5/12/2025 - AI's Creative Leap: From Text to Coherent Character Animation and Scene Control

Transcript

1. 0:02[Music] [Applause] [Music] [Applause] [Music]
2. 0:23The boys were sing shingling the summer that we You were 10 and 17. How could I
3. 0:38forget? All the stars from near and far were watching from [Applause] a watching two teenagers fall in love.
4. 0:54The way we danced was not a dance, but more a long embrace. We held on to each other and floated there in
5. 1:09space. I was shy to kiss you while the whole wild world could see. So shingling said everything for me. [Music]
6. 1:24And all the poor old folks, they thought we'd lost our minds. They could not make heads or tails of the young folks funny rhymes.
7. 1:39Uh, you and I knew all the words and we always sang along to old jam dong jam bang.
8. 1:54[Music] Well, after years and after days and after summer's passed, we held on like we held on to
9. 2:09each other. I don't know the [Music] words. Happy Monday, everybody. That is
10. 2:26Champ singing along. He knows the words better than I do at this point. producer Brandon checking in saying hi. Good day.
11. 2:32Good day. Good day. What's happening? Um, we, you know, sometimes these Mondays, it might be a meltdown Monday,
12. 2:45but I don't think it's going to be a meltdown Monday. But I I I've right now that is similar to a feeling I had back in December, January.
13. 3:01And you know what came after that was, you know, some of these really fancy models and some of the agentic stuff. I
14. 3:10have a feeling we're on the cusp of something big [Music] again. Kevin Stewart, happy belated birthday. Thank you very much. I
15. 3:21appreciate that. You got the number wrong. It's not 60. That was a rumor. That was, you know what it was? There
16. 3:29were there were internet trolls and the internet trolls were like, "Yeah, he's [ __ ] old. He's [ __ ]
17. 3:3660." And you know what's amazing about because of AI now? If you go back and you look up my birth certificate, the
18. 3:44trolls got to that, too. Like, my birth certificate would agree with that assessment. But we I mean, look at me. This is not 60. Come on.
19. 3:56So anyway, still can't tune my guitar for [ __ ] [Music]
20. 4:13[Music]
21. 4:30[Music] KKE. Hey Kyle, can you show us how you aggregate content about AI for post ideas?
22. 4:50I listen there are no bad questions. That's just a question that built into your question are some assumptions that I have a [ __ ]
23. 4:58plan. So I will absolutely tell you how I I put my thoughts together for things and what I pay attention to and
24. 5:07how how I pay attention to it. I'm happy to do that. Um, I mean to be quite honest, the the biggest the biggest thing Okay, Steve
25. 5:23has something for me in the regulars. The biggest thing that allows me to follow this stuff is just a massive sense of
26. 5:35curiosity. Like I just want to know like what's coming, like what's next? What's coming? How does it work? What's it do? What's it make
27. 5:42[Music] possible? I'm getting yelled at by my producer. It's very rude. He's saying, "Switch your cameras. Do the black bar.
28. 5:57Everyone's annoyed with your life. I just, you know, do do I deserve this?" No. I'm here. I'm here. I'm just
29. 6:06I I've got a guitar. I've got 17 chords, 15 songs, and a [Music] dog. Wow, this is so out of tune. It's
30. 6:21not playable. [Music]
31. 6:41[Music] That's better. A little bit better. Thank you, Kyle. I'm thinking of blog content topics. I appreciate it. I have
32. 6:55seen you do it before and like the thing. Um, you know what we'll do? So, I'll tell you what we'll do, Kake. I'll I'll
33. 7:05tell you my general how I do it. Like, I I'll share my methodology with you. If if if you're expecting my
34. 7:14methodology to have anything resembling an actual methodology, you'll be gravely disappointed, right? Um, no, I don't use Nomatic. I
35. 7:27created New Zomatic. I don't use New Zomatic. Um, I now have a couple of uh like Brandon will send me stuff. I've got
36. 7:39people that send me stuff occasionally here. I'll show you what I do. Let me just I I'll share my screen. We'll go look at some
37. 7:50stuff. I've got um Chat GBT doing some deep research on how to get Sydney produced. So, we'll come back to that.
38. 8:01Um, this is mainly what I do. So, okay. So, so I'm on I'm on the Twitter. So, it's mostly Twitter and then I'll I'll
39. 8:13occasionally if it's something big, I'll I'll occasionally look at LinkedIn, but it's mostly Twitter. And the key with it
40. 8:24is you need to groom your you need to prune or groom or garden your who you follow. And so what I did when I first
41. 8:35started getting into this is, you know, I found people I liked. I know Robert Scoble was way into this stuff. I knew
42. 8:42Ethan Malik was way into this stuff. Paul Ritzer. Um I'm trying to think who else were some of the early ones.
43. 8:49you know, Sam Alman, Greg Brockman, like, you know, players at the major at the major labs. And so I would follow
44. 8:57them and then I forget who told it to me. I don't know, someone probably younger said the people that you most
45. 9:07respect, go figure out and and Twitter will tell you who they follow. And then so I just spent like a week kind of
46. 9:15following people that are followed by um people I liked and then I would pay attention to who posted [ __ ] and then I
47. 9:24would go look at who they followed. And so I did that for I don't know maybe a month probably back in 2022. I don't
48. 9:30think I've updated it since then. Um so that's part of it. So part of it is just having and you can do
49. 9:37lists. There's there's things you can do where you can do searches and save those as like I've got I've got groupings of
50. 9:45people I follow like here's genai list but I just I just basically go to my for you page for the most part there's Riley
51. 9:56um and then what I do is this and then this is here's the big if if there's a secret that I have again like it's not
52. 10:05that much of a methodology I'll go down and I'll I'll like there are so many people now doing sensational headlines
53. 10:14that I I'm starting to recognize if it says um you know such and such is dead I just ignore that if it says um if it
54. 10:27says um Google just released right and then I look at what they're talking about what Google just released and if
55. 10:37Google released it two months ago. They didn't just release it. That's just some [ __ ] right? Because Google just released X followed by 10
56. 10:48crazy 10 crazy examples. Th those are like everyone's copying everyone. So, so those I just completely ignore. Then
57. 10:58what I do if I see someone like hater, I like um slow developer. I like um here's retrunk. I like him. he's an irregular.
58. 11:07Um Timmy is the guy from Runway. Like you just start to recognize people that post regularly and that you like. And
59. 11:15then what I do is if I see one person post about something, I'll look at it if it's someone I respect and I'll check it
60. 11:24out. If I see two or three, I'm like, "Huh, there's something here." If I see 10, you know, just in in like a single
61. 11:31scroll pass, then I'm like, "Something's up. So most of what I pay attention to, like I tend to ignore um open
62. 11:41source just because I'm not technically comfortable enough to just be filling up hard drives and switching things and maintaining
63. 11:53versions and just all that [ __ ] just exhausts me. So I just kind of ignore that. And and I figure at some point
64. 12:02people are going to use those open source tools to build some cool thing and a lot of the tools that we use today
65. 12:06are are that, right? So so I figure the open source will take care of itself. Um I tend to ignore things like iPhone
66. 12:18apps and things that are like little small things that are built on top of other people's tech. Um, and then I just and then I just pay
67. 12:30attention and and you know often what I'll do like there are a couple of people that I watch. Ethan Mllik's one of them. David
68. 12:39Shapiro is one of them. Um, occasionally Robert Scoville. His are usually one oneoff hits, but like Ethan Malik and Robert Sco or uh
69. 12:50Ethan Malik and uh [ __ ] who the hell was I did I say? Uh who was the other one I said? Ethan Mllik and oh David
70. 13:02Shapiro. They tend to go on topic benders, right? So like David Shapiro will do 10 things about universal basic
71. 13:11income or you know post post labor economics. Um and when he's on a bender something's something's simmering right
72. 13:19and Ethan Malik will often go on a bender or or Ethan Malik will do three pieces of research in a row that are all
73. 13:26related and that from those two in particular. They've got early access to stuff. They're they're up on this way
74. 13:34more than I am. that tends to point me in a direction and then I'll start looking for what they're talking about. Share the
75. 13:49live. Which which live? Oh yeah, we've got shitty numbers here. So do me a favor. If you're here, if you're on YouTube, if you're
76. 14:01watching this anywhere, can you send this to your friends? We only have a handful of people here. Danielle
77. 14:07Rockstar sharing like like a banshee. Love it. Thank you. Thanks for all the cool little What are all the cool little
78. 14:16heart hearty hearts? We're getting hearts and lipsticks and party favors. What is all that [ __ ] Did I miss a memo from Tik
79. 14:27Tok? Of course you did, Silver Fox. You always do. Wow, we got all sorts of cool new goodies instead of just hearts tonight.
