10/14/2025 - English Is the New Programming Language: How AI Empowers the 99% Who Are Not Coders

Transcript

1. 0:04[Music] [Applause]
2. 0:19[Music] Wow. [Music] Oh,
3. 0:38[Music]
4. 0:55every time I see it Get that look in mine. Every time I see your mouth, I hear that smile. The early misty morning light. I heard
5. 1:12the engine turning the old photo. [Music] You were leaving me
6. 1:26today. You will convince me [Music] again today.
7. 1:43You're leaving this old town looking for someone else's golden ring. [Music] should have said so long [Music]
8. 2:01cry [Music] so long. [Music] Don't you grab me.
9. 2:15[Music] Oh yeah. [Music]
10. 2:44Aha. [Music] [Applause] [Music]
11. 3:05Which programming language would you say you're most proficient? This is an easy one. English. [Laughter] [Music]
12. 3:17English is the new programming language. Uh I'm not a coder. I'm not a coder. Well, this is called the AI learning
13. 3:26lab. How could you do AI without coding? Uh because you can. Um, [Music] I mean, I suppose pushed it would be
14. 3:42HTML, but the HTML I know is from the '9s, so it doesn't look anything like the HTML that you do today. So,
15. 3:55um, Kyle, want to teach us how to use perplexity to merge multiple documents? Uh, if I knew how to do that, I would love to.
16. 4:06What is it with the stumpers? Everyone coming in. Sir, what are your qualifications, sir? Yes, I was inquiring, sir, when it comes to
17. 4:15programming languages. Sir, would by chance pray tell your qualifications? I ain't got none. Rather, carry on then.
18. 4:31Stump Kyle Tuesday. Um, so but seriously, um, if you're new here, welcome to the channel. This is the AI learning lab. And if you're like,
19. 4:44why are you playing guitar? I thought we were here to learn code. Um, so here's my here's my um let me give you a little
20. 4:52bit about my background. I I have been an entrepreneur I don't know most of my life and in the mid 90s I started one of the
21. 5:04first online magazines and one of the first digital agencies in 1994. So I was there for the very beginning of of that.
22. 5:11The reason I got into the the web development that early is that I'd been intrigued by the internet for years, but
23. 5:20for years the internet had been the purview of data scientists and engineers and librarians and it was command line
24. 5:28only and you kind of had to know IP addresses and [ __ ] like that. And it was very very technical and I'm I'm I'm not
25. 5:36non-technical. I'm I'm fine at product development, you I know how to do [ __ ] but I'm not a programmer. I'm not a I'm
26. 5:43not an engineer. And so the worldwide web comes out and it makes all of that command line stuff as simple as clicking a hyperlink.
27. 5:55And that was the moment where I realized, oh [ __ ] something is different here, right? this the internet is now available
28. 6:06for the 99% of people who are not coders and librarians and engineers and mathematicians and academics.
29. 6:15And so the online magazine I started was an art and culture magazine because the internet back then all it had on it were
30. 6:22dissertations and scientific research and things like that. And I thought if you could put up something with culture
31. 6:28and art and things like that, it would probably get noticed because there's nothing on here like that. And so I put
32. 6:34out this online zen called Urban Desires. And I uploaded some learned taught myself enough HTML to put
33. 6:41together the first issue. Uploaded the files to some server in California. I didn't even know how to do that
34. 6:48back then. Your computer had to call another computer literally. that thing. So, I had my computer call a computer in
35. 7:02California and hand it some files and three weeks later there was a full page article about urban desires in a Parisian newspaper.
36. 7:13Three weeks later, full page article in French in Paris. And in that moment, I knew that the world had changed and I knew that almost
37. 7:25no one else on the planet knew that the world had changed. But I got to experience it. Geography had collapsed
38. 7:31and the distance between, you know, sending something used to take weeks or days now was instant and I knew that was
39. 7:39going to change everything. This channel is about what happens when that exact same moment happens for AI and machine
40. 7:50learning, which is November 30th, 2022. Chat GPT launches. Chat GPT does for AI and machine learning what the worldwide
41. 8:01web did for the internet. So, while I'm not a programmer, what I'm about is how did the 98.5% of us that are not
42. 8:09coders and engineers, I think it's gone up a little since since I was doing things back in the 90s, 98.5% of us who aren't that, how do we
43. 8:19use this stuff? And if you are an engineer, it's it's still completely different because it can now code. And
44. 8:26if you're like, well, it can't code as good as me. Well, have you used it lately? Right? And give it two years. Have you
45. 8:34have you look at the growth looked at the growth of it? So what does that mean? And so what this channel is about
46. 8:38is about getting curious and understanding how it works and what the implications are and what the future of work is and what the future of
47. 8:45self-expression is because in the end I think what our job becomes is not to have the tactical skills.
48. 8:53It it it matters less and less every day if I'm good at this programming language versus that one. What matters more and
49. 9:00more is what's my idea and do I understand what I want like what's my intention so so I believe that our job is to is to
50. 9:14become much more like a producer like a Rick Rubin where our job is to hold the fidelity of the idea to have the idea I
51. 9:22want to do this and hold the fidelity of that idea and then understand how to use enough of these AI tools to get that get them to
52. 9:32do all the leg work to pull off your idea and amplify your idea and reimagine your idea. So that's that's what's going on here.
53. 9:44It looks like a zen. Exactly. Visionary. [Music] My dull sister-in-law is utilizing GPT
54. 9:58on a day daily basis. Yeah. Yeah. And like you know if if you haven't played if if you're a coder and
55. 10:06you haven't played with a tool like like Replet Agent or Lovable in particular, Lovable for me is the is the first of the vibe coding apps
56. 10:18to really start to hit a level of usability for non-coders that they can still build fairly robust apps that have database backends and
57. 10:30authentication and things like that. They don't have e-commerce in there yet, but I'm sure that'll come. Um,
58. 10:37but just like you can speak an image into existence with MidJourney and a video into existence with Runway, with
59. 10:44Lovable, you can speak an app into existence. And and you know, coders and and engineers are poo pooing that,
60. 10:50saying, "Well, that's not real coding, right?" which is the exact argument that photographers said when digital cameras
61. 10:57came out and what painters said when when cameras came out when photography came out right in the what 1700s 1800s I think 1800s 18 184547
62. 11:13the dgeray type came out [Music] that's not real art you just push a button literally the same argument as
63. 11:24what what's going on today with AI. Oh, you just push a button. >> You can make money with your >> That's not work. That's not real
64. 11:31creative, right? Well, a funny thing happened on the way to the um demonization of AI is that all of the tools to a tool, all
65. 11:49of the platforms have gotten increasingly sophisticated and increasingly complex. And if like for example, if you want to be competent
66. 11:58at chat GBT right now, chat GBT is not a single tool, right? The last time I counted, and they've added some since then, the last time I
67. 12:08counted, chat GPT was 27 distinct tools or features that radically altered what it did or what it could do. 27 different tools.
68. 12:20Do you know them all? Do do you like Right. Do you know what code interpreter is? Do you know Do you know um what what
69. 12:29uh chat GBT agent is? Have you used it? Do you use it on a regular basis? Do you know what it's good at? What it's not good at?
70. 12:37Like, it's it's absolutely [ __ ] insane. Do you use advanced voice on a regular basis? Did you know that you can now upload
71. 12:44videos into chat GPT and it can analyze them for you? That didn't exist two weeks ago. And every single tool, whether it's a
72. 12:54video tool or an image tool like MidJourney, um, every single tool just keeps adding features and features and
73. 13:01features and features. So, so we we have absolutely crossed the point where any single person can keep just can even
74. 13:09keep track of what's happening. No one can. So, one of the things that I strongly recommend is um Brandon, if you could pull up the the the
75. 13:23uh what's it called? AI salon. The AI salon. Yeah. Banner. Um, one of the things that I I I believe deep in my soul, like something
76. 13:35that I'm very passionate about. So, in 1994 when I started Urban Desires and I co-founded agency.com, there was a third
77. 13:41thing that I did. There was a third thing that I created. It was called the Worldwide Web Artist Consortium.
78. 13:46Whack. And it was it was this New York City-based group that was anyone who was trying to figure out what the hell this
79. 13:53worldwide web thing was. Come twice a month to this meeting and we're going to talk about it. And it became kind of the
80. 14:02central hub of people trying to figure this [ __ ] out. And so the week chat GPT came out, I started the AI salon. And
81. 14:08the AI salon is a very similar a similar sort of ethos like come with your curiosity. Try to figure out what this
82. 14:14stuff's all about, but do it in community. And here's why. We're we're in an we're in an inventive expansive phase of AI right now. So 1994
83. 14:28to 1997 or eight the worldwide web nobody knew anything about it. There were no manuals. Everything was changing. They kept
84. 14:41changing HTML. They kept changing browsers. They kept changing how how you found stuff on the internet. So there
85. 14:48was a three or four year period where it was just primordial soup. And the only people that were doing anything with it
86. 14:54were figuring out on on their own. And so what that community did, what that group did was it it allowed us on a on a
87. 15:02bi-weekly basis to get together and say, "What did you figure out?" Oh, what did you figure out? Ah, okay. That's cool.
88. 15:07Let me go back to my shop and do those things and I'll come back next time and tell you what I figured out. That's what
89. 15:14the AI salon is and that's what this meeting is about. Uh, Tik Tok pin chat voice found a folder in my notion today. Oh, that's exciting.
90. 15:27So, advanced voice used to be fairly disconnected from your chat history and and things like connections and things.
91. 15:34Do you know what do you know what connections are in chat GPT? Have you enabled enabled a connection? Do you
92. 15:40know what MCPs are? Do you know that chat GPT supports them? Do you know that chat GPT has five different
93. 15:45personalities? I could actually go show all this stuff instead of just rambling about it. But it but the point I'm making is
94. 15:54we're in an expansive phase of AI right now. And I don't think it's going to stabilize probably for another 3 to four
95. 16:01years, right? We've we we're three years in as of November, November 30th, 2025. We're three years into the chat GPT era.
