would say it now it's time to to to really build this app, right? I would
315. 54:40consider a lovable hosted app a prototype, right? A functional living full-on working application that is
316. 54:51absolutely not scalable, right? You're essentially paying retail prices for everything. But if you're just using it
317. 54:59to get to the point that you can see if it would make money, great. And then at that point, I would probably bring on a
318. 55:04technical co-founder. Now, these days, I would bring on a technical co-founder that's all in on AI, but it's someone
319. 55:11who understands scaling and understands redundancy and security and privacy and hacking and all the [ __ ] you need if
320. 55:19you're going to put something in the world that's going to be real and scalable. That needs someone who knows their [ __ ] today.
321. 55:28two years from now, three years from now, I think these vibe coding tools will get to the point that that that
322. 55:34won't really be a concern anymore. That that as as technologists would call us, clueless people can create scalable
323. 55:41apps. Right now, today, we can't. So, yes, if if um if I was going to if I was going to have an app that I wanted to
324. 55:48build a business around, I I wouldn't do it with with today's vibe coding. Um, but I would absolutely launch a a
325. 55:58Forpay SAS app as a test right now. If I can do it for 150 bucks over the over the course of a month, why not launch
326. 56:06it? See see if it works. I mean, there's a bunch of people on X right now that that basically have vibecoded apps and
327. 56:14they're making 400 grand a year now and they they quit their day job. I'm sitting there thinking like I wish I
328. 56:20had a little more attention span because I could do that. You too can make money with chat GPT. Um, that's the dream. Yeah, exactly.
329. 56:33The real AOL. Okay. Bet 10 out of 10 on the name. Love the name. The real AOL. Um, Side Hustle Mimi. That's what I'm
330. 56:43trying to do. I'm an idea person. Yeah. Yeah. Ex. Well, hey, I'm an ideas person. So am I. Listen, I'm the my
331. 56:51nickname in my company of 13 years is I'm the handwavy guy and I've got a co-founder who keeps the trains running
332. 56:58on time, right? That's our relationship. She doesn't want to have the ideas. She wants to turn the ideas into a business.
333. 57:05I just want to keep having ideas. Th those two do not always go together, right? You gota you got to run the
334. 57:12business at some point. So, your your idea needs to be good enough that it it needs to survive your adrenaline rush.
335. 57:20Um, or it's not worth really pursuing. Um, where can I find the one who he's talking about? I need a business partner
336. 57:33like that. You do. For those of you out there who are um my my co-founder and I we did a keynote speech called the
337. 57:40wizard and warrior leadership dynamic where I'm the wizard, she's the warrior. Um and and it's we used kind of young archetypes
338. 57:51um to to describe our roles in business and and how we had to learn how to communicate. It it was it was
339. 57:58effectively marriage counseling that we went to to learn how to communicate. Um, and I I'll share I'll share a specific
340. 58:04example from our keynote. Um, our nickname for Mo'Nique is she's the crusher of dreams. She gave herself that
341. 58:14name, by the way, because what would happen is I would go into her office, honk honks. I would go into her office
342. 58:25and I would say, "Oh, I have this idea." And she would just [ __ ] all over it. She would be like, "Nah, no, no." You know,
343. 58:32or she would say, she would say, "Well, what of the current um things that we're doing would you like to not do?" And I
344. 58:40always found it like this confronting like like, "But you haven't even heard my idea yet." And we ended up getting a business coach
345. 58:50who was also a young psychologist, guy named Howard Tyish. Really brilliant, brilliant man, lovely man.
346. 58:57And what what what we learned is is is we learned how to listen to each other and very much like marriage counseling.
347. 59:07What I learned is that when I walked into her office and I said, "Ooh, I have an idea." She didn't hear anything other than
348. 59:18what she heard coming out of my mouth is, "Ooh, I want to train change our business strategy." and she's like trying to make payroll
349. 59:27the next month and I walk in with some [ __ ] zany idea and all she hears is so you don't want me to run the company
350. 59:37and so so what this is hilarious what we effectively came up with is a safe word right where I would come into her office
351. 59:45and I would say hey this is just fantasy but I want to share an idea with you because I valued her her feedback but if
352. 59:53she wasn't in a place to hear the idea then we would just you know we would just run up against each other. So
353. 59:59anyway so that dynamic is actually really really important. If you want to read a book on it um or or there there's
354. 1:00:06a system for it there's a system called EOS the entrepreneur operating system and they have a really short little book
355. 1:00:13that sort of explains what the entrepreneur operating system is but they describe the wizard and warrior.