80. 14:39Um, so anyway, so that's basically that's basically what I do. And then, you know, it's funny. I go through these waves of I won't post
81. 14:50anything and I'll be like, uh, I don't have any ideas. And then someone will post something and and all of a sudden
82. 14:57just a really clear concept will pop in my head. And sometimes they're small and sometimes they're big. And that's
83. 15:04usually when I write an article on LinkedIn. Um, I used to post a lot more about this launched and that launched
84. 15:11and that launched. Now I'm much more interested in did something launch that's actually going to change the
85. 15:26game. So I don't know if that helps. [Music]
86. 15:49[Music] irregulars video. So I got to change tabs. What am I looking for here? Get out of here with your stupid pole.
87. 16:03What am I looking for?
88. 16:25Steo. I don't. Maybe I gotta [Music] reload. Oh, my first attempt at anything
89. 16:41other than an image on Dollali. It was fun to make. Hope it brings a smile. Oh, Champy singing. Okay. All right. Here. Let me let me get it
90. 16:51rolling. So, it's probably going to launch a new tab. What? Oh, it downloaded
91. 17:08it. Oh, I think I can do it without sharing tabs. Here, find a night. Every fella's got you
92. 17:24[Laughter] in. What you doing with a clown like me? Surely one of life's little
93. 17:39mystery. So, [Applause] that's you. You want to sing with you? Listen. [Music]
94. 18:00What did I say? That's awesome. The the uh the backup singer there was not great, but the lead dog was good. Pika dropped a new thing. Even
95. 18:15Champion is lip-syncing is adorable. Yeah, I know, right? That's awesome. Thank you, Steo, for that. That
96. 18:26was very nice. Very nice. Very nice. Maybe I'll share that on the uh on the Twitter. Oh, man. Um [Music]
97. 18:48[Music] [Music]
98. 19:06Is this place I can rest my [Music] forehead gather my thoughts in sweet silence. [Music]
99. 19:20There's this place where the feelings are [Music] dead. From an over this place I can slowly face. The
100. 19:34only one I truly can know. These are tears from a long time ago. Got these tears from a long time ago. I need to cry. 30 years or
101. 19:50so. These are tears from a long time [Music]
102. 20:07ago. You done? You good? Um, okay. I think Champy's good. So, there's two things that I think I want to do tonight. Um, and you
103. 20:20could derail me if you ask a question and I lose my [ __ ] on it. Cool. We can do that, too. Um, or if you come in and
104. 20:27troll me in an uncreative way. It's meltdown Mondays. We We just might get some fireworks. Who knows? Oh, Winston.
105. 20:33Oh, that's gorgeous. Thank you very much for that. Very generous. Starmap Polaris. Love it. Damn. Dag.
106. 20:43Um, so I don't know if you remember Mr. Strawberry. He he was he came out I don't know a year or so ago and was
107. 20:52hyping up hyping up you know feel the AGI and hyping up Open AI and then Sam Alman retweeted him and he annoys the [ __ ] out of me because
108. 21:04the stuff that he posts is just hypy [ __ ] and it like it rarely like it often doesn't come true and and it's
109. 21:14like it's just it's just hypy [ __ ] But he posted something today that looks like he didn't take his meds or he's seen
110. 21:23something. And so I want to read it, not because I particularly believe him, but there's something about how it's written
111. 21:32that I think is what we're about to experience and and it's kind of nuts. So even if the timeline's different from
112. 21:40what he says, I think this is about what we're all about to experience. So that's that. And then the other thing that I
113. 21:47thought might be fun to try and we'll see how fun it is because it this could also be not fun. Um I got access to
114. 21:55dscripts agent where you can just talk video editing into existence. Um and I tried it once and sort of it kind of failed on me and I
115. 22:06didn't I never really dug deeper. So, I thought it might be fun to, you know, sort of talk about the point that he
116. 22:13made or prediction that he made and and what that's going to look like because I I think we need to get
117. 22:22ready. And again, I think, you know, the the folks that hang out on this channel regularly, um, we're as prepared for what's coming
118. 22:31as anyone. So, I, you know, I think we're fine. Um, but I think shit's going to start getting weird.
119. 22:37Um, and then I thought we'd just go play with Dcript and just like I'll just sort of, you know, learn and and fail
120. 22:46publicly or learn and succeed publicly. Um, damn, you and the dog and the guitar make a great trio. Oh, thank
121. 22:53you very much. I appreciate that. Yeah, you know, the doggy the doggy's the lead singer. I'm just here backing up the uh
122. 23:00the the diva. The diva we call champ. [Music]
123. 23:34All right, Let's let me go. Let me go find this thing and then and then once I find it, we can we can share it. Let's see. Uh I
124. 23:50[Music] [Music] rule. Hold please. Oh, wait. I responded to him. So, let me I can find it on my
125. 24:05thing. Did I respond or did I did I reply? I don't think I replied. Hey, I I'll be with you. Just calm down. Here it is.
126. 24:21Okay. All right. So again, the context here is that the guy whoever's behind this
127. 24:35account is mostly full of [ __ ] and it's mostly hypy [ __ ] and I I you know it's it's got a crypto bro
128. 24:46kind of feel to it. So I have a feeling this is someone that's kind of connected with Worldcoin or you know World ID or something like that.
129. 24:56But I'm just gonna I'm just gonna read what he said and then so take it with a massive grain of salt. Um and and then I'll sort of unpack at
130. 25:12least like this is one of those things where this has me now starting to starting to pay attention to what's
131. 25:19coming. So he goes, "It feels quiet, doesn't it? You are not prepared for the next four weeks. Not at all. And by the way, if
132. 25:28you want to see this account, it's I rule the world mo as in master organization mo. Um, just spent six hours with Deepseek
133. 25:42R2 and I literally can't process what I witnessed. The cognitive architecture is [ __ ] alien. No one is ready to admit
134. 25:50the intelligence gap between the last gen models and what's happening now. We're not seeing 2x or 10x jumps
135. 25:56anymore. We're seeing whole order of magnitude leaps in capabilities capability landscapes that didn't even exist before.
136. 26:09Um, there's a good possibility this is horseshit. Jim Ross is now officially in love with Lovable. I'm telling you,
137. 26:22Lovable, if you have not played with lovable.dev, go there right now and instead of listening to me, just go tell
138. 26:29it to make you an application of any kind and watch what it does. If you haven't played with it yet, it's insane.