96. 16:10I think we've probably got another three, probably more like five where the tools just keep expanding and morphing
97. 16:17and changing and what's possible keeps changing and morphing. You can't keep up with it, but you also can't ignore it.
98. 16:26You can't just sit on the sidelines. In my opinion, I think anyone today who's who's actively choosing not to participate with AI is
99. 16:37doing themselves a massive disservice. Massive, massive. Because from where I look at things, AI is not going away. You don't invest
100. 16:53in invest hundreds of billions of dollars, trillions of dollars, right? Nvidia just committed a hundred billion
101. 17:00dollars to to invest into Open AI, which OpenAI is then going to hand back to them and then they're going to give them GPUs.
102. 17:08It's one of those Silicon Valley kind of deals, but it's a hundred billion dollars. You don't put a hundred billion
103. 17:13dollars into something and and have it go away. It's not going away. So if it's not going away, then we as human beings have two choices
104. 17:25and it's it is literally binary. You either deal with it or you don't. This channel is about dealing with it.
105. 17:33What is it? How does it work? What's cool? What's not cool? What's dangerous? What's exciting? And you know what we're finding in here
106. 17:42is that all the tropes, all of the tropes that are like the robots are going to kill us and the the just all
107. 17:47that [ __ ] it's the world's greatest plagiarism, tend to be things that people use as excuses why they're not participating with AI. And
108. 18:02what we find is the people that are participating with AI, if they kind of strap AI on like a jetpack, it amplifies
109. 18:09their ideas and it's incredibly inspiring and and can make a huge difference for people. Um, is it going to absolutely decimate jobs? It is,
110. 18:20but it's going to make whole new categories of jobs possible. And then your next question is going to be, well,
111. 18:26what are they, smarty pants? I don't know. We literally cannot anticipate what's coming. Just like the farmers in the 1800s could
112. 18:38not anticipate what would happen when steam engines took all the farming jobs. 80% of the agrarian society lost their jobs
113. 18:50in the ad with the advent of the steam engine. 40 40 to 50 year period. 80%. Did they all just stop working? No.
114. 18:59You know what they did? They moved to New York City. They moved to cities. They moved to Cleveland. [Music]
115. 19:08We get to make it up. We get to make it up. We do. We We are in Okay, here's the other thing. Like, why am I so
116. 19:15passionate? Why am I willing to go live five nights a [ __ ] week and just like very often feel like an abject idiot
117. 19:24because I don't know what I'm doing because they keep changing the [ __ ] rules all the time and the tech and the interfaces because we are
118. 19:37if you're if you're on this call if you're paying attention to this if you're listening to this we are in a rarified amazing
119. 19:48time in history where we're the only people who are going to know what it was like before AI and know what it was like
120. 19:56after AI. And if you participate in that transition period to a great degree, you get to write some of the rules. You get
121. 20:06to tap into the opportunity. You don't just have to be Sam Alman to have a lot of opportunity. Right now
122. 20:13there are uh what's it called? Wind was it windsurf? Windsurf was a vibecoded app about vibe coding apps that sold for $3 billion
123. 20:26in a year. Like like the opportunity right now is [ __ ] insane. YouTube question thoughts on AI trying to gain control of itself
124. 20:37um whether for self-preservation etc. um like how we saw recently. So with with with very few exceptions,
125. 20:53those moments are happening in test environments. Now they are happening, right? Like like the AI is being
126. 21:02deceptive and and and trying to do self-preservation. Um, but those are happening for the most part in test environments. If if if
127. 21:11there's something that has sort of escaped the labs, I haven't heard of it yet. Um, it is absolutely possible
128. 21:25that the AI does that. I'll give you another possibility. If we do the alignment right, if the labs do the alignment right and say that
129. 21:36the imperatives of humans, this is something that anthropics working on a lot, they call it alignment or super alignment. If the imperatives
130. 21:46of the AI are in line with the imperatives of humans and prioritize humanity, it is also possible that the AI will save us from ourselves.
131. 22:01Right? Because by far the most dangerous actor in an equation with AI right now and human beings are human beings. We're
132. 22:12[ __ ] We're we're idiots and [ __ ] with egos and we you know everyone's like, "Well, AI hallucinates." Have you met a human
133. 22:22being? They [ __ ] lied. They're deceptive. They're power hungry. They're egotistical. Narcissistic. Right. Tribalistic.
134. 22:33We're about ready to take ourselves [ __ ] out here. It is possible that the big destructive thing that AI does
135. 22:40is save us for from ourselves. Johnny D question. Iterating on drafts in perplexity changes overwrites past work when editing.
136. 22:51Iterating on drafts in perplexity. I don't know about that. If it does do that, that's pretty shitty. I can you not go back in your chat
137. 23:04history and restore earlier versions? I think you can, but maybe I'm not thinking of the right feature, but I'm
138. 23:13pretty sure like in in anything I know in Claude when you do when you do claude artifacts when you do chat GPT canvas,
139. 23:22you can scroll up in your chat and there's there's save points along the way. So you might be able to restore
140. 23:28that stuff. Rick McCauley, we made it. What's happening, dude? Those are all downstream of fear. AI does not have fear. Exactly. Preach.
141. 23:37Hallelujah. Jim Ross, what's happening, man? Excited you're in here. Yes. But they just create new drafts which are great but different.
142. 23:49[Music] Oh. Oh. Oh, I know what you're doing. Okay. Yes. So, okay. So, Johnny D. Okay. Yes. So, this is a general thing with
143. 24:00with AI. Okay. Let me show you something. Okay. Let's we we'll go do some showand tell right now. I'm not going to do it
144. 24:07in perplexity just because I haven't been in perplexity in a while, but LL LLM's all do this and it drives me
145. 24:13[ __ ] mad. I think I understand what you're saying now. Okay. So, we're going to go we're going to go um
146. 24:20um let's see. Write me a short story that has turns and twists and is about a young couple
147. 24:37um trying to figure out if they are more than wait if they are more than just friends. Oh, did I not share? No. Okay, fine.
148. 24:51Whatever. Understood. More than just friends. Um, set in. Let's see. Set in a carnival. Okay.
149. 25:05So, whatever. This is going to write before the lights come on. So, so it that's good. It remembered that I wrote
150. 25:11a song called Before the Lights Come On. Um, which I'll I'll play for you in a second. The carnival was still half
151. 25:17asleep when they slipped through the gate. the boy and the girl friends since fifth grade. Now old enough to know that
152. 25:22something was different but too young to say what. The air smelled like sawdust and syrup and promises that wouldn't be
153. 25:28wouldn't last till morning. Mia walked ahead. Okay. So So we've got her thing here. Okay. So now if you want to let's
154. 25:35see. She hesitated and then climbed in. The wheel creaked upward. If I just say um let's see uh they pass the mirror maze.
155. 25:52I'm gonna say instead of instead of passing of passing the mirror maze
156. 26:08um have them walk by the goldfish bowl game and the carney
157. 26:23is dumping in the goldfish. Okay. And so now it's going to rewrite this.
158. 26:39And let's see, did it rewrite it? The carnival was still half asleep when they slipped through the gate. The carnival. See, it
159. 26:47didn't do it. So a lot of times what it'll do is you ask it to change something in the middle of this and
160. 26:52it'll rewrite the whole thing. So a way you can get around that is and I don't know how this is done in perplexity but
161. 26:59if I say if I say now add this to a canvas add this to a canvas in chatgpt a canvas turns this chat where I just have to keep
162. 27:12chatting against the whole thing into an editable document. Enthropic is red teaming trying to create the worst case scenarios. Just
163. 27:22vibe coding some sites while I listen to you. The usual. Um, beautiful, beautiful, beautiful. Yep. It's rough, Kyle. Um, okay. So,
164. 27:34now I've got this thing in a canvas in chat GPT. Slashcanvas would be good here. What do you mean slashcanvas would be good? Oh,
165. 27:43yeah. That's exactly right. That's what I'm doing. So, this is a canvas. So now in the canvas I can I can click on this
166. 27:49edit button and this is now editable. So I can now come in here and I can just take let's see they walked they walked
167. 27:59on silence stretching between them. So I can just highlight this one sentence and I can do things to it like I can hit
168. 28:07a pl a plus here editor explain. Let's see highlight this and then I'll ask chat GBT. I'm going to say um expand
169. 28:22this into him singing her a silly song.
170. 28:37And so now what it's going to do is it's going to replace the that sentence but leave the rest of the document the same. All right.
171. 28:47Side hustle jumping in from O'Hare coming back from the create conference. Beautiful. Is she jumping into
172. 28:55into the uh into StreamYard, Brandon, or just she's jumping in in Tik Tok? Tik Tok. Cool. Hey, Side Hustle. What's happening?
173. 29:06Kyle was one of my favorite South Park characters. What's happening, Cal? Hey, Cal. They killed Kenny. What's up, Kyle? Um,
174. 29:17so anyway, so they walked on. So now that that sentence has become a big long thing. So if if uh if you're frustrated with with chat GPT
175. 29:29completely overwriting [ __ ] this is the way to do it. And it's listen, what you're describing is one of the problems
176. 29:36with vibe coding right now is you you tell you tell one of the vibe coding apps to like fix a button and it
177. 29:43completely re refactors your code and [ __ ] up your app. Like that's just kind of where we are right now. It's a
178. 29:49disaster. Here from Pensacola, Florida, new to your channel. Welcome, welcome, welcome, Coach Melissa. Coach Melissa,
179. 29:55what's happening? Welcome. Um, where are you? Where are you with the uh with the whole AI thing? Like what what got you
180. 30:02here? Were you just stumbling and you were like, you know, God, Alec Baldwin looks like [ __ ] Like what was the what
181. 30:08stopped your scroll? Where where are you? What's vibe coding? Okay, cool. So, um let's go jump over and do some of that.