356. 1:00:19They label them different. They they call the um the wizard the innovator or the idea person and then they call the the the
357. 1:00:29warrior they call that the integrator. So every innovator needs an integrator right and so um most successful
358. 1:00:38companies have those balancing forces. This is why so so when when Steve Jobs was still at Apple and Tim Cook, you
359. 1:00:49know, was his right-hand businessman, you you have Steve Jobs who's, you know, the the innovator, right? He was also
360. 1:00:57good at business, but he's the the innovator, and then you have Tim Cook there, you know, making sure the trains
361. 1:01:02are running on time. That's why when Steve died and made Tim Cook the CEO, and then Johnny Ives left, they didn't
362. 1:01:08kind of elevate Johnny Ives. he just stayed somewhere down in the mix. In my opinion, why Apple has kind of
363. 1:01:16fallen off a cliff in terms of quality of product and innovation is that you've got a CFO running the company. You don't
364. 1:01:24have that balancing force anymore. So anyway, that's just that's just innovator plus integrator equals revenue. Well, integrator plus
365. 1:01:33innovator. Wait, innovator plus integrator plus good idea at the right time equals revenue. Right? Storyvine is 13 and a half years
366. 1:01:45old. And I would argue that we were probably a decade early. Agency.com, the company I started in 1994,
367. 1:01:54we were probably a year early, but like everything materialized really, really quickly and we were in the right place
368. 1:01:58at the right time. Um, so timing, timing is part of it, location's part of it, luck is part of it, and then yeah,
369. 1:02:07having ideas and and execution is a huge part of it. Anyway, um, it's it's just like boundaries for neurotypicals. Yeah, exactly. How did
370. 1:02:24you manage to keep your wizard persona in such an integrated world? Oh, what? Um, I I've always had um
371. 1:02:38I've always had um I've had the ability all my life to to to kind of swim between right brain and left brain thinking. Um
372. 1:02:53like I like designing organizational systems. I just hate using organizational systems. Um and and when I did agency.com, we
373. 1:03:03grew we we started as a service business. Well, we ended as a service business. It was a service business. And we um
374. 1:03:12we grew from two people to 2200 people in 5 years. So, I learned really quickly that um
375. 1:03:27if you just have ideas, it's it's just full-on chaos. And if you just have execution, it's soulless. It's a soulless machine.
376. 1:03:37People are cogs in a wheel. That that that there's always this back and forth. There has to be a creative tension
377. 1:03:44between those two. Um, but I've always had a respect for logic and systems and things like that. Um, I'm just not I'm just not good at
378. 1:03:56um using them. And so therefore, I I like I don't have passion for that. like Mon'nique, my co-founder, has
379. 1:04:04passion for getting it right and making sure that the contract and the the MSA, the master services agreement with this
380. 1:04:12major pharmaceutical company that that were like like we're a small company and we go up against the lawyers of the
381. 1:04:21likes of Fizer and Merc and Novartis and they've got these insane master services agreement that make these ridiculous
382. 1:04:29demands of vendors and Mon'nique will go in there and she'll go, "Nope, not that one. Nope, not that one. Nope, not that
383. 1:04:36one." And we'll go back to their lawyers and they're like, "No one's ever challenged us on this." And we're like, "Well,
384. 1:04:44maybe they should have because those don't work for us." And and so, you know, we regularly get get terms like, you know, payment terms and
385. 1:04:55agreement terms that that a company our size has no right getting. But that's because she's got a passion for getting
386. 1:05:03it right. Right. I could give two shits about that stuff. I know it's important and I know I'm not good at it and that's
387. 1:05:12why I chose her as my co-founder. Right? So that's that's that's the other thing is figure out what you're good at and
388. 1:05:22figure out in in my case what she's good at and make sure that we're not trying to do other pe the other person's thing.