139. 26:36Uh, as in Strawberry. Yes, this is Mr. Strawberry is is what what he was called when he was doing all that. I would
140. 26:42snark about Nicks on social media sites, but I'm a sea slug. Okay. Um, I watched it solve three impossible math conjectures while
141. 26:54simultaneously building a theoretical framework of consciousness. Like, as I'm reading this, I'm thinking, is this just
142. 27:05horseshit? This could just be horseshit. Let me keep reading it and then just assume that everything I'm reading is
143. 27:17horseshit because I I I do want to talk about it at the end because I do think that what he's describing here is going
144. 27:25to happen. I just don't think it's going to happen in the next four weeks. Um, I wa I watched it solve three
145. 27:31impossible mathematical conjectures while simultaneously building a theoretical framework for consciousness
146. 27:37transfer. Uh, consciousness, yeah, consciousness transfer that made actual sense, not metaphorical sense, not
147. 27:43philosophical sense, engineering specification sense. Researchers in the room were physically shaking. One kept laughing
148. 27:51uncontrollably uncontrollably saying, "This isn't supposed to be possible yet." 03 Pro and Gemini Ultra are
149. 27:58running similar architectures now and the emergent behaviors between interconnected systems are producing
150. 28:06capability spikes that nobody predicted on any timeline. forget AGI debates. Um, uh, Grock 3, Grock 3 is the only one
151. 28:20being somewhat transparent about what's happening under the hood. And even they're downplaying the weirdness by at
152. 28:25least 60%. That's a weird number, but like why is that number there? So again, like I I feel like he might be off his
153. 28:33meds. When when these systems start self-modifying beyond our ability to trace execution paths, the acceleration
154. 28:41curves be become actual mathematical singularities, not metaphorical literal computational explosions that create capabilities no one can
155. 28:53predict. Um, and so he goes on, he says this is going to happen in the next four weeks. So whether this happens in the
156. 29:01next four weeks or the next four months or even the next four years, um I do think there's going to be kind of a snap um in
157. 29:12capability. And I also agree with him that all of the from from what I've experienced over the past year, all of the frontier model
158. 29:26companies, they're either copying one another or they're starting to use the AI to improve itself and it's
159. 29:34improving the models in similar ways. But but I think we're going to keep seeing this thing where you're going to
160. 29:42get one of them release something remarkable and then you'll have a bunch of very quick follow on. So next week is
161. 29:51Google IO and Google's supposed to be releasing new [ __ ] and everyone else is probably going to launch new [ __ ] to
162. 30:00counteract that. So what we don't know is is anyone going to launch new [ __ ] that is actually really gamechanging and
163. 30:08is kind of a snap function that is way beyond what we have right now. And I don't know but um I just have this simmering feeling
164. 30:18like something something is coming that is not going to feel we're not going to know what to do with it is I guess what I'm saying.
165. 30:30Um, but you know, [ __ ] I feel like I don't know what to do with the current tech we have. So, so there's that. Uh, thoughts,
166. 30:41questions. As now that I've now that I've read that, don't know lovable. If you need to go play with lovable. Hi, H. Hello, Stacia. What's
167. 30:50happening? [Music] I have ranch dressing on my word [Music]
168. 31:14salad. Co-pilot lied to my face repeated repeatedly today. Co-pilot. Yeah, all of these things lie repeatedly. They just do. They're just
169. 31:27bad. They're just bad. We're going to so look back on this time and be like, "Oh, remember when they used to lie and now they just flat out
170. 31:40manipulate." All right. Um, what are we going to do? Maybe we'll go make something lovable. I have a couple of use cases
171. 31:48lined up for lovable. What should I make in lovable? You guys that are doing it. Jim Russ, what's something I should make? Hey, Mr. It,
172. 31:57what's happening? I'm doing well. [Music] Uh, the more I read of that Mr. Strawberry thing, the more I think it's
173. 32:10[ __ ] So, that's why I kind of cooled down on not digging deep on that. I do think that we are going to have at
174. 32:18some point one of these frontier model companies is going to launch something that's just going to be very different.
175. 32:24Um and then and then and then stuff starts to get crazy. All right. Um maybe I should tell it to be impatient. Oh, I'm impatient. B Yeah,
176. 32:37you can listen. If you're not yelling at your if you're not yelling at your uh your your large language models, you're
177. 32:44not doing it right. I yell at them all the time. Tell them they're being idiots. Tell them to dig deeper. I like all the time. I'm like it
178. 32:54just they're exhausting. They really do give you sort of sort of like the the average answer, the average non
179. 33:06shitty answer to every question unless you, you know, trick it into thinking deeper. So, um, all right. Here we are on lovable. Let's
180. 33:19see. What do I want to make here that Oh, you know what would be fun that that I haven't done before? Okay, we'll we'll do this.
181. 33:29Um, I want to create a synthesizer. Synthosizer. How do you spell that? S I
182. 33:46Z er. Yes. I want to create a [Music] synthesizer musical
183. 34:00instrument that plays like a video game. You know what? We're not going to we're not going to architect this here. We're going to
184. 34:12architect this in chat GPT because you only have so many hits with lovable. And so we're going to do this.
185. 34:24We're going to do this new. We're going to do this with 04 mini high, which is great at coding and visual reasoning. Or do we want to do
186. 34:3403? Let's do 03. We'll do the big MacDaddy model. All right. So I want to create a synthesizer musical instrument that plays like a
187. 34:44video game. I want it to be very visual.
188. 35:02Incorporate music theory in a way that the user just benefits from its
189. 35:17knowledge but doesn't need to know theory.
190. 35:35Um, the video game component should score you higher as you layer on.
191. 36:00[Music] sounds and beats. Act like a senior product visionary and manager.
192. 36:19And let's uh create a spec for this bad boy or
193. 36:34girl. How are you all? My father-in-law. I got my father-in-law to stop start watching Kyle over the weekend. Sorry
194. 36:42about that, father-in-law. It but it doesn't get better than this. This is what you're experiencing right now. This
195. 36:49is the pinnacle of what this channel is. So, if you're expecting more, ju just don't because what where it very quickly can
196. 37:02devolve is things like this. You know, you know what I'm saying? And I'll just talk about [ __ ] with a pink bow on for no reason.
197. 37:14Well, mainly the reason is I was told by the woman that gave me this, I'm wearing it wrong. Every headband I've ever worn,
198. 37:23this is how you wear it. Apparently, I got I got woman explained how to wear a pink bow. Okay. Okay. He and Kyle are the same. Well, that's
199. 37:39bad. If you're expecting more, just [Laughter] don't. Oh, my daughter who writes grants for a nonprofit had chatbt write an
200. 37:52entire grant first time. She's pretty excited. Oh, that's cool. Yeah, when you when you do stuff like uh when when someone hasn't had their
201. 38:01first Kevin Mallister moment and if she's in grant writing and this thing just knocked out a [ __ ] grant. Yeah.