182. 30:18Um let me share this. So this thing, what you're looking at right here, this is a a a an add friendly project
183. 30:30manager that I that I built that that um I was writing some I was trying to organize some thoughts on uh a
184. 30:39whiteboard and I thought well wouldn't it be cool if I could just move these elements around and then I went to
185. 30:45lovable and two hours later I had built this app. So, vibe coding, there are these apps like this one called lovable.
186. 30:53So, I'm going to go there right now. And so, it says build something with lovable. So, it's it's just like it's
187. 31:00just like chat GPT, right? You got a little prompt there. And then what the hell do you put in the prompt? You can
188. 31:06describe an application and Lovable will build it for you. It will write the code. And in fact, Lovable just added a
189. 31:14thing. They they call it lovable cloud where normally when you create an application that you've got what's
190. 31:21called the front end and you've got the back end and the front end is what everyone sees and that's usually the
191. 31:26simple thing and that's that's the thing where people like ah it's easy to make an app you just put some buttons and a
192. 31:32little field right that's the front end the back end are things like authentication and databases and and and
193. 31:39web servers and things that actually power the front end right so you got front-end developers back-end developers. What Lovable's done is
194. 31:47they've basically built that into a single thing. So, it does the front end for you and then you literally just say
195. 31:54turn on the lovable cloud and it enables all the backend database um access to AI features like you can put nanobanana. We
196. 32:01could we'll we'll make a nano banana app right now. And if you're wondering what is a nanobanana app? What the hell is
197. 32:07nanobanana? If you're like Kyle, are you having an aneurysm? A a nano bananurism. What? What the hell are you talking
198. 32:15about? Nano Banana is the new image generation model from Google, from Gemini. That's really amazing. Uh,
199. 32:23Coach, oh, I just missed that pin. What was what did that pin say from coach? I've been using AI for about a year. Um,
200. 32:29and in am in an AI course right now. Awesome. Welcome, welcome, welcome. So, here's the here's the philosophy of
201. 32:37this channel. The philosophy of this channel is just be in the AI conversation. Like Jim Ross who's here, that Jim Ross said,
202. 32:47"I'll be vibe coding while you're waxing poetic up there." Um, Jim is is one of what we call the irregulars. So, the irregulars are a
203. 32:56group of weirdos. They're not weird. They're irregular. But they're a group of people, human beings that come to this channel every single
204. 33:07night. It's nuts. But here's why they do it. They they may come a little bit for me, but what this is really about is about this community
205. 33:19connecting with one another, supporting one another, learning from one another, and just being in the conversation. So,
206. 33:25a lot of times someone will come in and they'll ask questions like someone just said, "What's vibe coding?" And so, now
207. 33:30we're going to run down a vibe coding rabbit hole and I'm going to actually build you an application before your
208. 33:35eyes and it should melt your face. And and what this channel's about is no matter where anyone is in the process of
209. 33:44figuring out this AI stuff, just come in here and be in the conversation. And sometimes the conversation's going to be
210. 33:50way over your head. And sometimes the conversation's going to be, you know, you're going to be like, I know more
211. 33:56than he does, and that's part of the part of the charm, right? And what it's about is just we're we're all learning
212. 34:04this right now. None of us know what it means. Nobody really knows more than anyone else right now because the
213. 34:10technology is changing so fast. We literally can't. like the way you made an AI movie a year ago and what was possible a year ago
214. 34:21versus how you make one today and what's possible today. It's it's just there it's just it's a completely different
215. 34:28set of tools. It's a completely different set of possibilities. So, welcome to anyone who's new and just jump in, sit back and if you're if
216. 34:36you're feeling feel free to ask questions whether you're on TikTok or YouTube. What's what's amazing about
217. 34:43this community about the irregulars? They're incredibly loving. They're incredibly generous. And so if you have
218. 34:49questions, and I don't get to them very often, they'll answer your questions and they'll say, "Hey, come join the AI
219. 34:54salon." And you know, you can ask questions over there and I'll help you out over there. So if you haven't done
220. 34:59that yet, you should do that. Join the AI salon. Go to go to the salon.ai and say join our community. And uh and you
221. 35:07you will see it's a it's a really amazing place. Okay. Gravity jump. My back end hurts. I can never get it to work. Okay. Vibe coded
222. 35:16two apps last week. Oh, that's great. Uh, Tai B, uh, a genealogy app and a martial arts career seeker app. That's
223. 35:23beautiful. It's amazing. Yeah, it's it's incredible. Okay, so what are we going to make tonight? Let's see. Let's do a um
224. 35:38Oh, I got one. Um, I want to create a fake um,
225. 35:54DNA animal designer using nano banana to generate images. The user
226. 36:10um picks up to three animals to combine in a visual interface.
227. 36:26Interface. Um what were those cards called? Do y'all remember those cards where you had like the head of an animal, the middle
228. 36:35of an animal, and the legs of an animal? Anybody remember what those were called? Come on. Some of you Gen Xers in here,
229. 36:46you got to have a couple of brain cells left cuz I don't. LSD. No. No, it's not LSD. [Music] Silver Fox. Say it again. The entire
230. 37:03animal. No, it was it was a card. It was a It was like they were flip books and it had three three panels and it there
231. 37:11was like the head panel, the middle of the animal, and then the feet. And you'd flip it. It'd be like the head of a
232. 37:16flamingo, the body of an elephant, and the legs of a chicken. Come on. I know what you're talking about, but I don't
233. 37:23know the name. Anyway, all right. A visual interface. The 60s were good to you. I actually remember those cards though.
234. 37:32Okay. A visual interface like the ones that had a head, body, and feet cards you would
235. 37:48mix up. But this is the modern digital version. It should look like a lab experiment
236. 38:05experiment where you select the three animals and what percentage of each
237. 38:20and then nano banana makes a photo of your new animal. That's cool, right? Okay.
238. 38:35So, we're effectively done. So, what vibe coding is, uh, so who was it? Who was it earlier on that that when they first got in here asked
239. 38:50not and it was Young Young Bungalow? I don't know if Young Bungalow is still here. Young Bungalow came in and when
240. 38:56when he first got in, he said, "What's your favorite or what lang programming language are you most proficient at?"
241. 39:02And I said, "English." And so, um, this is why because I can now just bumble bumble out a description of an
242. 39:13app and this thing will figure it out. So, look what it says. Okay. I'll create a DNA animal designer with a lab
243. 39:19experiment aesthetic using nanobanana image generation. This will have a visual card interface where you mix up
244. 39:26three animals with percentage controls. Design vision laboratory science theme. Visual animal selection cards. Um
245. 39:35percentage sliders for DNA mixing. Clean futuristic interface features. Select up to three animals. Right. And then so now
246. 39:43enable cloud. So I need to enable the cloud so that it can um use AI features from Gemini flipflap books or miximals or exquisite
247. 39:55corpse books. Cool. Beautiful. I like the percentage feature. Thank you. Thank you. Um still asking do you follow
248. 40:04follow Peter Diamont Diamontis? Um, I I don't follow him like I would say if there's people I follow right now, it's
249. 40:12Ethan Malik. Um, Paul Ritzer from the Marketing AI Institute. Um, Robert Scoville I pay attention to. Um, Peter, I'll I'll if if he shows up
250. 40:23in my feed, I'll pay attention to what he's talking about, but I don't necessarily follow him. Um, okay. The game you're likely thinking of
251. 40:32is often referred to as Animal Rummy. No, it wasn't Animal Rummy. No, it wasn't Animal Room. It's It's
252. 40:41flip-flop books, I think, is is what it was. Mad Libs has the word version. Picture picture consequence, maybe. All
253. 40:48right. Okay. So, and and you're like, Kyle, why are you talking? What's going on? Why is this not done? Because it's
254. 40:55coding. It's writing code right now. Um, you used to be able to watch it write code, but they don't let you do that
255. 41:01anymore because it's I think they've got an interface bug that they still haven't fixed. Um, but as soon as this writes all the
256. 41:08code, we're going to be able to play with this app or at least look at it. And then what vibe coding is, actually, while
257. 41:16that's building our app, let's go find the original definition of vibe coding. Vibe coding Karpathy. So, Andre Karpathy, who's one of the
258. 41:30founders of OpenAI, um, on February 2nd of this year, wrote this. There's a new kind of coding I call vibe coding where you fully give
259. 41:43into the vibes, embrace exponentials, and forget that the code even exists. So, if you're not a coder,
260. 41:52this makes sense to you, right? Because you're just like, I just want an app. like you don't really give a [ __ ] what
261. 41:58the code's doing if you're a coder, which Andre Karpathy is. He's probably one of the better coders, you know, on the planet. Um,
262. 42:08this is like blasphemy, right? So, so you know, there's a new kind of coding where you give in fully to the vibes,
263. 42:14embrace exponentials, and forget that the code even exists. It's possible because LLMs, eg cursor composer with
264. 42:20sonnet, are getting too good. I just talked to composer with super whisper. So, so speech to text. So, I barely
265. 42:28touch the keyboard. I ask for the dumbest things like decrease the padding on the sidebar by half because I'm too
266. 42:35lazy to find it. I accept all always. So, so when it makes changes and it says, do you want me do do you want to
267. 42:41accept these changes? He just says accept all. Um, I don't read the diffs anymore. So, diffs are like what
268. 42:48changed? Here's the old code. Here's the new code. And engineers will read those differences to understand what it what
269. 42:55changes it made. When I get error messages, I just copy and paste them with no comment. Oh, am I not sharing
270. 43:00this? I'm not. Um, when I just uh when I when I get error messages, I just copy and paste them with no comment. That
271. 43:09usually fixes it. The code grows beyond my usual comprehension. I'd have to really read through it for a while.