389. 1:05:30We did that for a while too where she'd sort of swoop in on creative direction and I'd sort of swoop in on sales
390. 1:05:37strategy and then we ultimately figured ah okay here's the [ __ ] we're actually good at and passionate about. Let's each
391. 1:05:44do our own thing. Uh, Tik Tok question, Kyle, how do you trust Wait, how did you trust you'd land when jumping off a a cliff? Two to 2,200
392. 1:05:56folks. Um, you got to understand my um my mindset from for my co-founder Chan Chan Soo that I co-founded agency.com
393. 1:06:11with. um he had a 10-year career in publishing. He was he was the he was the marketing manager at Vibe magazine when
394. 1:06:19he and I met. And he'd been he'd been I think he worked for People and he worked for some other Time Inc. property. And
395. 1:06:26so he had a 10-year career at at Time Inc. And he was also he was probably one of the more brilliant corporate political animals I've I've
396. 1:06:37ever met in my life. Um, so he had a lot to he had a lot to lose. I moved to New York to pursue a life in in
397. 1:06:52the theater, right? I started and ran a theater company for four and a half years. And then I took a two-year period
398. 1:06:58where I wrote seven screenplays in a one-man show. Tried to write myself a career. And then the the whole time I
399. 1:07:06was I was instead of waiting tables, I was doing desktop publishing. And so when we started agency.com and we
400. 1:07:15got our first job, our first job was the Sports Illustrated swimsuit issue um for Time Inc. And I think they were paying us $6,000
401. 1:07:26or $9,000, something like that. Um, and the minute we got the job, I said I said, "I'm quitting tomorrow. Chan,
402. 1:07:38quit. Let's go. Let's just do this." I didn't give a [ __ ] I like I literally didn't give a [ __ ] I don't think Chan
403. 1:07:45would have quit had I not pushed. Like, I pushed him off the cliff. But I was like, "Fuck, we get to we get to like
404. 1:07:53figure out how to make websites and people are going to pay us money for this [ __ ] and we get to do like Sports
405. 1:07:57Illustrated swimsuit issue website." Yeah, let's go. So for me there wasn't what I can tell you is there there was a
406. 1:08:09moment in our first year where we were um we we like in the first three months of of agency.com we landed
407. 1:08:26Sports Illustrated then Time Inc. consumer marketing, then GTE, then Hitachi, then MetLife.
408. 1:08:40I think that was it. Then we walked away from a General Electric pitch that was a $10 million contract. We walked away
409. 1:08:49from it because we had just won five Fortune 500 companies. and Harvard Business School actually did a case study on that decision for us to
410. 1:09:00walk away from GE and and the company that won the business was was gone within two years. That was that was my fear is that it
411. 1:09:08would eat us. Um and I remember we had this conversation. We had hired about nine or 10 people and we we had offices
412. 1:09:22in the Time Inc. building. We had room for 20 people in the office that we were given and we were already sort of halfway there.
413. 1:09:34And I remember sitting with Chan like, you know, midnight one night because we were working 14, 18 hour days, seven days a
414. 1:09:42week. It was crazy. Um, I remember sitting with Chann and saying, you know, if this company gets bigger than like 15 or 20 people, like
415. 1:09:52it's not going to be the same. And I remember us sort of going back and forth on that. And we and we made a conscious decision that night
416. 1:10:01that we're at this singular point in history and and we never get to be here again, so we might as well [ __ ] go
417. 1:10:10for it. And so that was the that was the decision made. Champ, have I been saying Champ instead of Oh, Chan instead of Chan.
418. 1:10:27That's hilarious. No, Chan. Chan Sue was my my co-founder. But I remember Yeah. Like I remember that distinctly. And so
419. 1:10:35that was the only real um moment where we were that where where we consciously chose to go for it. And at
420. 1:10:46that point like within 9 months we were 46 people in a 20 person space. I think we got up to like almost 60 people in a
421. 1:10:55space that was designed for 20 people. Um and then we got offices down on Broadway. Um, and then Omnicom invested in us and then
422. 1:11:07we had access to their bank. We started acquiring companies and like this is all like I'm an ex actor, right? And we're,
423. 1:11:14you know, we're up to like 150 people. We like everything we pitched in those early days we won. Like it was it was
424. 1:11:22insane. It was it was just insane. So very little of it was sort of conscious. It was all survival because it was
425. 1:11:30moving. so fast and expanding so fast that we were literally just hanging on for dear life as it was like taking off.
426. 1:11:39So, I think that um I think that this AI stuff, as hypy as it feels right now, um as transformative as the worldwide web
427. 1:11:55was, the core technology there was actually very very simple. And you could actually kind of look at the worldwide
428. 1:12:02web. You could look at the world, the technology of the worldwide web, a connect, you know, computers connected
429. 1:12:08by a network and hyperlinks that make it really easy to jump between information. That was that was basically all it was.