202. 38:07It's, you know, because you got to figure large language models have been trained on all the [ __ ] that's been put
203. 38:14on the internet. Well, what [ __ ] that's been put on the internet? Every [ __ ] grant application, you know, grant
204. 38:20writing course, every grant, right? Every failed grant, every succeeded grant, every analyzed grant, every business school
205. 38:28grant, they're all in there. So, it knows everything. So, like that's a great use case where it's just going to
206. 38:34be really good at that [ __ ] All right. 03 thought for three seconds. It's called Synth Quest, a gamified visual
207. 38:41synthesizer experience. Product vision. Invent the first Play It Like Mario, sound like Brian Eno instrument. That's
208. 38:50brilliant. Play it like Mario, sound like Brian Eno instrument. Players sculpt beats and melodies by jumping,
209. 38:58sliding, powerup, stacking audio layers. Under the hood, we enforce solid voice leading harmonic rhythm and modal
210. 39:06interchange. I don't even know what the [ __ ] that is. Tik Tok pin. Yeah, if you're expecting more, don't. Well, welcome to the AI learning
211. 39:17lab where ain't no one learning and ain't much AI. If you're expecting more, don't. All
212. 39:31right. Invent the first. Play it like Mario. Sound like Brian Eno instrument. Like that's a [ __ ] marketing scheme right there, isn't
213. 39:40it? Player Sculpy. Okay. Under the hood, we enforce solid voice leading harmonic rhythm and modal interchange. So
214. 39:49whatever they do sounds legit, but the player only sees color, shapes, and points climbing. Target personas. The bedroom
215. 39:58producer, streamer Sam, the music music ed mentor, core gameplay and UX loops, invisible invisible theory
216. 40:12engine, circle of fifths, world map ring, warp pipes, modulate keys. Okay, voice leading rules. It auto renders
217. 40:21collect. Okay, it doesn't [ __ ] matter. This is already above my head just in the product description. This is
218. 40:30unbelievable. Oh, so Oh, yeah. Base game. Okay. So, wait. Okay. Okay. This is great. You can pull out the marketing stuff for now and the
219. 40:49timeline stuff. I want you to write this as a product
220. 41:04spec assuming we can just make it happen order. Uh let's see. I'll be using an AI
221. 41:23vibe coding app um to execute it. So I just need it very well
222. 41:39articulated. Uh, make sure that the game play and visuals are world class. Don't get lost in the geeky music
223. 42:00[Laughter] [ __ ] All right. Fantastic. Fantastic. Bob, tell him what he's won. Uh, he hasn't won anything. It's a This is not
224. 42:16a game show, right? Create the first game field synthesizer.
225. 42:30Wait, where? Okay, we're going to go steal the the product vision. So, we're going to pop that into lovable.
226. 42:44Boom. Then we're come back over to chat. Jetpack. We're going to scroll on down to the bottom here. End of spec. Oh,
227. 42:56Jesus. Game play note router. Unity render loop theory engine. Good lord. I wonder if I can I copy all this
228. 43:08together? I bet I can. All right, fine. So, here we're going to go. All right. Purpose and North Star audio
229. 43:20requirements less than 10 milliseconds. All right. Here we go, people. Boom. This is a big [ __ ] complicated ass thing.
230. 43:31Paste. Make it. Oh, sign in. Damn it. Shut up. You know what? Does he have any qualific What are his qualifications?
231. 43:46Does he have any qualifications? Is he a professor? He he seems he seems like a an atypical If he's a professor, he seems
232. 43:56atypical. He He doesn't seem like he he knows anything. is that that this is what you do here. Huh. It's a novel approach to
233. 44:10education. It takes my breath away. It's just a lot. All right, we're
234. 44:25off. Spinning up the preview. Okay, so now it's going to do its [ __ ] This is vibe coding. If you haven't seen
235. 44:31vibe coding before, it's like come up with an app idea and tell it to go make it and then it does [ __ ] and then it works or it
236. 44:42doesn't. He got a certificate from a cereal box. Yeah, but I lost the cereal box before I cut the certificate off it.
237. 44:49That's the problem. Nobody else hear that? Hear what? Joker, are you hearing those voices again? You You gota You gotta get off the the
238. 45:04ether there, buddy. We're off like a herd of turtles. Oh, man. All right. What's it doing? What are we doing
239. 45:25here? Too many mushrooms. Exactly. Exactly. Duty-free designs World Bazaar. I like it. Solid name. Solid
240. 45:36name. Okay. I'm excited to create this innovative music game hybrid that combines performer mechanics with audiovisisual synthesia.
241. 45:47The concept blends the tactical feel of Mario platforming with Brian Eno's ambient soundsscapes in a way that makes
242. 45:55music creation intuitive and visceral. Drawing inspiration from games like Rez Audio Surf and Hi-Fi Rush, I
243. 46:05envision a colorful fluid experience where movement directly translates to music with vibrant visuals responding to
244. 46:13every note. The cell-shaded neon aesthetic will with particle effects will certainly will create sinthetic connection between sound and
245. 46:27visuals. For the first version, I'll implement basic platformer mechanics, sound generation, visual feedback, theory engine. Cool. I'll use
246. 46:39vibrant color palette and neon stuff. All right. It's making [ __ ] now. Now, if you haven't played with
247. 46:48Lovable before, like what it's doing right now, I think. Can we go look at uh open preview tab? Yeah. Here's all the code it's
248. 47:00writing. So, see this little block over here to the left? This little thing right here? It's got the little code. It says writing the
249. 47:08code. So if you click on that, here's all the code it's writing. So here's index.tsx game context. A lot of game
250. 47:24context, then audio context. If you've never coded before, you can just ignore this [ __ ] But if you're like curious, like a good
251. 47:36way to learn coding is watch what these things are doing. They're doing it way fast. Way fast and way competent. The intro
252. 47:50overlay configure. Oh, spinning up preview. We just went on to our next thing. So, it created Whoa. Whoa. All right. What do we
253. 48:02got? Error. Try to fix it. Vibe coding baby. Kyle, this is Mr. It trapped in my
254. 48:19tablet in my other Tik Tok account. Dutyfree designs is is not really a duty-free design shop. Well, it might be, but it's Mr.
255. 48:31it. So, it's an irregular. Ah, okay. Harmonic jumper. Create beautiful music through movement. Jump and collect orbs. Okay.
256. 48:42Arrow keys to move. Okay. Space or up arrow to jump. Collect sound orbs to create melodies. Land on platforms to trigger beats. Oh, error.
257. 48:55Try to fix. Damn it. Start creating. Oh, didn't work.
258. 49:11Okay, we're facing an issue with the context provider router. Use audio context must be used with an audio
259. 49:21provider. All right, it's it's fixing it. But look, we got we got neon glowy [ __ ] [Laughter] Harmony. See if we can get something
260. 49:35working here. So, does Lovable make apps? Yes, Love Well, Lovable makes web apps. So, this is important. A web app can only run in a
261. 49:47in a web browser. I guess you could probably do some sort of container where you run it on the desktop, but it's
262. 49:51basically just a it's running in a browser. um that doesn't mean it's not valuable and powerful, right? Because there's all
263. 50:03sorts of ways um all sorts of apps that we use are just web apps. Um but if you want to make like a native iPhone app,
264. 50:12you can't do that on it. Or if you want to make a native standalone Mac OS app, OSX app, it's not doing that. So, it's
265. 50:19basically making web apps, which the advantage of that is that it can run the thing it writes right in your browser
266. 50:26while you code it. Web apps only. Yep. Exactly. Can our main character be a turtle? Yes, we'll do that. Start creating. Okay.
267. 50:41Um Okay. Um, the main character should be a turtle. And when I try to move, the main
268. 51:01character disappears. Also, the Um, there are
269. 51:16only dots on a grid and it doesn't feel like a video game. Make the game
270. 51:32play better. Start jumping, Kyle. Yeah. Oh, let me let me reload this and then maybe we can reload. Let me jump first. Start
271. 51:47creating. No, I can move right and left, but then it disappears. The issue seems to be with the character disappearing in the
272. 51:59overall game not feeling engaging enough. Let's fix that. Let's But anyway, like even if this didn't work, like the fact that it like
273. 52:10just understood the game, spun this up, did the splash screen, you know, did all the other stuff, it it's it's just
274. 52:20crazy. And again, what what you need to do is do what Jim Ross did with this is like, you know, just play with it. Try
275. 52:27to make an app for your business or try to make an app for your I don't know recipe organizing what whatever you want to
276. 52:35[ __ ] do your your uh your obsessive compulsive uh seed organizing by time of season for planting whatever. Uh what what champ? What do
277. 52:50you want? Oh, you want to go out because you hear people out there and you want cheese now. Is that what's going on?