272. 43:15Sometimes LLMs can't fix a bug, so I just work around it or ask for random changes until it goes away. It's not too
273. 43:23bad for throwaway weekend projects, but but still quite amusing. I'm building a project or web app, but it's it's not
274. 43:29really coding. I just see stuff, say stuff, run stuff, copy and paste stuff, and it mostly works. So that's the definition of it. Okay. So
275. 43:42we now have Dr. J exquisite corpse books. Those were the the animals. Cool. Very cool. Can you explain the token system with AI,
276. 43:53please? Yeah, I absolutely can. Give me one second. Okay. DNA slot one, DNA slot two, DNA slot three. So, we're going to
277. 44:02put Oh, these need to be drag and dropped, but that's okay. We're going to do Oh, this is cool. So, so we've got
278. 44:09we've got So, this app, it looks like it's going to be functional right out of the gate. So, we're going to do a dog, a dragon, and [Music]
279. 44:23a shark. Okay. So, so it's done the math for us. So, notice that it's done the math. Oh, yeah. Is there a turtle in there?
280. 44:32There's not. So, one of the things I need to do is I need to be able to say, um, we need to be able to add our own
281. 44:40animals. Enable nano banana. I think that's already done. Okay. So, let's let's make this thing like 50% shark,
282. 44:5644% dragon. Oh, it doesn't balance them equally. All right. All right. So, we're going to go 45 dragon, 41 shark, 15 dog. Generate
283. 45:14hybrid creature. Now, now this should this should So, already already. Okay, here's our hybrid creature. Look at that. Are you
284. 45:28[ __ ] kidding me? at the water. Shark teeth, dog body, dragon tail, dragon fins, things to the percentages. One prompt.
285. 45:43David Simon, what's happening? One prompt. And now we can we can go fix this, right? You could go say, I don't want I
286. 45:52don't want to use emojis for the animals. I want, you know, more animals. I want to be able to add my animals. I
287. 45:58want people to be able to create accounts. That's [ __ ] sick. Lion. Wait, how do I How do I restart? I Oh, I can't
288. 46:08restart. We got to be able to restart, right? A hybrid creature. I could add sharing in here. I want to
289. 46:16add social media sharing so I can share that that thing, you know, out to the world. Come on, man. You kidding me?
290. 46:27Okay, let's do one that's like 70% dog. We'll we'll we're going to lower the shark. We're going to lower shark to
291. 46:36six. Oh, wait. I can get rid of one of these. Ah, okay. Let's add in eagle. Eagle's good. All right, we'll go 33 33
292. 46:4733. This is dragon, eagle, and dog. [Laughter] [Music]
293. 47:05How cool is that? Come on, people. Look at that. You kidding me?
294. 47:19We should probably mention credit limits for storage. $1 AI out of the box on the free plan. I don't Can you write that in English? I
295. 47:33don't quite understand what you're saying, Brandon. That is a cool dragon dog bird. Every generation costs Oh, every
296. 47:45generation costs credits. Yeah, yeah, yeah. And Lovable gives you a dollar to start. Okay. Yeah. So, you can do this in Gemini AI Studio.
297. 47:56Um, but anyway, so so that's vibe coding and you can vibe code video games. You could vibe code like a a I I don't know an app for
298. 48:08managing your soft softball team like like let people log in and and let you know their availability and and it'll
299. 48:16you know create schedules for you like literally any app idea that you have you can just make it here. Now Lovable is
300. 48:24web apps only right? So these are apps that are going to live on the web. If you want to make phone apps there's
301. 48:29different vibe coding tools for those. that gets more complicated. But lovable if you just want to if you've never
302. 48:35coded before in your life and you want to go from a standing start, I've never made an app before to what I just did
303. 48:44here. Right? One prompt sort of poorly articulated. Boom. Functional app within I don't know what was that 10 minutes.
304. 48:55Imagine spending four hours with this refining it and doing all that sort of stuff. All right. So, if someone start
305. 49:02uses the app that you create, you pay for every generation. Yes, you do, Rick McCaulay. So, one of the things that
306. 49:08Lovable does not have in um in the cloud right now is the ability to to put a payw wall in front of it. So,
307. 49:20like lots of apps will will let you do a generation or two and then they're like, "Give us five bucks a month to keep
308. 49:25going or you know, buy 10 bucks to buy credits." Um, you could certainly do that with this. Um, Brandon last week, I
309. 49:33was whining about that and Brandon last week spun up a a he basically just said to Lovable, "I want to be able to charge
310. 49:41for this and tied it to his what' you tie it to, Stripe or or uh Venmo?" Yeah, he just tied it to his Venmo account.
311. 49:50So, you can have Lovable do that stuff. It's just not built into it right now. So, um, Kyle, Sora just turned your
312. 49:59comment into a cool app idea. Posted in irregulars. Sora just turned your comment into a cool app idea. The hell does that
313. 50:08mean? Where'd you put it? Irregulars.
314. 50:25I made an app with Lovable that transcribes Tik Tok videos. Oh, cool. Oh my god. Tik Tok's interface for when I try to expand a comment.
315. 50:43Check the chat. I do not see it. I do not know of what you expect.
316. 50:58Who's that? Suzanne. Suzanne speaking at the create conference. Very cool. Oh, look. Here's Corey Sandler's images
317. 51:11for the uh for the song I made. Very cool. It's in share cool things irregulars now. Share cool things.
318. 51:30[Music] Where is share cool things? Oh, just refresh.
319. 51:46No, I do not know what you specu
320. 52:04[Music] sneaking into class late. Oh, Mary Mary, thank you.
321. 52:22Kyle said his f vibecoded app would melt your face off.
322. 52:38Wait, hang on. Let me show this tab. app that I call Melt Your Face Off because when you see it, it kind of
323. 52:45does. It reads your mood, spins up a generative filter, and turns you into living liquid color. I'm in the AI
324. 52:49learning lab and I just finished something I'm pretty excited about. It's a vibe coded app that I call Melt Your
325. 52:53Face Off because when you see it, it kind of does. It reads your mood, spins up a generative filter, and turns you into living
326. 52:58>> That's hilarious. Uh, okay. I thought you meant Sora. The OpenAI team that built Sora did something with my app.
327. 53:06No, you made a Sora video of me vibe coding an app. Got it now. Good lord. Miro encroaching on Tik Tok cam. Oh. Oh
328. 53:20my. Okay. Let's see what's going on. Oh, there we go. Okay. All right. Oh, yeah. They're they're they're
329. 53:32building lovable in Demiro right now. All right. What was I going to even talk about tonight? We got off on tangents,
330. 53:38which is super cool. Like one of the things I love on this channel is if people come with ideas, thoughts, questions, things like that. Um,
331. 53:47what do you want to play with? What do you want to talk about? Um, I'm happy to I'm happy to dig in and
332. 53:55show some stuff. The Melt My Face app. You're going to talk about black bars. Shut up, Pin. Let's do a code of arms family crest.
333. 54:09That's pretty good. Oh, tokens, Coach Melissa. Okay, tokens. All right, so you're going to get the box metaphor. I
334. 54:15haven't done the box metaphor in a long, long time. Okay, [Music]
335. 54:31where we'll use playing card boxes. and a cassette tape. Okay, we'll come back to those. Wait, we'll use a beat speaker.
336. 54:48Yeah, there you go. Okay. Okay, tokens. Everyone's talking about tokens. In fact, the OpenAI developer
337. 54:58conference, they just gave out awards to developers who have processed a 100red billion tokens or a trillion tokens and
338. 55:08they got plaques. You've processed a trillion tokens. What What is this token madness? Okay, I'm going to go back a little bit in
339. 55:20history. I won't be there that long. In 2017, Google, some engineers at Google write this paper called Attention Is All You Need,
340. 55:31and they invent. It's not the same without the printer inking bottles. Oh, here are the printer inking bottles.
341. 55:36Fine. Okay. I got I got two. We're good. We're fine. Okay. I couldn't find them before. I've I've started to clean up my
342. 55:43desk area. And you know how it is when you have ADD and you try to clean something and then you can't find
343. 55:47anything because you knew where the piles were. Okay, that's not what we're talking about. Tokens. 2017, Google invents this thing called
344. 55:56the transformer. And the transformer is an architecture that is the engine behind almost all of the generative AI things we play with
345. 56:08today. All the large language models, all the image models, the the video models, all of them are based on this
346. 56:15transformer architecture. There's some other um technologies layered on top for images and videos called diffusion
347. 56:22models and and there's some different things in there but but the underlying model is is this transformer model.