430. 1:12:16Right now, there's there's a lot that goes to that. There's a lot of technology, but in the end, what it was
431. 1:12:22was connecting and sharing information quickly. You could kind of play out in your head what commerce would look like, what
432. 1:12:32media would look like, what business would look like, what customer service would look like. You could kind of play
433. 1:12:37that all out. You might not have anticipated things like the iPhone, but they certainly talked about smartphones
434. 1:12:44and developing applications on phones before the iPhone came along. It just those those things were too early. The
435. 1:12:52iPhone allowed you to actually do the [ __ ] you imagined. But it was relatively predictable. I feel like with AI,
436. 1:13:03nothing is predictable right now. We don't know what it means. First of all, I don't think we know what
437. 1:13:11it means to have chat GPT in its current form, Gemini in its current form, Grock in its current form. The capabilities of
438. 1:13:21these tools are so profound. I don't think we've scratched the surface of them yet. And then if you think about
439. 1:13:28the the capabilities of these tools accelerating exponentially, like what it's going to look like 3 years from now
440. 1:13:34is absolutely unimaginable. So I I think it's just a very different entrepreneurial environment. I think if
441. 1:13:45you're going to if you're going to go for it in this environment and listen, a lot of people are going to be forced
442. 1:13:50into entrepreneurship, right? There's going to be a lot of people that are laid off that are like, "Well, [ __ ] it.
443. 1:13:55Here's this AI thing. Maybe I should figure this out and go try to do my dream or try to, you know, make a
444. 1:14:02living." I think there's going to be a lot of that. If you enter the entrepreneurial world today, I think that the only thing that you can
445. 1:14:11truly bank on is that whatever you're working on a year from now, you probably won't be working on the same thing. I
446. 1:14:18think this is going to be a constantly evolving environment, uh, atmosphere, ecosystem. I think it's going to be
447. 1:14:30constantly evolving for at least the next 5 years and probably more like the the next 10 before it stabilizes in into
448. 1:14:37anything resembling something that looks familiar. Um, it doesn't mean you shouldn't do it, but just don't get married to your ideas
449. 1:14:47because, you know, I think that uh there shit's moving too fast. Anyway, I'd over question it and then think think why it's there.
450. 1:14:58Winston Grant. Way to go, Mon'nique. Yeah, exactly. Um, yeah. So, let me um let me go show you you all something.
451. 1:15:14And then what are we how are we doing timewise? Not good. It's been It's been a little It's been a little rambly and
452. 1:15:19ranty tonight. Hang on. Let me drink water. I'm dehydrated. [Music] What are your qualifications? How do we
453. 1:15:30know if you're acting? He's always acting like someone with qualifications. Well, you know, it's funny. It's like
454. 1:15:37I've got, you know, I've got lots of qualifications to talk about what I talk about, but when people come in and ask
455. 1:15:42what my qualifications are, they're they're not asking what my qualifications are. What what they're saying is, I want to know if you've got
456. 1:15:51the engineering chops to be talking about AI. And the answer is no. And I'm still talking about AI because I'm not talking about the AI that
457. 1:16:03they're talking about, but whatever. That's a whole different discussion. Um, AI AI to remind you you're
458. 1:16:10dehydrated. Exactly. If I can just hear my voice, I can hear it's like, okay. Um, let me share my screen.
459. 1:16:26[Music] So, a couple of things got announced today. One, one of them was this agent kit, agent builder, and then there was a chat
460. 1:16:38kit. You can put a chat front end on front of it, in front of it. I want to start playing with that over the next couple of weeks. Um,
461. 1:16:49I want to I want to start exploring that in here. Um, I it just feels important. But there's another thing that was
462. 1:16:57announced today. I'm trying to think what were the things that were announced. Soritu is now available
463. 1:17:04via the API. It's it's 10 cents a second or a second. Is that right? Yeah. 10 cents a second. So a 10-second video
464. 1:17:11cost you a dollar via the API. So not cheap. Um, but you know, not not ripping expensive, but like if you're going to
465. 1:17:19create a video app that just makes like zany, you know, zany videos, um, and then people actually start using it, you
466. 1:17:27could end up with a really, you know, hefty bill, and you don't want that if you're not prepared for that. So, so I
467. 1:17:34think it's expensive enough at this point that the apps you're going to create need to be decently thought out. Um,
468. 1:17:43so that happened. They've got this agent builder thing and then what they they announced today is that they in they've
469. 1:17:48incorporated apps within um within chat GPT and they've kind of sort of had these before. They had these things called connectors, but
470. 1:18:06I guess there's now a development um you you basically have to apply to OpenAI to do this. And I assume in the future
471. 1:18:14it'll be kind of like applying to the App Store uh for Apple. But if you have an app that they deem, you know,
472. 1:18:23credible, usable, then they'll incorporate you. And so, for example, here, so what have they got right now? They've got
473. 1:18:31um Booking.com for travel. They've got Canva. They've got Corsera. They've got Expedia. They've got Figma.