278. 52:56Yeah. Yeah, you a cheese eater. Oh, please. We'll let that thing keep writing. Come
279. 53:22on. Still writing. All right, let's go. Let's go over to decco here while we're doing Well, that thing's working. Let's
280. 53:32do some chat add here, shall we? I hear people and I want cheese, too. Nice site, by the way. Um, you can't do anything while it's
281. 53:46working. Yeah, I know it's a drag. Okay, wait. So, I'm going to go to decode here. Not decode here. Descript. vscript
282. 54:02web.dscript.com. And if you haven't seen dscript, dscript is a very weird though very powerful AI video editing tool. What's
283. 54:14weird about it is you upload a video and it transcribes it and you kind of just interact with the script of the, you
284. 54:23know, the transcript of what people say, but they've but they've now added an agent to it that you can just ask it to do [ __ ] and it does
285. 54:35it. Dropping you something in a regular is very nice. Why is this taking so long to load? Let me just go to dscript.com. This was
286. 54:45web.dscript.com. So something's up. Um, well, it's [ __ ] up again.
287. 55:18Reload that. Okay. Well, I I don't even know where to begin. Okay. Turtle is fine. The
288. 55:33background star noise is distracting and not musical. So, let's kill
289. 55:48it. the left and right. Just jump the
290. 56:04turtle to opposite sides of the screen. Up arrow doesn't work.
291. 56:19It doesn't scroll. In fact, it sucks. Make it
292. 56:33playable. So, yeah. Uh, tabs for audio outside of Tik Tok. Yep. Dropped you something in irregulars. All right,
293. 56:40let's go look at irregulars while we're at it. Well, that thing's working. [Music] new activity archetypal architect. What are your
294. 56:55coalifications? Yeah, I see what you did there. You see that? That was for Steo. Um, certified AI consultant Karen,
295. 57:10thank you for sharing the live. Much appreciated. Uh, if you haven't shared the live and want to share it and annoy some people
296. 57:18in your life, that would be a good thing. Now's the time to do it. Um, if you have people in your life, or if
297. 57:26you're one of them that are actively ignoring this AI [ __ ] you got to knock that [ __ ] off. Like, seriously, it's
298. 57:33getting serious now. If your friends are like, "I don't care about AI is evil. I don't know why you like that AI." You
299. 57:40gota they got to get their [ __ ] together. They got to [ __ ] snap out of it because this stuff's coming. And here's the here's you want
300. 57:49to know the dirty little secret about all the bitter people about AI? I promise you, they're going to be the
301. 57:56ones that fall the hardest where they're just like, "Oh my god, it's so amazing. I didn't realize it could do that. I
302. 58:03just thought it was evil." But now that I use it, wow, it's gonna be a lot of that going on. So anyway, yeah.
303. 58:14Yeah, Dennis referred you. Great. Awesome. Welcome, welcome, welcome. So, okay. So, if there are some new folks
304. 58:21here, my name's Kyle Shannon. Uh, this is a neurospicy friendly channel. So, I have lots of uh lots of lots of issues
305. 58:31uh conditions, ADHD being the primary one. Um, I am also hopelessly optimistic and um, uh, encouraged by generative AI and
306. 58:45what's coming. Um, most of the people that are really angry about this in my experience haven't used it. So, there's
307. 58:53a lot of people on soap boxes that are on the sidelines. So, I don't care if you're on a soap box, but if you're on a
308. 59:00soap box and you don't know anything about what these tools actually do and how they work and what the issues really
309. 59:06are, then I don't want to [ __ ] hear it because not for nothing, this is the the kind of purpose of the channel, and I'll get to the AI
310. 59:18salon in a second. The purpose of the channel is not to judge AI one way or the other. My opinion about whether AI is good or
311. 59:31bad or your opinion about whether AI is good or bad is completely independent of the fact that it's here and it's not
312. 59:39going away. So what this channel is about is is the basic premise of the channel is if AI is not going away and
313. 59:47if its capabilities are accelerating which they are, what are we going to do about it? How do you learn? How do you keep up
314. 59:56with it? I've been trying to keep up with it for two years, two and a half years. It's absolutely impossible. You
315. 1:00:02cannot keep up with it. Okay, so what do you do with that? How do you build? What do you do? How do you incorporate this in your
316. 1:00:09business? How do you incorporate this in your job? How do you incorporate it in your family? I don't know. So, what we do
317. 1:00:17here is we just hang out five nights a week. So, I come in generally around 8:00 p.m. Mountain time and I just go for about two hours
318. 1:00:25and just ramble. So, it's like chat add or chat TMZ when we're doing speculation about what's coming. Um, and then sometimes I'll play
319. 1:00:36with stuff, sometimes I'll teach stuff if I know it, sometimes I'll experiment with things and just watch it fail over and over
320. 1:00:44again. It's so sad that it's the 20somes that are the most opposed to it. Well, so here's the thing, archetypal
321. 1:00:50architect. The [ __ ] education system told them that if they used it, they'd be kicked out of school. So So the the Gen Zers and the
322. 1:01:02Gen Alphaers have basically been educated that AI is cheating and so they're not using it. It's It's [ __ ] inexcusable. It's
323. 1:01:16[ __ ] inexcusable. Um, anyway, good for us Gen Xers. Gen Xers are kicking ass with uh and elder millennials are kicking ass with AI
324. 1:01:28because we got to experience what happened with the worldwide web and how that changed everything, right? And now
325. 1:01:36with AI, we're like, "Holy [ __ ] this is a [ __ ] thousand times more powerful than what the worldwide web was." And look what that did.
326. 1:01:45Um, plus we're also just ornery pricks where it's like we just want to [ __ ] some stuff up and AI is the perfect
327. 1:01:55thing to do that, right? It is shattering, you know, it's going to shatter every everything we do. Every
328. 1:02:03way of working is going to be completely upended and transformed. And who gets to benefit from that are the
329. 1:02:11curious. That's what this channel's about. Okay. So, um, go to the AI salon. Uh, Brandon had the address up earlier, the
330. 1:02:21salon.ai. Scroll down a little bit and click on join our community. And that'll take you over to the active community. There's about
331. 1:02:303,000 people that are there that are AI optimists and adventurers and people trying to figure this stuff out. And
332. 1:02:39it's a remarkably like remarkably generous community. And so if you go there, the second step, we ask you to basically,
333. 1:02:49you know, come in, take a look at what we're about. Step two, introduce yourself. So I would strongly encourage
334. 1:02:55you go to the salon and introduce yourself. Um, and if you're like, "Yeah, but I don't know anything about AI and
335. 1:03:02just someone told me to come here and I think you're kind of weird and I don't even know why I'm here, but I'm still
336. 1:03:08here and this is stupid and you don't have any qualifications." Yeah. Yeah. All that. Go in there anyway and
337. 1:03:15introduce yourself and just say that. Just say, "I don't know what I'm doing with AI, but I it seems like something I should do."