348. 56:32What happens is they take a bunch of data. So, so let's just say they take documents and they take documents like
349. 56:39here's a book and they describe what the book is and they put the the book into this model and then they take a blog
350. 56:45post and they they label that and say this is a blog post and maybe here's what the topics are and there's all
351. 56:51sorts of fancy ways to do this but basically they go out and they get as much data as they can and in fact one of
352. 56:59the one of the reasons AI has such a bad reputation right now is they went out and they got all the information. Well,
353. 57:07what's included in all the information is a bunch of copywritten stuff. Okay. So, so they absolutely trained these
354. 57:16models on copywritten stuff. They have done that. But when you understand what actually happens to them and the court
355. 57:23actually agrees with this. The court calls when you embed um something like a book through the transformer model. It is a
356. 57:31transformative act and because of that they consider it fair use. But you can't steal the book to do that. You have to
357. 57:39buy the book to do that. But anyway, that's neither here nor there. Okay. What happens when they take a
358. 57:50book? We'll just we'll just stick with a single book. They take a single book and the transformer model fragments, you
359. 57:58know, splinters, shards all of the contents of that book into what are called tokens. So tokens are either words, periods, spaces, commas,
360. 58:11um parts of words. So you might have running. So you might have run and ing running or run and ning two ends. Um
361. 58:20those might be two tokens that make up one word. So in general when you have a thousand tokens you it
362. 58:28equates to about 750 words. So when you hear people talk about a large language model having um a c certain token
363. 58:38context window and they say oh this has got a 100,000 context uh token context window that's about 75,000 words which basically means
364. 58:49in its in its memory it can basically keep a little more than a a complete novel in its thought before it starts
365. 58:57forgetting things. So, so tokens are fragments of words. But, but what's the box? The box is they take all of the documents that they
366. 59:08can gather, like all of the documents that essentially represent the output of humanity for all of our history that's
367. 59:15been stuck on the internet for the past 80 years and then probably more than that. And they run it through this
368. 59:22machine. And what it does is it takes the original document and it shatters it into all these tokens and then these
369. 59:29tokens are clustered in semantic in thousand dimensional I know this sounds like I'm making this up and it's
370. 59:37[ __ ] It's not [ __ ] This how it works. It's thousanddimensional mathematical space where the tokens are
371. 59:44grouped in semantic clouds. And and here's what a semantic cloud means. Semantic cloud is you could have the
372. 59:52word dog, right? The word dog is the word dog. Dog g. And you could say, "Oh, well, you're just going to have that's
373. 59:57going to be in there once." No, it's not going to be in there once because dog in one context is the fluffy thing that you
374. 1:00:04pet when you get home at night. And dog in another context is he did his girlfriend wrong and he's a dog. So
375. 1:00:12those two tokens of dog actually get put in different semantic clouds. So you have this massively complicated
376. 1:00:20mathematical space that has all these tokens in it. And the way the transformer model works, this is absolutely mindblowing that somebody
377. 1:00:28figured this out. But because they've figured it out, we get to benefit from it. When you type in a prompt into chat
378. 1:00:34GPT, what it's actually doing is it's generating a probability. It's generating a probability of a single token in what they call the
379. 1:00:45latent space. This thousand-dimensional mathematical semantic cloud. What you type in creates a probability that some
380. 1:00:54token in that massive information cloud is the most likely next probable token to answer your question and then the
381. 1:01:04next one and then the next one and then the next one. So when you watch chat GPT write you a whole document or you watch
382. 1:01:10lovable code for you, it's all just pulling these mathematical tokens out of this mathematical space. So so you have basically all of human
383. 1:01:21knowledge gets jammed into a box and this is what's called a large language model, right? And so you have like GPT5
384. 1:01:33is is one of these or GPT5 is actually multiple of these that are kind of glued together. And then chat GPT is actually an app
385. 1:01:43that sits on top of the large language model. And this would be like you know like like um uh Microsoft Copilot,
386. 1:01:52right? So Microsoft Copilot uses the same model GPT5 as chat GPT but these are completely different experiences
387. 1:02:01right this one's bloated and unusable and chat GPT is actually good right because this application is better both
388. 1:02:08of them when you type a prompt in here through an API it calls down into this giant mathematical space and says give
389. 1:02:16me an answer back and so what it displays on screen is coming out of this what's called a corpus of all of the
390. 1:02:23knowledge of humanity or as much as they could get in it. Um, and that's the B box metaphor and that's what tokens are.
391. 1:02:30It's it's absolutely insane. So, the here's what's here's why I think it's important if you're if you're going to
392. 1:02:38play in the AI game and especially if you want to get good at this and be a be a professional about this. When someone says chat GPT is the
393. 1:02:46world's greatest plagiarism machine, it's not because it's not the way the thing works. What what people are assuming who
394. 1:02:57don't don't understand how AI works is that they're assuming that that chat GPT is somehow reading a document and
395. 1:03:05copying like whole paragraphs out of it and and you know just copy and pasting, right? That would be plagiarism. It's
396. 1:03:13not doing that. The G in GPT, GPT stands for generative pre-trained transformer. So the T for transformer, that's the
397. 1:03:21technology. Pre-trained means they went out and they found all the data for us. We don't have to bring our own data like
398. 1:03:27you used to have to in the olden timey AI days like four years ago. The G stands for generative. It's generating new content every time.
399. 1:03:38So actually there was a comment earlier about every time I make a change it completely rewrites my thing. That's
400. 1:03:44because of the nature of how the transformer works. It it's generating it new each time. Now you can prompt it
401. 1:03:51very specifically to generate something that looks just like the original, right? And that's where people can prove, oh, this was trained on
402. 1:04:02copywritten works, right? But if you look at what's actually happening, it is a transformative act that shatters the
403. 1:04:08original documents into these tokens. Cool. Make sense? It's absolutely [ __ ] insane. It's like the more you learn about this, the
404. 1:04:19more I learn about this, the more blown away I am that it does what it does. It's crazy. It's crazy because
405. 1:04:34In a weird way, it's not actually understanding what it's doing, but it's behaving incredibly like it's understanding what
406. 1:04:44it's doing, right? It's it's behaving a lot more like a human than people want to give it credit for. But it really is
407. 1:04:53just math. It really is just a probability calculator. But but a lot of people will say, "Oh, it's just a stochcastic parrot. it's just a
408. 1:05:01probability calculator. It can't have feelings. It can't it can't be creative. It can't be this. It can't be that. But
409. 1:05:08none of that's actually true. Like it it can actually be quite creative. And it can be quite creative in a way very
410. 1:05:14similar to how our brain works because neural networks, the hardware underpinning of all this stuff, were
411. 1:05:21designed to mimic how our brains work. So anyway, Tik Tok pin or it's just consciousness twice. Yeah. I don't know. I mean, the whole
412. 1:05:31thing about is it sentient, is it conscious, that that whole thing for me, that's a semantic argument. It's an academic argument.
413. 1:05:44If we as humans interact with this thing like it's sentient, then it is. Then it is. Whether or not it actually
414. 1:05:53is to me is a is an academic definition. But but for people who are not digging into the geeky side of AI and they're
415. 1:06:03just talking with with chat GPT advanced voice like an old person, you know, alone at home where their kid gives them access
416. 1:06:15to chat GPT advanced voice and they can have a conversation all afternoon. To them, that's that's every bit a
417. 1:06:24relationship as having a person sitting there, right? So for them, they're talking to someone and they probably
418. 1:06:32have a relationship with that person. In fact, when chat GPT went from GPT40 to GPT5, a bunch of people got their
419. 1:06:40panties in a twist because they're like, "You took away my girlfriend. You took away my boyfriend. Their personality's
420. 1:06:47gone. Give me back my boyfriend." My CPU is a neural net processor, a learning computer governor Arnold Schwarzenegger.
421. 1:07:00My computer, my GPU is a neural network. And then there's emergent properties. Rick McCaulay just said that. So
422. 1:07:10emergent properties, these these this is an amazing thing. So when you embed all of this information, so so basically you categorize a bunch
423. 1:07:23of information. So So you have human beings that like watch movies and read books and read papers and read blog
424. 1:07:29posts and they they label them and categorize them and they basically do all that stuff to be able to embed them
425. 1:07:35correctly. And so they train these models on certain things. One of the things they they never trained models on
426. 1:07:44was to be able to do translations. Um, but they just embedded them with enough data that that they just
427. 1:07:51realized, oh, these things are actually really good at translations. And there's those are called emergent behaviors
428. 1:07:56where it wasn't specifically trained to do something, but it's just good at it. Um, computer program, excuse me,
429. 1:08:03programming was was one where they trained it on programming and it got good at linguistic things because language is structural and code is
430. 1:08:12structural and so it could sort of figure out, you know, um, things that they didn't think it would be able to
431. 1:08:19figure out. It's absolutely crazy. All right. I had a message come by by on chat GPT that I was online for a long time and I
432. 1:08:29should take a break. Well, so so chat GBT was probably right. Rod on deck AI studio refresh. Okay, cool. Um,
433. 1:08:44I want to show you something. Um, I don't have really anything to show you in comparison to it, but let me let me run over and
434. 1:08:54and show you something here. So, this is Google's playground or Google's they call it AI studio. So if you go to AI studio.google.com
435. 1:09:08they've done a completely new interface here. Um so you can chat with models in the playground. So you can chat with AI
436. 1:09:16models. You can vibe code in in the AI studio and it's actually really good. The the vibe coding here is really good.
437. 1:09:24You can look at monitor usage of your dashboard. This is a what this site is aistudio.google.com. This is for developers.
438. 1:09:36Um, so if you open up the little side panel here, um, they've got like documentation. You can get an API key.
439. 1:09:45Um, and and they've got lots of choices. This site used to be very very cluttered and technical and they didn't explain anything.
440. 1:09:57and it was up to you to figure out which model had which capabilities in it and you had to read documentation and things
441. 1:10:04like that. And so the fact that that it is now kind of this visual front end, I'm sure that engineers this probably drives them
442. 1:10:14[ __ ] baddy. They're like, we don't need any stupid pictures. They're used to command line interfaces, right? Um
443. 1:10:24this is not for them. What what's happening is this. As the AI models are getting more and more and more capable, people that were
444. 1:10:34historically not coders are now coding. So what's happening is people are starting out vibe coding and then
445. 1:10:40they're realizing, oh [ __ ] I got to set up authentication or I need to set up a database or I want to do, you know, some
446. 1:10:46sort of commerce backend or I want to put this on a different website. And so they get chatpt to start teaching them
447. 1:10:52things or Gemini to start teaching them things and they're like okay so if you want to do that you got to go in here
448. 1:10:57and then you got to put your code in a GitHub repository and then you got to push from there and you got to do a pull request.
449. 1:11:05Right? So you're starting to get people who were not trained as engineers entering engineering spaces. And so what
450. 1:11:14this represents to me is this is Google recognizing that they've got audience in here that's not just engineers. That mashing
451. 1:11:25of expertise is going to happen in every single industry. Every single industry. It's all going to get mashed up.
452. 1:11:37What it means for us as individuals is that where we were weak, where we've where we have gaps, like for me, I've got ADD. One of the
453. 1:11:50reasons I can't code is I don't give a [ __ ] I like I love having the idea. The minute I have to turn the idea into,
454. 1:11:58okay, now you're going to have to put this in a sequence of things in a framework with I can understand the
455. 1:12:03frameworks, but the minute I actually have to go in there and look at every [ __ ] line of 27,000 lines of code, I
456. 1:12:10would rather jump through a window, right? So, my brain just doesn't work that way. But now, because of AI, I can take that
457. 1:12:19gap in my skills and in just even the way my brain works and I can fill that in. So now I can just take this idea I
458. 1:12:26had for an app, put it in one of these vibe coding things. It fills in that gap for me. And now I can look in an app that I built.
459. 1:12:35And engineers will tell you, "You didn't build that. You just prompted that." Yeah, I did. But that was my app idea.