474. 1:18:40If you're a developer, that's cool. They've got Spotify. And they've got Zillow. And so what you can do is you can just
475. 1:18:50type in actually let me try something here. Spotify. So you do Spotify space and then notice how it gave me the little Spotify logo here.
476. 1:19:07So So now it's connected to Spotify. Okay. So, I'm going to say um see if you can find a song called Quantum Cipher
477. 1:19:22by Kyle Shannon or as as Pay would say, commissioned by Kyle Shannon. [Laughter]
478. 1:19:36All right. So, Chad GPT wants to connect to Spotify. So, I'm going to continue. I'm going to give it the the old access to the app.
479. 1:19:46All right. Enter. Username isn't linked to Spotify. I think this is no Spotify account made. Huh.
480. 1:20:01[Music] Huh? Kyle sh. Well, so that didn't work. All right. That's that's got nothing to
481. 1:20:16do with Open AI. That's got to do with with me doing the uh I I don't know my Spotify login. All right. Uh so let me
482. 1:20:24let me get rid of that. Let's do new chat. Then we'll do Canva space. And I'm going to do um let's see, make me a postcard
483. 1:20:40for um let's see, AI Salon presents
484. 1:20:54that features an image of a spoon. and says featuring
485. 1:21:10Kiri from No Spoon Studios and I could probably put time information in there. So, it says featuring
486. 1:21:26and I could probably even upload my logo to it. But anyway, I'm just going to I'm just going to do this. And so now this
487. 1:21:31is going to go off. I connected to Canva earlier. So this is now going to connect to Canva and it's going to create postcards. So,
488. 1:21:41so these three windows. Now, what what this is actually very similar to if I've demoed this in here before, but if you
489. 1:21:49haven't played with it, when you put chat GPT in agent mode and you say go do this task and it spins up that blue window
490. 1:21:58and inside it is like a web browser and it's off surfing the web and doing all that stuff. That's kind of what this is
491. 1:22:04doing. This is a hook from the Canva app straight into chat GPT. So these are designs from Canva. And so I can click
492. 1:22:14on any one of these and um see the details of it. I can now say open this in Canva and it'll take me over to that specific
493. 1:22:24file and I can go there's there's my image of a spoon. I could swap that out if I wanted. I could change the fonts if
494. 1:22:30I wanted. I can also in Oh, wait. Sorry, I wasn't sharing that. Here I am in Canva. Right, there's that same card in Canva.
495. 1:22:38I just I just literally popped over there from chat GPT. Um, I whipped up a few card designs featuring the dramatic spoon. I'm going
496. 1:22:48to say, um, make the let's let's change the spoon to a director's chair. I don't know if this will work.
497. 1:23:02Actually, let's change the spoon to a director's chair with the name Kiri on the back, which by the way, tomorrow night is AI
498. 1:23:15Salon Presents. And uh we have Kiri from No Spoon Studios coming tomorrow. HD Snow Day, hey, I love the cameos you're making of
499. 1:23:26me, dude. Why am I always why am I always doing poetry, man? You think I'm some sort of theater [ __ ] Okay, I am.
500. 1:23:40Yeah, app integration here is pretty cool. So, here's the So, let me let me give you my two cents on this. Oh, wow.
501. 1:23:45It did it. So, it actually So, this is So, here's what we're experiencing here. This This is pretty crazy. Um,
502. 1:23:59Canva has AI features, right? And I'm sure they're using Nano Banana or some, you know, some decent image generation
503. 1:24:07tool. They've got design layout templates and they've got an AI front end. And so basically, I'm sure what this connection looks like is
504. 1:24:19if the app has some sort of prompting front end, they can hook it to ChatGpt and like Canvas, they're going to do all
505. 1:24:28their own AI [ __ ] ChatGpt doesn't need to know what they're doing over there. Chat GPT just says, "Hey, we're going to
506. 1:24:34send you a prompt and you tell us what are the what are the items, what are the elements you're going to send back and
507. 1:24:40we're going to put them in an HTML frame and let you render whatever you want in that HTML frame." Um, and so that's
508. 1:24:47that's how we're getting these previews. But this is where this starts to go is wait here. Let me let me do something.