338. 1:03:24Great. It's a very welcoming community. Um, and if you know what's going on with AI, get in there and share what you're
339. 1:03:32learning. As you start figuring this stuff out, start sharing it. And in fact, one of the things we talk about in
340. 1:03:41the salon, actually, the first thing we talk about in the salon is what we call the AI readiness cycle, which is play first, mindfully
341. 1:03:50create, generously lead, play, play with AI. Why do I act like an idiot in here? Because everyone's taking AI so [ __ ]
342. 1:04:01seriously. Lighten up, people. Sure, it's gonna decimate the job market, but let's have fun on the [Laughter]
343. 1:04:13way. There's actually some seriousness behind the play first though, and that is this. Everything is moving so fast and there
344. 1:04:21is so much and it is so insanely powerful that you can't right now if you haven't been playing with the this AI
345. 1:04:32stuff. You can't right now even imagine what's now possible that wasn't possible two years ago. Like literally wasn't
346. 1:04:42couldn't do it. It is mindshattering what is now possible. And so one of the best ways to do it is get in a community
347. 1:04:51like the salon and just start playing. And as you're playing, you'll learn some [ __ ] You'll learn, oh, I can make like
348. 1:04:58a really cool song on Sunno and I can make really cool images on MidJourney or Chat GPT. And oh, I can write a
349. 1:05:05marketing plan. I didn't think I could write a marketing plan. It does a whole marketing plan. Well, I bet it can't
350. 1:05:10write social media copy. Oh my god, it wrote social media cut. Like you you will just what we call them your Kevin Mallister
351. 1:05:18moments right from home alone. Have some of those and then mindfully create is once you've learned a couple of things now go
352. 1:05:29oh okay let me try to do this thing for my business or do this thing for my mom or do this thing for my uncle who wants
353. 1:05:35to start a business. Maybe I can do that. Right? That's the mindfully create. try to actually use AI to do
354. 1:05:42something once you play with a pile of tools and then generously lead is about teaching what you're learning. So here's the the constant
355. 1:05:54that I see in the salon is that people consistently undervalue their contribution. So they're like, "Well, I
356. 1:06:06just I I just I'm like really into gardening and like all I did was use chat GPT to come up with an automatic
357. 1:06:12gardening thing and it's not that big a deal." Well, to people that are not into gardening, it's a pretty big [ __ ]
358. 1:06:19deal. It's gardening season. Maybe they want to get out and garden, right? So you teach people what you're learning
359. 1:06:26and then someone else will go, "Oh, I never would have thought about it like that." That's what this community is all
360. 1:06:32about. And as you then share that, people give you feedback. Oh, here's a thing I did. And then you're like, oh,
361. 1:06:40that's inspiring. Let me go try that. You get back to play, right? You just keep doing that. If you keep doing that, you will win.
362. 1:06:49You will win. Whatever the next 5, 10, 20 years is is going to bring is going to require nimleness, curiosity, adventure,
363. 1:07:02adaptability, judgment, curation. It's not going to require taking a class in prompt engineering. It's not going to require I've got to
364. 1:07:13learn Python. It's not going to require I've got to go back to school for music theory. All of that's out the
365. 1:07:22window. Everything that you thought you couldn't do in your life, you either now can do or you will very shortly be able to do
366. 1:07:33anything. Well, I always did want to be a quantum physicist. Okay, you can probably do that now. And like a year from now, you'll have
367. 1:07:46like a research lab of like a thousand PhD level quantum physicists working for you. So you want to prove that the Earth
368. 1:07:56is flat and it actually lives on the head of a pin in some 1870s Taylor shop. Cool. Go prove it.
369. 1:08:13We all get to reinvent who we are and what we do. We all get to We all get to ask ourselves the question. This is the
370. 1:08:20thing I'm most excited about with AI. If I could do anything, like period, if I could do anything, if I could accomplish
371. 1:08:34anything, if I had the skills to do anything, what would I choose to do? what difference do I want to make in the
372. 1:08:43world? And then you can go do that. Now, right now, because we're so early, you have to go play with all these tools and
373. 1:08:49figure all this [ __ ] out. That's where all the opportunity is. So, welcome. I love philosophical nights. Doubletapping on my tablet and my
374. 1:09:01phone. Thank you very much. Um, all right. Where were we? We were at Lovable. Let's see if Harmonic Jumper is back.
375. 1:09:11It already wants to refactor code. I've made several improvements. Start the adventure. Error. Try to fix it. All
376. 1:09:19right, that's broken. Back to dcript. Loading is taking longer than normal. You think I don't know what's up with
377. 1:09:29Dcript. Maybe they're redoing their servers. [Music] [Music]
378. 1:09:47Um descript sign in. Yeah, it seems to be broken. It seems to be broken.
379. 1:10:07Oh, the other thing that I thought would be cool to experiment with, I saved in my little my mind, my little this
380. 1:10:18thing. Yeah, let's go. Let's go look at this thing. Go back to lovable and refactor files. It suggested you didn't read the
381. 1:10:28bottom. Oh, fine. What did it say? The issue was with the function. Oh, okay. Refactor. Can I do both of
382. 1:10:43these? No. All right, it's off. Thank you, Vicki. Vicki Vicky serves as my second brain. Brandon Brandon serves as my um
383. 1:10:58Put the black tab there for the Tik Tok people. [Laughter] You'll have to learn to create video for [Music]
384. 1:11:11starters. What's new in the AI space? Well, good lord. Um, let's see. deep research in OpenAI right now can
385. 1:11:26now export um PDFs, fully formatted PDFs. So, did I is that what this is? No, that was that one. Hang on. Innovative Pathways to
386. 1:11:40Broadway. So if I click share, download as
387. 1:12:00PDF. Let's see what it does if I if it formats it nicely. Launching a new musical and shephering it to Broadway has traditionally been a
388. 1:12:12lengthy, high-cost endeavor. The traditional path versus recent changes. Um, you know what I'm going to tell this thing? I'm going to
389. 1:12:26say, listen, [ __ ] Uh, I'm gonna say, um, um, have you met me? I have ADHD and a wall of
390. 1:12:43text makes my eyes cross. It feels like homework. Use bullets and
391. 1:12:58[Music] uh and uh catchy uh uh techniques to keep me
392. 1:13:17interested. Again, one of the best things you can do with AI is just tell it what you don't like. If it gives you
393. 1:13:23something shitty, just tell it. If Oh, this is a new thing. If you're if you're not paying attention to this stuff, about a month and a half
394. 1:13:31ago, six weeks ago, eight eight weeks ago, um, chat GPT now has long-term memory. So, it now can remember all of
395. 1:13:39your previous chats. So, you can say based on what you know about me, what are some ideas that we've talked about
396. 1:13:45that I never, you know, that I only mentioned once and it'll go find them and bring them back. It'll make some up
397. 1:13:52too, which is frustrating, but you know, um, Runway now has a thing called references, which is pretty flipping
398. 1:14:03cool. Um, Dcript now has an agent that's in beta testing. Um, Gen Spark is [ __ ] crazy. Um, it's kind of like Manis but but
399. 1:14:18slicker. Um, I don't know if it's as robust, but it's pretty slick. Pretty slick. Look, I got tables and bullets for my marketing
400. 1:14:28plan. Yay. Um, now I'm going to say I'm going to turn on deep research again. I'm going to say this looks very top line. Now deeply
401. 1:14:44research exactly what I should do and who I should talk to. Make it a world
402. 1:14:58class plan that will change Broadway forever. All right, so AI is off working for me now. Um, if you haven't played
403. 1:15:14with the 03 model in chat GPT, it's [ __ ] insane. Oh, I've got to answer questions for it. Hang on, please hold, please.