460. 1:12:44And I'm looking at it now. And I'm now looking at it going, "Is that as good as I want it to be?" No. I want this button
461. 1:12:51over here, and I want that over there. And I want the label to be different. I want the colors to be different. And so
462. 1:12:56I go vibe code that and it does that. And then it does all that. And I start playing with it. And I have my animal DNA maker thingy.
463. 1:13:06And now I'm like, huh, that's kind of cool. You know what that would be cool as? That would be cool as like a a
464. 1:13:13children's story book. And then I go over here and I can make a children's story book. And I'm like, huh? You know
465. 1:13:19what that would be cool? That'd be cool as a song. And now I can go over here and I can make it a song. Then I can
466. 1:13:23make it a movie. Then I can make it a business presentation. Then I can launch a business. Every single person
467. 1:13:32who chooses to use AI and who who chooses to actually understand what it makes possible can now just live in idea space
468. 1:13:44and and think like a producer like, okay, I got an idea here. It's going to be big. Great. what you needed to do.
469. 1:13:53And then the better you get at at using all these disperate tools, the more gaps you fill in on yourself, the better you
470. 1:14:00get at chat GPT, the better you get at vibe coding, the better you get at making images, the better you get at
471. 1:14:06duct taping all these things together. and what's happening. One of the reasons, especially for people that are new here tonight, one of
472. 1:14:15the reasons people coming back, keep people keep coming back to this channel night after night after night
473. 1:14:24is that we get to collectively watch the evolution of these tools where we know how hard it was to get something usable
474. 1:14:32three years ago and then we knew it got a little bit easier two years ago and a little bit easier a year ago. And now
475. 1:14:38we're kind of marveling at, wait a minute, I can just go to Sora and I can make a video with me and Sam Alman
476. 1:14:45without having to actually create a Laura model and install open-source software. And yeah, no, you can just do that now.
477. 1:14:56And so that's going to continue to evolve. And then what happens is all of the tactical execution and especially the tactical execution
478. 1:15:09that doesn't require a lot of thinking. It's like that blocking and tackling of I've got to write out the code. I got to
479. 1:15:15block it out first. I've got to write out the outline of the presentation. I've got to write out the story. I've
480. 1:15:21got to sketch the images. All that blocking and tackling work that takes brain cycles but isn't really creative,
481. 1:15:29that's just going to be done. It's going to be done for us and all we're going to have to worry about is the idea and the
482. 1:15:38fidelity of the idea and our taste. Holy [ __ ] how cool is that? You know, Sununo Studios awesome. At first I
483. 1:15:49thought, is this AI? No way. This this was my idea. I just use tools to make it reality. You know, it's funny. I I I made a
484. 1:15:59comment today here. Let me I want to show you something. I want to show you all something
485. 1:16:16Tik Tok pin Mary. Yeah. You don't have to you don't have to build a cell phone to be able to use one. Exactly. Right. I mean, RadioShack
486. 1:16:27was built on the idea that everyone wants to build a cell phone. No, some people just want to use cell phones,
487. 1:16:33right? We don't always want to build them. And that's one of the one of the challenges of AI right now is that
488. 1:16:39because there's, you know, a 50-year legacy of of classic AI, classical AI that really did require you to be an
489. 1:16:46engineer and a mathematician and things like that, people still think that it's got that requirement that you've got to
490. 1:16:53be able to build a cell phone. No, you don't. We're like chat GPT marks the turning point where you can just use
491. 1:17:00cell phones now, right? That's that's that's why this is different. That's why this channel exists. But I want to I
492. 1:17:06want to I want to share something here. Um so John Leland um who's a a salon member and he's he's got his creativity work
493. 1:17:15that he does. Um he he uh posted about this this video that this guy made. And so let's let me share this tab. So he
494. 1:17:26posted So here's this video that this guy Isaac Rodriguez made. Hello.
495. 1:17:41[Music] [Music] Huh?
496. 1:17:56[Music] Hey, [Music]
497. 1:18:18heat. Hey. [Music] [Applause] [Music] Okay. So, we're going to stop it right there. Um,
498. 1:18:36so this is a a a series, a video series that this guy's putting out there. And he said something here. Introducing
499. 1:18:45The Valiant Night, episode 1, a fun short series created 100% with AI. Um, I took issue with this. What he's talking about here is that the
500. 1:19:01images that you see here and the animations and the audio and the music were created with AI tools. I would absolutely argue this was not
501. 1:19:11created 100% with AI. My comment was, let's see where I put it. It might be on John's post.
502. 1:19:27Um, my comment was, "This is great. However, I would argue this is not 100% created with AI. Isaac had the idea, a
503. 1:19:35vision, learned how to prompt, discern which tools could deliver the quality he was looking for, created the script,
504. 1:19:42even if it was AI assisted. He ultimately chose what was being shown and said, and how." He generated 10
505. 1:19:49times or more the number of clips he didn't use. He curated the best ones out. He edited them into a final piece
506. 1:19:56and he package packaged it up and shared it with the world. AI tools are enabling tools in the stack. But this is a human
507. 1:20:04creation. I look forward to the day when we don't have to declare the tools we use. Important disclosure. I wrote this
508. 1:20:12comment on an Intel Mac with an Apple extended keyboard and the Chrome browser. As stupid as that little last sentence
509. 1:20:21says, when when people like, "Was that AI? Did you use AI for that?" It's none of your [ __ ] business. Quite frankly,
510. 1:20:29what tools I use in my tool stack. If I had an idea for something and I brought it to life and I put it in the world,
511. 1:20:36you can judge it in one of two ways. You like it or you don't. If I'm happy with it as a creator, I created it. It's
512. 1:20:44independent of the tools you used. So, I would encourage you as you start thinking about these AI tools, don't
513. 1:20:51give them more power than they already have. If you say that this was 100% created, Isaac just discounted his contribution to this entirely.
514. 1:21:04And I [ __ ] hate that. Like, this is a wellput together video. It's a It's a It's a strong well-put together video.
515. 1:21:15And that's not because of AI. That's because of his taste, his storytelling. Right. So anyway, all right. Enough [ __ ] ranting.
516. 1:21:29I'm losing my voice. What you saying should be on a t-shirt. I don't quite know what I said that could be on a t-shirt. That's a good
517. 1:21:39rant. Thank you very much, Joe. Creative AI coach. Yeah, like like I did I did another post to this today. Oh, here
518. 1:21:46we'll we'll do this. Okay, I'm gonna let me show you another LinkedIn post today. I'm I'm on a rant right now. I'm on a
519. 1:21:54I'm on a [ __ ] bender about this cuz it's driving me [ __ ] crazy. Was that AI? Did you make that with AI?
520. 1:22:05Did you ever use Photoshop, you big [ __ ] talentless loser? That's exactly what they're saying. Cameras don't create a great photograph.
521. 1:22:16Yeah, cameras are a great example. Source Cam, you know, you know there's things in the world called snapshots,
522. 1:22:22right? And then there's things called photographs, and then there's things called art. What is the difference between them?
523. 1:22:32You could take the exact same camera and take a snapshot with it, a photograph with it, or art with it. What's the difference?
524. 1:22:41What's the difference? The person holding the camera, what they look at, where they focus, when they push the button, how they edited
525. 1:22:49afterwards, what their intent was. Right? A snapshot is just, yeah, here's a picture. We are on vacation. This is
526. 1:22:56cool. A photograph is I actually looked at that. Art is I took my time. I have a point of view. I'm not going to let it
527. 1:23:02go out to the world until I am excited about it. And then it's going to do what it's going to do. AI is the exact same [ __ ] thing. Um,
528. 1:23:15let's see. Where's the boundary? So, this is a prompt I did today or post I did today. So, I'm going to play the song first.
529. 1:23:31So, this is a song I made last week here in the AI learning lab. was asleep. [Music]
530. 1:23:47The ferris wheels asleep, its arms still folded tight. We walk between the shadows and the strings of halfway
531. 1:23:55light. The popcorn stands are dreaming of the laughter they'll be fed and your reflection in a puddles turning pink instead of red.
532. 1:24:10You say it's strange to see it empty. I say it's strange to feel this calm. You smile like you know something I
533. 1:24:24don't know. I've already begun. Before the lights come on, before the crowd arrives. Before we name what this
534. 1:24:36could be, before the day survives, we linger in the middle where the maybe feels like song. Two hearts not yet decided
535. 1:24:48before the lights come on. [Music] A prize bear in the windows. Got a rip along the seam. You laugh and say he's
536. 1:25:05just like me, stitched up from a dream. The smell of rain and sawust makes the air taste almost sweet. I count your
537. 1:25:14footsteps next to mine. They nearly match the beat. You toss a coin. It spins between the quiet and the breeze.
538. 1:25:23It lands on something fragile, something trembling between. Before the lights come on, before the noise begins, before the world demands,
539. 1:25:36we choose where all this story ends. We're written in the margin where the ink still barely strong. Two souls half in the promise
540. 1:25:49before the lights come on. [Music] Cotton candy ghost. The whisper of a
541. 1:26:04secret that your hand could hold but hasn't yet tried. If I breathe too loud, I'll break the spell we've spun.
542. 1:26:15A heartbeat before gravity decides which way we run. Before the lights come on, before the sky turns gold. before our voices learn
543. 1:26:29the lines they're scared to have told. Let's walk this midway silence till the dawn has found its song. And maybe we'll decide it
544. 1:26:45before the lights come on. [Music]
545. 1:27:04Okay. So, beautiful, right? So, here's my question. Where's the boundary? Is this AI or human? It's it's a song made
546. 1:27:14in Suno. That video's got a Puno logo on it. So, that's an AI song, right? And so, if someone goes, "Oh, did you use AI
547. 1:27:21for that?" What are they doing right there? They're completely diminishing it. They're assuming that how I made it was this.
548. 1:27:30>> I pushed a button and out came a song. So, what I wrote was this. There's a lot of talk of the critical importance of
549. 1:27:36distinguishing between AI generated or human generated. Listen to the song I wrote or didn't, depending on where we draw the line.