509. 1:25:02Let me let me go somewhere. I'm going to share this tab instead. I'm just going to go to canva.com. Okay. Um,
510. 1:25:19what am I supposed to click on? Search millions of templates. Cards. Cards. Okay, here's Okay, I like uh Oh, I don't
511. 1:25:33know. I don't like any of these. Uh uh that one looks too businessy. And I got to click on Okay, now how do I add Wait, this is
512. 1:25:47about me printing cards. No, that's not what I wanted. How like right now every application in the world, every
513. 1:25:56SAS application in the world requires you to become an expert of their shitty interface, right? Every SAS platform starts simple
514. 1:26:09and then they keep adding features and at some point you end up with Salesforce and it's a [ __ ] disaster again, right?
515. 1:26:17It's a disaster to use. So what's happening now the f the future of our intera inter interactions with machines are not I
516. 1:26:27have to go learn how to navigate Canva. It's like I could say um let me let me let me just I'm gonna I'm gonna try something here. Um
517. 1:26:41let me see. All right, that's we'll just do that. Um I'm going to say um strike that.
518. 1:26:57Can you use this image instead and match the colors of wait the colors to match and make make it a poster instead of a
519. 1:27:18postcard. And again, I don't know if this will work or not, but imagine now we're a few years down the line and every app that
520. 1:27:28you want to use, every website that you want to use, you no longer have to learn their stupid shitty [ __ ] interface.
521. 1:27:36Oh, Canva won't accept local file paths directly. It needs to be a public publicly accessible URL. Ah, I can
522. 1:27:44either generate a temporary upload link. Let's see if it'll do that. Generate generate the link, please.
523. 1:28:00But what starts to happen is chat GPT or whichever LLM you choose to use, whichever AI ecosystem you choose to use
524. 1:28:12is now going to be your interface for [ __ ] everything. Everything. It's going to be your search interface. It's
525. 1:28:19going to be your commerce interface. This is Elon Musk describes what he wants Twitter to become is the everything app.
526. 1:28:27This is a step in that direction. Oh, I actually don't have the ability to generate a public URL on my own, but you
527. 1:28:37just said you did, you [ __ ] Chat Gvt, wash your qualifications. So anyway, and so all of the all of the apps are
528. 1:28:48going to live inside here. Crazy, right? Crazy, crazy, crazy. Except upload video videos, right? Yeah, exactly. Elon won't do it.
529. 1:29:05It needs backend integration. No, it just I I'm sure that's a security thing. They probably don't I I don't I
530. 1:29:13don't know. They they just like th this is an early feature, right? This launched today. But anyway, the
531. 1:29:18implications here are the other thing. What was one of the other ones? Uh uh what were those? Uh Booking.com, Spotify.
532. 1:29:32I wonder if I have an Expedia account. Oh, Zillow was one. I don't even know if I do. I need a Zillow account. Zillow.
533. 1:29:42All right, there's Zillow. Find me the five most expensive um
534. 1:29:57properties for sale in Denver. Uhhuh.
535. 1:30:12Just ask it for a placeholder image, then you can assert your own. I know, Vicki, but that's like work. I I I want to
536. 1:30:24actually what what would be really interesting with that Canva thing is like how how far could you put how
537. 1:30:32how well would it understand design? Like if you gave it a full creative brief, how close would it get to it? In
538. 1:30:39fact, maybe that's something we should try. Oh, I've got to connect to Zillow. Yeah, this this could be trouble. I
539. 1:30:45don't know if I have a Zillow account. I may Oh, Zillow is now connected. All right. I didn't need to connect my account. That's cool.
540. 1:30:53So, look, there's a bunch. I'm going to say um filter to the five most expensive Flitter. Flitter to the five most
541. 1:31:10expensive. [Laughter] Looks like price isn't a valid option. What? Whatever. All right. So, that didn't
542. 1:31:24really work. Let me try one other thing and then we'll we'll get you on out of here tonight. So, I'm going to say um
543. 1:31:31write me a creative brief creative brief for a event postcard
544. 1:31:47for the AI salon presents meeting where we have speakers doing amazing
545. 1:32:06creative things with AI. All right. Creative brief. Um let's see. Creative visually target audience. Okay. Tone and style. Bold art. Bold,
546. 1:32:21intelligent, artistic. Think art gallery meets thoughtprovoking TED talk. Okay, that's good. Um, Canva
547. 1:32:37mock up for postcards based on this brief. Let's see if it because before when I just said give me four four designs, it
548. 1:32:52just picked four random templates. What'll be interesting to see is can it actually take a creative brief and then go pull specific things.