404. 1:15:23one. What's the current state of your musical? Uh script and soundtrack finished two. What kind of support do you already
405. 1:15:38have? uh a co-writer and some interested producers and theaters but nothing
406. 1:15:55concrete. Um three, are there specific cities you plan to launch from like Denver? Are you open to digital first? prefer New York City, but
407. 1:16:09um open to digital first since we use AI as a collaborator
408. 1:16:24already. Oops, I didn't do the other questions. Whatever. Now it's going to go. What were the other questions? Do you want help finding All
409. 1:16:37right. So, that's off now. So, it basically goes into thinking mode and just starts searching the web
410. 1:16:45and thinking and reasoning and doing all of its [ __ ] Speaking of reasoning, how are we doing here? Um, error build
411. 1:17:02unsuccessful. Try to fix it. I probably, for the purposes of this channel, shouldn't have created an ambitious genre
412. 1:17:16smashing music synthesizer that plays like a video game. We we could have gone with a game of Frogger and and know that we'd have
413. 1:17:25gotten somewhere with [Music] it. All right, let's see. What else do we have here? Yes, join the salon.ai. It's free.
414. 1:17:41Lots of people in there to help and share. Yep. It's great people. It's great people. It's also great people in
415. 1:17:48here. If you're if you're new to this channel, whether you're watching it on YouTube or LinkedIn or X or on Tik Tok,
416. 1:17:55um just keep coming back night after night. We do this regularly. Although next week's going to be [ __ ] up. I'm in New York. In New
417. 1:18:04Jersey, New York and New Jersey. I fixed the type mismatch errors by properly handing the note value within the collect orb function. Thank
418. 1:18:16goodness. [Music] Um, let me reload this. Uhoh.
419. 1:18:40Uh, everything flies immediately off screen. Ah, vibe
420. 1:18:56coding. Uh, what kind of video is that? What kind of video is what? what I was just looking
421. 1:19:15at. Yeah, there's there's lots of people that are good with video. You got the high score though. Try preview. Oh, maybe. Yeah, let's try
422. 1:19:24preview. [Laughter]
423. 1:19:46Yeah, she she's done broke. Can't get D from here. I fixed the issue where everything flies off screen. I'll be the judge of
424. 1:19:57that. Yeah, you fixed it. Wait. Okay. Um, you need to establish platformer levels and the turtle needs
425. 1:20:13to start at the beginning and go right. you know, like a
426. 1:20:30game. Uh, next year I'll get AI to make my own smartphone OS. It'll be better than iOS. It can't be worse than iOS
427. 1:20:39right now. iOS has just turned into the biggest pile of dog [ __ ] It's so bad. It really is bad. It's The only
428. 1:20:47thing worse than it is OSX. like Apple has just they apparently Apple needed an [ __ ] CEO um that just was relentless with people
429. 1:21:02like it can't go out until it's brilliant. Um no one's doing that anymore and hence the result. Not to mention it sounded like
430. 1:21:16Uncle Bob after Thanksgiving dinner. run. All right. Thanks, Tim Apple. Exactly. Um,
431. 1:21:32let me go to Let's go to runway. Oh, so this is what I wanted to show you. Oh, here. This is what I wanted to show you. Okay.
432. 1:21:49So there's the image that runway generated. There's the storyboard scene lay layout. Tom, Brad, happy sad. Tom,
433. 1:22:04Brad, happy. I don't know if it's sad, but at least, you know, like he's trying to hold in a fart. Characters Tom and Brad.
434. 1:22:16It's so funny. Every time I see Brad Pitt, I'm like, "Wait a minute. Is that me? Am I looking in the mirror? Shut
435. 1:22:23up." And then here's what they're wearing. So, and then there's what they're wearing. So, like it can do this
436. 1:22:34now. Well, this post claims it can do it. So, let's let's just go ahead and steal We'll just steal their storyboard. Copy
437. 1:22:51image. Oh, no. That's image address. Copy image. And let's go to runway. [Music]
438. 1:23:13[Laughter] Oh, it's so funny. All right. References testing. Wait, where? How do I do
439. 1:23:29new Gen 4 for image and video? Okay, here's our here's our test. So, let's get rid of these things. How do we get rid of them?
440. 1:23:40with that. No. Oh, I see. Okay. Can I paste? Back. Paste. No. Damn
441. 1:23:55it. All right. So, save image as uh scene. Save that to the
442. 1:24:11[Music] desktop.
443. 1:24:25Let's I guess we can go I'm just trying to figure out how much photoshopping I actually want to do. Oh, I guess we'll do some [ __ ]
444. 1:24:37work. It's 9:23. See, this is a this is a [ __ ] rabbit hole. Hi, Champy. What's up? You got your cheese? You back to bug me now? Sit
445. 1:24:50down. Lay down. Lay down. Do you want to sing again? You want to sing again? You're wagging your tail. You want to sing? Ready?
446. 1:25:04[Music] Well, I heard there was a secret
447. 1:25:21call David played and it alone. You don't really care for music, do you? It goes like this. The fourth, the
448. 1:25:35fifth minor fall in a major baffing composing. Hallelujah. Hallelujah.
449. 1:25:49Hallelujah. You're not into it. You were just faking me out. Go lay down. Go lay down, you big fat soul. That dog is fat.
450. 1:25:57I don't know if you knew that. That dog is large, Marge. Okay. Yes, I'm fat shaming my dog. Um, I grew up in a time
451. 1:26:07where that was allowed. Okay. All [Laughter] right. You people you people with your your treating people with dignity. I
452. 1:26:19don't know how to do that. Okay. All right. Scene layout's going to be that. Let's go grab another picture.
453. 1:26:33We're going to go back over here. We're going to go grab and we'll grab this [Music]
454. 1:26:50one clothes. Let's get rid of Tom and Brad. And then maybe we'll go find some cooler clothes. All right. Who should we go?
455. 1:27:04Who should we put in this? Let's put Ann Murphy. Hang on. If since we're since Since we're already bismerching her reputation with
456. 1:27:18AI, let's keep that tradition up. She's on vacation, so she can't defend herself unless they have Wi-Fi at the fancy
457. 1:27:26resort she's going to. Okay. Um, where did I have her picture last? I had it Oh, I had it in in uh in runway, didn't I? We can go get it
458. 1:27:41there. Copy image. All right. Paste. Oh, it's too small. That's all right. We can scale that [ __ ] up.
459. 1:27:55All right, Tom. No. All right, Ann Murphy. Boom. Done. We could put her with Brad Pit, but that seems that seems
460. 1:28:04too It's not appropriate. It's just not appropriate. I think who she needs to be with. Let's see. Let's see who would be good. She's
461. 1:28:24tall. Any ideas? She's smoking right now. William Defoe. Okay, that's good. Silver Fox. Silver Fox is always there with the
462. 1:28:35brilliant William Defoe images. And we got to get we got to get the right William Defoe. It can't be the when he was handsome.
463. 1:28:48No, no, that's not the uh we won't we won't make him that ugly,
464. 1:29:03but let's see which which William Defoe is the question. It's got to be one where he was out on a bender. Let's see. William Defoe.
465. 1:29:17Paparazzi. Let's get him when he looks like [ __ ] Come on. Oh, that's is that William Defo?
466. 1:29:31That looks like Dennis Liry or Dennis. Yeah, Larry. William Defoe. All right. Bearded William Defoe. This is good.
467. 1:29:46Right. Oh, for [ __ ] sake. Everything's a [ __ ] salesfunnel. There we go. All right. So, here's William
468. 1:29:59Defoe. This is the one we're going to use. This is This is who Ann's going to be in her little scene with.