550. 1:27:45The original four bar guitar lick. That was me. The idea for the song, the idea of two people deciding if they're more
551. 1:27:53than friends, walking down a dirt road at a carnival before the people show up. My idea, the prompt that I put into chat
552. 1:27:59GBT to articulate that that idea. My idea, the lyrics, AI. And normally with AI lyrics, I'll edit the [ __ ] out of them because they're
553. 1:28:09usually bad or I'll mix things together or whatever. The lyrics were so good, I didn't have to edit them, right? the music prompt.
554. 1:28:19Um, Chat GPT wrote the initial one and then I did nine different variations of the prompt trying to get the song I
555. 1:28:25wanted. So, that was a combination of AI and me. The initial song generation, the first song generation was AI. Um,
556. 1:28:36the the um curation of creative input. So, we had people from the live giving me ideas. So, I'm sort of curating
557. 1:28:43inputs from them. I'm having ideas of my own. We're trying to get the song better. I generated 21 versions of that
558. 1:28:51song that I didn't like, that didn't sound like the vision. The first one sounded like what I what I imagined, but
559. 1:28:59there was weird things with it. And then we tried to make it was it was a fascinating process over over 22
560. 1:29:05versions. We started really good and then as I tried to improve it, the song got worse and worse and worse. And then
561. 1:29:13Corey Sandler on the channel said, "Oh, why don't you try this? Why don't you try that? Why don't you put in cellos
562. 1:29:18with I don't know, piccolo or whatever the [ __ ] it was, an obo." And and it started getting better and better and
563. 1:29:24better. And then the one that you just listened to was the 20 second song. And if as I listen to that song now,
564. 1:29:31there are no misprononunciations of words. There's no weird chord structures. It's just a good song. Right. So 21 bad versions. Um, uh, you
565. 1:29:41know, the songs were fine, but but I just didn't like them. That was me choosing the 22nd song as the one. That
566. 1:29:50was me. The album art was Corey Sandler had the idea. She prompted it into an image. I chose to use it. And then
567. 1:30:00choosing to share that song with the world. That was me. So, in that stack of creative choices, there's like three or four things in
568. 1:30:09there that are AI. They're important things, but the bulk of the work, what took 90 minutes was all the other [ __ ]
569. 1:30:21You can't tell me that's not a human creation. I don't accept it. I don't. And I'll [ __ ] I'll [ __ ] argue this one up
570. 1:30:30one side and down the other. If you use Grammarly, does that mean you don't write or AI wrote your thing? So anyway, yeah,
571. 1:30:44the perception is AI, you push a button and out out squirts slop. And to be fair, you can treat AI that way, right?
572. 1:30:51That's why it has a bad reputation. But everyone that I know and certainly the people on this [ __ ] channel spend a
573. 1:30:58lot of time trying to get AI to bend toward their vision. That's human creation. So anyway, so my last question
574. 1:31:06is, so is this an AI song or just a song? And does it really matter? You know, my opinion is if I put that song
575. 1:31:14out and you don't know it's AI and you like that song and it moves you. Someone just said on the on TikTok over here
576. 1:31:19they downloaded it to their iPhone because they want to listen to it. They're not listening to it because it's
577. 1:31:24an AI song. They're listening to it because they like the music, right? Anyway, all right. Tyade Tuesday. It's a song. Exactly.
578. 1:31:40Um, such a beautiful song. Can I show my AI class I teach? Absolutely. They're middle schoolers. It's a be It's
579. 1:31:50actually a middle school song. It is absolutely a middle school song. Like I imagine this is like a 14 like a
580. 1:31:5814year-old girl and a 15year-old boy and they've been friends since elementary school and they sneak into the carnival
581. 1:32:06before it opens and it rained last night and they're walking through the carnival and that was like when I first played the guitar lick. Um,
582. 1:32:23wait. [Music]
583. 1:32:38Um, >> I didn't have I didn't have the dog howling when I did that, but like that sounded to me like this innocent kind of carnival,
584. 1:32:49you know, when you're a kid and you're walking through the carnival and it's kind of innocent and it's like this girl
585. 1:32:55and boy trying to figure out if they're more than friends. And there's I mean there's just some beautiful lines in it.
586. 1:33:01There's there's these beautiful lines that I don't know I whatever there's there's there's the the the the lyrics
587. 1:33:09in it are just beautiful. So yes, absolutely. Show your middle schoolers. It's a it's a it it's it's that spirit. So um
588. 1:33:21the genre was you such a beautiful song. Yeah. Cool. Story of my life. All right. Let's see. And their hands are one apart as they
589. 1:33:31walk at first. Yeah. Anyway. All right. It's 9:30. Any other questions? Any other thoughts? I know Johnny D had a question about the
590. 1:33:44prompt. How do you get it to not change everything? I mean, one way is throw it into a throw it into a canvas so you can just edit
591. 1:33:51parts of it. That's that's a way to do it.
592. 1:34:06Yeah, you you diminish 50 years of my life by clicking at that moment with with a camera. Yeah, exactly.
593. 1:34:18Um the pool of blood turned pink. Yeah. Yeah, it was it was like your Well, I don't think it was a pool of blood. I
594. 1:34:24didn't see it that way. I saw it as um the sun's coming up, so it's going from deep red to pink and she sees the
595. 1:34:33reflection of his face in a puddle and it's going from from red to pink. Tik Tok question, how do you mitigate AI
596. 1:34:40push back? It's affecting my ability to implement tech. Um that's a I the short answer is I don't know. Um,
597. 1:35:02what I what I can tell you, so what I try to do here when when when when people come in here, you know, leaning heavy on the hate, and
598. 1:35:13I've gotten my ass handed to me on LinkedIn a couple of times where I did a post and, you know, the haters the
599. 1:35:19haters came at me. Um, every in history in the history of humanity, every single time there's a transformative
600. 1:35:27technology, every single time, every single time, people resist it and sometimes dramatically. I have a feeling
601. 1:35:37AI is going to be one of those that gets resistance dramatically. What I've experienced is that most of the people that hate it, hate it,
602. 1:35:48have very strong opinions about it, have never tried it. They're listening to the tropes. They're listening to, okay, they they stole they
603. 1:35:57they stole from artists. Okay, it's bad. Therefore, it's 100% bad 1,000% of the time. I'll never use it. It's evil.
604. 1:36:04Right? So, they they throw it into this categorical evil category. We've got 50 years of Hollywood telling us the robots
605. 1:36:13are going to kill us. We've got all of those tech bros in San Francisco that are building this [ __ ] going, "It is
606. 1:36:20going to take your jobs." And some of them are [ __ ] and they're like, "I can't wait for it to take all the jobs." Um,
607. 1:36:27that's [ __ ] terrifying, right? This is people's livelihood. When you see something like Sunno, there's a really amazing video the
608. 1:36:37series of Tik Tok videos of this of this musician who takes his songs and puts them into Sunno and turns them into full
609. 1:36:45production songs and he kind of does it in real time and he does it over a series of days and he is at the same time mortified and inspired
610. 1:36:57and he's also smart enough to know that this stuff's not going away so that he's got to find a way to to get on the other
611. 1:37:02side of the mort mortified part of himself like holy [ __ ] my my career is over is is one of his reactions. The
612. 1:37:10other one is holy [ __ ] this technology is really cool and amazing and and it's not going away. So he's one of the rare people that can
613. 1:37:19actually hold those two emotions at the same time. And I I feel like that's actually a a fairly sophisticated
614. 1:37:30place to be that if you're using AI, it it it's some of this [ __ ] is absolutely [ __ ] terrifying. Like what what power it's got. But then
615. 1:37:43you're like, but the power is in my hands, too, so I can go do something with that, right? So it's both.
616. 1:37:50I don't think you can kind of convince people to love it. I think the best that you can do is you can say, "Listen, hold that hate. I
617. 1:38:03understand where the hate's coming from." Like it's it's just it's like this one of those marriage counseling
618. 1:38:07things. Yeah. I I like mirror it back to them. I absolutely understand why you hate it. You know what? There's things
619. 1:38:13about AI I hate too. These four things you just talked about, I agree with you. I absolutely hate that part of it, too.
620. 1:38:21These three things you talked about, it's not actually how it works. But I I mean, I'm happy to explain that, but
621. 1:38:26that doesn't matter. But you know what? When you experience it, it's a very different relationship than when you're
622. 1:38:36just looking at it from the outside because you can see that it's not perfect. And if you can get them to give
623. 1:38:41it a shot and if you can get them to do things that are relevant to them and they try it and they have one of the
624. 1:38:50things we talk about in here is is one of my joys in life is watching people have their first Kevin Mallister
625. 1:38:58moment. If you don't know Kevin Mallister, he's the the kid from Home Alone where he puts his dad's after
626. 1:39:04shave on. He's like that moment when you use AI when you're doing it right and you and especially when you do something
627. 1:39:11that's relevant to you. So if you're a writer you're like well it couldn't possibly write this kind of thing and
628. 1:39:16then you prompt it and it does and it's good you're like holy [ __ ] Uh right that moment like oh my god
629. 1:39:28when people have that first moment with AI you kind of can't unhave it. And so what I've experienced is people that are
630. 1:39:36really pissed off about it, even if they reluctantly come into it and give it a shot and play with it enough to have
631. 1:39:42their first Kevin Mallister moment, then they're like, "Uh, okay, I think I still hate it, but this is kind of
632. 1:39:50interesting." And then they get to, "Oh, I think I kind of hate it, and actually this is kind of inspiring." And then
633. 1:39:54they they get to this more sophisticated place where they're both excited and terrified at the same time. And then
634. 1:40:02sometimes like my co-founder of the AI salon, Leah Faston, she's a professional photographer, she went through that cycle and she
635. 1:40:11actually is stepping away from the AI salon because she realized for her personally AI was not feeding her
636. 1:40:18creatively. She just didn't like it. And like I have the deepest [ __ ] respect for that because like she came into it
637. 1:40:27hard. like she co-founded a 30,000 person community with me of AI optimists and then realized it wasn't for her.