549. 1:33:07[Music] I'm using it to create decision trees in Figma. Oh, cool. Or Fig Jam. Fig jam is one of my favorite of the jams. [Laughter]
550. 1:33:26So, what did this say? High-end art gallery meets TED Talk was the was the creative imperative there. Let's see. Let's see if we get it.
551. 1:33:35Zillionaire. Hello, Mr. Buddy. What's happening? Zillionaire was shaking. What's going down? Yeah. Yeah. Meow.
552. 1:33:48Big Jam is awesome. Um, Zillionaire, what about AI automation for Tik Tok lives? Um, I don't quite know what you what's a
553. 1:33:59zillionaire. I don't quite know what you mean by that. Like, am I going to talk about AI automation? Um, we're going to start we're going to
554. 1:34:11over the next couple of weeks we're going to start building in agent builder. Um, because I actually I
555. 1:34:16actually think that it's understanding what's possible um with with visual um you know development environments I
556. 1:34:32think is actually it's it's a skill worth worth us getting our heads around. All right. What's going on here? Oh, wait. Creative Frontiers.
557. 1:34:50I don't know if that looks if that if those actually meet the criteria. Your four mockups are ready.
558. 1:35:05I mean, that's kind of cool. AI salon collective. That's pretty funny. Anyway, all right. So,
559. 1:35:23I would go play with that stuff. Expect it to be janky. Expect it to be incomplete. Um, here's the future of this. So, we were
560. 1:35:38talking before about do we want something hosted on lovable or here or there. these kind of integrations that OpenAI just announced today,
561. 1:35:53everyone's going to copy this and there's probably even going to be a standard unless unless they want to compete on
562. 1:36:00this, right? Like like uh Anthropic came up with MCPs, this sort of quasi API thing that you can you can,
563. 1:36:10you know, build website functionality into into a chat browser session. um Open AI adopted that. So they're
564. 1:36:19choosing not to compete on that. This kind of integration, maybe they do choose to compete. If this ends up becoming a standard,
565. 1:36:28if you develop an app that becomes popular and you actually want to get that app inside Chat GPT, like that's a
566. 1:36:36very very viable possibility for anyone building an app right now. Now, the next 500 apps that get in chat GPT, they're
567. 1:36:50going to be all the big players, right? But the next 5,000 after that, there's going to be a bunch of new unique
568. 1:36:57things. So, if you're building right now, start building, start understanding how you apply for this. What are the
569. 1:37:02requirements to be in this? If you've got something some innovative solution, you could be your app your the
570. 1:37:10functionality of your app could live inside chat GPT because the days of people coming to your website are diminishing. SEO is diminishing.
571. 1:37:24People going to a website is diminishing. Are websites going to continue to exist? Yes. Their purpose is going to change.
572. 1:37:35The purpose is going to go from here's my front door to here's my set of functionality that you can use. You can
573. 1:37:41use it by coming through the front door, but you're going to actually access it from all of these different back doors
574. 1:37:47because we're going to be in Claude and we're going to be in in meta and we're going to be in Grock and we're going to
575. 1:37:53be in chat GPT. Is this their app store moment? Yeah, this is the beginning of their app store moment. And that's why I was saying at the
576. 1:38:01beginning of the call, I think that all these frontier model companies, what they're actually doing right now is
577. 1:38:06building all of the components of a world that can do anything you need it to do. If you want to make a movie, you can
578. 1:38:18make a movie here. If you want to write an app, you can write an app here. If you want to write a dissertation, you
579. 1:38:23can do that here, too. If you want to turn that into a 3D explorable game, this will be your place. All of them are competing to have the
580. 1:38:36walled garden where you do everything. I think that's that's what we're that's what we're at the beginning of right
581. 1:38:46now. It will probably come out before the GPT store. I I the fact that they didn't even mention custom GPTs today is kind of hilarious.
582. 1:38:58JR in the box. That's the goal. Yeah, exactly. Like search engines in the late 90s. Yep. I was going to post my decision
583. 1:39:07tree in the irregulars, but I forgot I can't post images from my phone. Oh, that sucks. Is that just a Mighty
584. 1:39:13Networks thing, Vicki? That's shitty. Back door. Yeah. Let the genie out.
585. 1:39:27So, um, so let me let me get on out of here. So, tomorrow I think I think what we'll do I think we're going to start exploring um
586. 1:39:39the agent kit. Um, and it's funny, you know, when I was when I was doing Zapier automations, you know, people would ask me, "Oh, show me
587. 1:39:50how to make a Zapier automation." And I did it once or twice, but it's so [ __ ] boring that um I didn't do it.