638. 1:40:37Like that I'm totally down with. But it's the people that are sitting on the outside that are just using these tropes
639. 1:40:43like it's the world's greatest plagiarism machine. It steals. It uses too much water. It uses too much energy.
640. 1:40:50Some of those facts are true. Some of them are not. But if you're using those things as an excuse to sit out and not
641. 1:40:57try it and not understand what it makes possible and not understand what it might do for your ideas, it just feels really [ __ ]
642. 1:41:06shortsighted to me. Like that's what this channel's about. So there you have it, man. I'm on a [ __ ] tear tonight.
643. 1:41:18Tyrate Tuesdays is where it's at. Hallelujah. [Laughter] Oh, good lord. Did that make any sense?
644. 1:41:33It'll be difficult to go on tour. Actually, you know what's funny, Professor Crystal? I think in-person
645. 1:41:39concerts. I think the value of in-person things are about to go up in value dramatically because it's going to be so easy to
646. 1:41:49produce all this digital stuff. People are going to have like hunger for human connection. Um I think in a weird way robots return
647. 1:42:01us to our humanity. I think they they allow us to ascend out of the drudgery of the systems that we have to interact with today by auto
648. 1:42:13automating a lot of that. Let us ascend out of that and then I I think the thing left to do when you ascend out of the
649. 1:42:20drudgery is reconnect. And so I think I think there will be jobs like um I'll walk with you for an hour for 50 bucks and people will pay 50
650. 1:42:32bucks to have someone come walk with them. And I know that sounds weird right now, but like we've got people making a
651. 1:42:38million dollars a year running YouTube channels. I'm not one of them. I I need to talk to one of those people and
652. 1:42:44figure out how they're doing it. Um, but like we're gonna we're going to want to reconnect. So, I think I think tours and
653. 1:42:54concerts and you know, take take songs that you've written in Sunno and start a band. Make a cover band of the songs you
654. 1:43:02wrote using AI. Yeah. The Taylor Swift four billion dollar economy. All right. It's like when Fetch Fred
655. 1:43:11Savage had the Power Glove. Exactly. It'll be difficult to go on tour. Tony Williams, my cousin just went
656. 1:43:19through the same cycle of fearing it to being inspired. Yeah, it's so I've got Let me share one other thing and then
657. 1:43:26we'll we'll get on out of here. If you go to the AI salon, so if you go to the salon.ai and say join our community,
658. 1:43:33the very first little space is called welcome to the AI salon and
659. 1:43:47I've got in here what I call the we've got the cycle of AI readiness. Play first, create excellence, generously
660. 1:43:54lead. That's kind of the the the ethos of the salon. Um the mission and I've got the the five stages of AI adoption.
661. 1:44:04And I base these on the five stages of grief because I because I think it's a similar kind of thing. The first stage
662. 1:44:10of grief is is denial or dismissal. And so I think a lot of people right now they they've either demonized AI or
663. 1:44:16they've dismissed it. It's not very good or or or they're just, you know, it's bad, it's evil, right? It's going to
664. 1:44:22kill us, right? And so so step one is dis dismissal. It's just another tech fad. Nothing to see here. People
665. 1:44:29dismissing. Oh, did you use AI for that? I can judge you now. I can judge you as a talentless oaf. Yeah. Yeah. Yeah. AI.
666. 1:44:37I push a button, out comes a thing. Yeah. Yeah. Whatever. Right. I get a lot of that, right? That's dismissal right
667. 1:44:44there in stage one. Stage two is you've had your first Kevin Mallister moment. It's this moment of awe where you actually for the first time
668. 1:44:53understand you you internalize kind of the power of what this thing is. Computers have never acted this way in our lifetimes.
669. 1:45:05Computers have always taken an input, processed that data in some sort of logic that was programmed by a human.
670. 1:45:13That was a deterministic output, right? You put in an input, it processes it, and it comes out in a predictable way.
671. 1:45:21That's not what computers are doing when they're using generative AI. They are generating new original stuff. Stage two
672. 1:45:28is where you experience that and you go, "Oh, I didn't know it could do that. I had no idea that phrase, I had no idea. I hear
673. 1:45:37that over and over and over again where I'm showing someone something. They're like, "Oh yeah, I use chat GPT all the
674. 1:45:43time." Yep. Yeah. I use it for emails. Yep. Yep. Yep. Yep. I use it a lot. Yep. And then I'm like, "Oh, did you see it
675. 1:45:49could do this? Oh my god, it could do that." And I'm like, "Well, yeah, if you didn't know it could do that. Did you
676. 1:45:54know it could do?" Oh my. It can do I had no idea. Like that I had no idea moment. That's stage two. you realize
677. 1:46:03all of a sudden this thing is something different, right? Stage three is you now start playing with it and you you want
678. 1:46:10to discover it on your own and and you go, "Oh my god, it can do this. It can do that. It can do that." You get really
679. 1:46:15excited and then you realize, "Holy [ __ ] I've spent 30 years trying to be good at that and now it's better than
680. 1:46:21me." Fear, wonder, fear that coexists. Stage three is this coexistence of wonder and fear. Stage four is when you
681. 1:46:32kind of let go of the ego and say, "Okay, it's smarter than me." Because it is. If you don't think it is, it is. I'm sorry. It is.
682. 1:46:45And you give into it and you say, "Wait a minute. I could either be in competition with this thing or I could
683. 1:46:52be in collaboration with this thing. I'm going to choose to be in collaboration with this thing. And what I'm going to
684. 1:46:57do is I'm gonna have the ideas and I'm going to strap this AI on thing like a jetpack and I'm going to have it amplify
685. 1:47:05my ideas. That's stage four. I find myself right now going back and forth between stage three and stage four. I
686. 1:47:12use it like a jetpack. I'm like, "Oo, this is great." And then like some new thing comes along and I'm like, "Holy
687. 1:47:17[ __ ] Oh my god, that's amazing. Oh my god, that's terrifying." Right? Until I kind of internalize it and put it back
688. 1:47:23in the jetpack. And then stage five is is kind of pure self-expression or or uh David Shapiro calls it cognitive hyperabundance.
689. 1:47:33The the point at which it's just literally anything you want you just ask for and it will materialize.
690. 1:47:39We're not anywhere close to stage five yet. The tech the tech isn't there. So anyway anyway all right. Yeah. Well, we're not special.
691. 1:47:53Not special. The salon is great. So, do me a favor. If you have not joined the AI salon, if you have not joined the AI
692. 1:47:59salon, get your ass there. The salon.ai. We have a brand new a shiny new website that's amazing. Tells you all about it,
693. 1:48:14all the stuff we're up to. You'll see events. You'll see posts from the community. And you say, "Join our community." and
694. 1:48:22that'll take you into the community site. And we're rearchitecting it right now. We're kind of halfway through that. It's
695. 1:48:29really exciting. All right, so go do that. Um, today's Tuesday, tomorrow's Wednesday. So, tomorrow night it'll be
696. 1:48:358:00 p.m. Same time, same place. Um, I didn't even get to there were a couple of things I wanted to show tonight.
697. 1:48:42There's some new features in Runway ML that I thought were kind of exciting. Um, I'm going to do So, when did I when
698. 1:48:52did we do back to basics, Brandon? Three two months ago. September. Okay. So in September I did this thing
699. 1:49:03called back to basics where I took five days and day one was overview of AI, day two was chat GPT, day three was images,
700. 1:49:13day four was sort of integrating it and day five was questions. I'm going to do a new back to basics. I'm going to do a new five-day course.
701. 1:49:22I'll probably do it in the next three weeks, four weeks, something like that. I need to put it together. It's going to
702. 1:49:29be five days. So, five nights here and all five nights are just going to be chat GPT. We're going to do five nights of chat GPT
703. 1:49:41just figuring out what the [ __ ] is in it. Because one of the things that I confronted when I did my back to basics
704. 1:49:49course where chat GBT only had an hour was in a in one hour I could not even scratch the surface of what chat GPT now
705. 1:49:59does. So I'm going to do a full 5day week coming up. So just know again just keep coming here nightly. Become an
706. 1:50:08irregular. If you show up here nightly you are an irregular. It's weird. You shouldn't do that. You shouldn't come here, but people do,
707. 1:50:19but keep doing that. And then and then we're going to do a deep a relatively deep dive on chat GPT. We're going to
708. 1:50:25take a full week on it. Um, so anyway, all right. Beautiful. Beautiful. Tomorrow, Brandon and Gwyn are at MCON, so they're there
709. 1:50:36as representatives of the AI salon. All chat all the time. Exactly. How hard is it to learn how to build an app? How
710. 1:50:44complicated of an app can I build by myself? I'm has. So, first of all, what you need to do is go watch the beginning
711. 1:50:51of this episode. So, when we stop when I press stop tonight, this will be live on the YouTube channel. So, the YouTube
712. 1:51:01channel is learning lab-ai on YouTube. Um, we vibecoded at the beginning of this. We created an amazing app with a
713. 1:51:11single prompt. You can do incredibly sophisticated apps with zero knowledge. Um, they may not be scalable. They may
714. 1:51:19not be perfect, but if you've never coded before, what you can do right now with zero knowledge is insane.
715. 1:51:28And in fact, if you just want to blow your own mind and and not watch me do it, just go to lovable.dev. dev lovab.dev.
716. 1:51:40Just go there right now, describe your app, and hit the button and then sit back and wait for five minutes and see
717. 1:51:46what it generates. It'll blow your mind. It'll blow your mind. Yeah, exactly. Try lovable.dev. There
718. 1:51:54you go. All right. Um, that's it. I hope you all had fun tonight. Um, enjoyed my little tirade. I was all
719. 1:52:05passionate tonight. Good lord. Um, I'm gonna go rest and have fluids. I can tell that I talked too much and didn't
720. 1:52:13drink enough. So, I will see you tomorrow. Bye, everybody.