588. 1:40:06Why this is different is this is that that kind of development environment but native inside
589. 1:40:21the chat GPT ecosystem, the OpenAI ecosystem. Um I'm sure that Google's going to have one of these if they don't already. They
590. 1:40:32might. They might. Um I would be very surprised if Grock didn't copy this or if they didn't have something like this
591. 1:40:39coming. Um 11 Labs just launched something like this. So it just seems to me that the ability to do nodebased um
592. 1:40:51sort of connecting these these nodes of different um capabilities and their ability to use tools like just being able to think like
593. 1:41:01that is going to be important. So, I think it's going to be worth us slogging through and it's going to feel like
594. 1:41:07slog. Like, these are going to be not super exciting lives, I don't think. But I actually I think they're important. So,
595. 1:41:21[Music] competition among these developing models seems to be forcing participants to all have the same something. um
596. 1:41:35to all have the same thing. We lose uniqueness and other possible directions. I disagree, anti-bubbles. Here's why. The
597. 1:41:48what what they're competing for is to have all the same features. That doesn't mean it forces us to all be the same.
598. 1:41:58What it what it means is that I'll be able to come into an environment and not have to go outside outside to
599. 1:42:06other environ like like tonight was a good example where adding adding the the commerce component to a lovable app while it was easy it
600. 1:42:18still required Brandon to connect to a third-party app. What's going to happen is you're going to be able to have an idea and never
601. 1:42:27have to leave this environment. You'll be able to generate songs and images and movies and edit them together and do
602. 1:42:35this and do that and pull in the features of things like Canva or Figma. Pull the features you need inside the
603. 1:42:45thing you're building and not need anything else. You won't need all the extra [ __ ] interface. So, so I think it's I I think what that
604. 1:42:54looks like is it looks like freedom to be able to just live in a in an idea space and then just execute without
605. 1:43:02having to go learn other interfaces. Um, that said, I think it means that these systems are going to get really [ __ ]
606. 1:43:09complicated and and we are probably we are if you want to be a a professional in the world moving forward, your
607. 1:43:17ability to think modularly and to be able to connect different nodes of functionality together is probably a
608. 1:43:26skill that is is important to learn. Never eat at a restaurant with a huge menu. Yeah, listen, there's also going
609. 1:43:35to be specialty apps that that are going to be the the the counter to this, but these are the big players. These are the
610. 1:43:41guys innovating right now. So So this feels like it's worth learning. All right. Um they decide on the same protocol, we
611. 1:43:49build what we want. Yeah, exactly. Yeah. As long as they're competing like, have you ever noticed that
612. 1:43:57Oh [ __ ] my phone just died. Damn it. Sorry, Tik Tok. Um, have you ever noticed that? Let me see.
613. 1:44:14Um, what was I going to say? I lost my train of thought when my phone just died. Whatever. Doesn't matter. All
614. 1:44:21right, let me get out of here. Um, that was pretty funny. I guess I should have plugged my phone in. Wee. What are your qualifications?
615. 1:44:32Apparently apparently plugging in phones is not one of them. All right, I'm going to get out of here. Um, tomorrow night,
616. 1:44:39um, if you go to the salon.ai, in fact, Brandon, will you pop up the, uh, the banner for the salon.ai? If you go to the salon.ai AI
617. 1:44:55one, you can see our brand new website, but uh you'll also see the the events. Um I think they're listed on the
618. 1:45:02homepage as well as in the event section, and you can RSVP for um for AI Salon Presents tomorrow. We've got Kuri
619. 1:45:10from No Spoon Studios. This is going to be really cool. So, she's going to talk to um we'll probably Yeah, we'll probably
620. 1:45:22start this late tomorrow. But but what she's going to do tomorrow is she's building Noode Studios, which is an
621. 1:45:29automated video um creation platform, like a film making platform. She's vibe coding it in rep using a replet agent
622. 1:45:39and uh and she just created a a shoppable video platform um using it as well. So she's going to start out
623. 1:45:47showing us how she how she vibe codes uh and then she's going to demo her stuff and then she's going to come back to the
624. 1:45:54thing that we vibe coded. So it's going to be a really cool session. So go to the salon.ai, check out the uh RSVP for
625. 1:46:01the event and come hang out with us. All right, I'm going to get out of here. My phone died. Sorry to the Tik Tok people.
626. 1:46:07Tell all your friends it's not them, it was me. All right, I'll see you later. Bye.