469. 1:30:12Hey, if any of you tell her that I'm doing this, that would be inappropriate and rude. Would be rude of you to do [Music]
470. 1:30:22that. All right. Now, we go back over to photo slop. Paste that bad boy in there. Let's chop off part of
471. 1:30:37that. All right. So, we cover up All right. So, now we're going to change some Wait, I got to go to this layer. I
472. 1:30:49got to pull this [ __ ] out. Kill it. Oh my god, this is exhausting. You see why I didn't want to do this? Everything's a [ __ ] production.
473. 1:31:03And And then this is We're going to call him uh Billy. Okay. So, we got Ann and Billy. Now, let's So, what else do we need? We need
474. 1:31:18their clothes. And then, what was the other thing? We had the layout. All right. So, we're going to make Ann sad and Billy
475. 1:31:28happy. Okay. So, this is going to be Ann and this is going to be Billy. All right. So, we got that. That's good.
476. 1:31:42Now, we need clothing. All right. Back to the Google. Um, let's see. Uh, co dress. Let's get something fundamentally
477. 1:31:58ridiculous for an um Oh, well, let's see.
478. 1:32:13There we go. Is that good? Yeah, that's
479. 1:32:29good. All right. There's There's Ann. We got to resize that bad boy. Is Is anyone still here? If you're wondering what this channel's all
480. 1:32:46about and why anyone watches it, I don't know either. Okay. Um, and then what are we going to put Billy in? I think Billy needs to go
481. 1:32:57in like a gas station. Uh, let's see. Um, gas station. Uh, jumper uniform. Yeah. Here we go. That's good
482. 1:33:11right there. Right. Yeah. That looks like Billy Defoe. [Laughter]
483. 1:33:27Uh, if one of the questions you're asking yourself is, is this channel really just an opportunity for Kyle to
484. 1:33:35entertain himself? Yes, that is exactly what this channel is. Okay, so now we need some letters. So this is Billy
485. 1:33:51and this is Ann. Okay. And then did we need one other thing? I think we might have needed one other thing. Oh no, that's just the output.
486. 1:34:03Okay. So we've got our three things, right? Okay. Beautiful. Back to Photoshop. So we're going to do export as PNG. So this is clothes.
487. 1:34:15to the desktop clothes. Um, if you're new here, um, pro file management tip. Save all of your files loose to the
488. 1:34:25desktop with completely generic names you'll never be able to find again. You're welcome. All right. Export quick PNG. This is
489. 1:34:37characters. All right. Fantastic. And then uh and then this is scene layout export quick export is so let's call
490. 1:34:54this scene. Okay, back to runway. Anybody have any questions? Have I lost anyone? Is anyone still here?
491. 1:35:07This is why the irregulars show up night after night to defend our reputations. Yeah, exactly. You go on
492. 1:35:15vacation as an irregular, this shit's going to happen to you. So, yeah, get over it. All right, here we go. So, we're going to upload
493. 1:35:28desktop scene character clothes. Then we're gonna say an
494. 1:35:43and Billy sit in a horserawn carriage in
495. 1:35:57Central Park, New York City. Uh, I guess I'll just say
496. 1:36:14follow the image guidance, I guess. All right, we're going. We're off. Generative AI. Yes, that's what
497. 1:36:30we're doing here. So, this channel is primarily about generative AI, not about weights and embeddings and
498. 1:36:38math. Not the guy. Um, there are people in here who are really good at that. I'm not the guy though. I talk about
499. 1:36:46generative AI. How are we using these tools? Why did that not generate? What's going on? Oh, generating. Here we go.
500. 1:36:53Oh, okay. We're gonna have two sets. And B, what's happening? Good to see you. Welcome. Welcome. Welcome. [Music]
501. 1:37:09All right, these are almost done. Let's see how lovely they
502. 1:37:26are. Damn. Come on. Come on, runway. You can do
503. 1:37:45it. Oh my god. It's too perfect. It's too perfect. Is the processing time based on their server or yours? It's all
504. 1:37:53theirs. I'm not doing anything locally. So, so on this channel, I tend to just ignore open source. Not because I don't
505. 1:38:02love it and think it's important. I just don't have the attention span for it, the attention to detail for it, the hard
506. 1:38:09drive space for it, the computer, the computational resources for it. Um, and I also just feel like, you know, about
507. 1:38:17the time I'll get something up and running, it'll just have moved on to the next technology. So I just basically
508. 1:38:23what I do is I focus on the commercial commercial big players like who are the ones that are the the least waste of
509. 1:38:30anyone's time and that's what I focus on. Um the fact that we have this level of of coherence right now is is actually
510. 1:38:39kind of mind-blowing to me. Um, look at Ann. Oh, sad. Frozen. Um, why are they sitting so far
511. 1:38:53apart? Cuz she wore her dress and he came out in a [ __ ] Look, he looks giant. She looks
512. 1:39:08teeny. Oh my god, this is good. This is so good. Okay, there's one where she's really pouting. Let's see
513. 1:39:23that it's sitting really far apart. Although the carriage looks [ __ ] up there. Let's see there. She's pretty pouty. All
514. 1:39:37right. Okay. So, now we're gonna make a movie. Hey, Joy, if Joy, have you How long have you been here? Have you seen
515. 1:39:47this? Oh, I'm buffering. Oh, that sucks. Leave her alone. Is this generated locally? No, this is generated uh this is RunwayML. So,
516. 1:39:57app.runwayml.com. They have this new thing called references, and you can put references in that do things like put
517. 1:40:04her on the right, him on the left, her sad, him happy. Um, just got here. Okay. So, so here here's what we
518. 1:40:14did. So, I went I went on Twitter and someone someone had this with Tom Cruz and Brad Pitt. Um, and so here's the scene layout, right? So,
519. 1:40:25the person on the left is happy, the person on the right is sad. Um, it's Billy on the left and on the right. So,
520. 1:40:31I did that. Then I said, you know, here's what the people are wearing, right? Billy's wearing the mobile gas
521. 1:40:39shirt and she's wearing this big ass fancy dress. And then, you know, it's William Defoe and Anne Murphy are the sources. All right.
522. 1:40:50And then we went, it just [ __ ] did it. So Joy, for you in particular, the character character coherence here is
523. 1:41:04insane. Uh oh my god, this is hilarious and awesome, isn't it? Well, now we're going to make it into a
524. 1:41:10movie. Let's see. So we're gonna go All right. Um camera tracks with carriage as
525. 1:41:27it rolls through Central Park.
526. 1:41:43The man is laughing and the woman is [Laughter]
527. 1:42:01pouting. So naughty and fun. Dragonfly alchemy. What does Ann have to say about it? She's probably crying, laughing. She's on vacation. So,
528. 1:42:13so what I what I inadvertently did, so when I first started testing um this this reference thing, um I did it first
529. 1:42:26in Midjourney. And in midJourney, it kind of sucks. It's not really good. And so then I came to runway and I just I
530. 1:42:35just had used I had a picture of Ann from when we were setting up the podcast. And so I just grabbed it and threw it into into
531. 1:42:43um into runway with two two photos I've taken. One of a photo of clouds that I took and one of this motel. And so I
532. 1:42:52put, you know, the clouds, the motel and her in the picture. And if [ __ ] came up with that and then we turned it into
533. 1:42:58a video and it made this. Wait, hang on. This is [ __ ] hilarious. [Laughter]
534. 1:43:17Now, that might seem random, right? And it might seem like glitchy, like no one can hold that much smoke in their lungs,
535. 1:43:24but it's like, you know, have you ever partied with her? You don't know. Maybe she maybe she's she's a toker. Um, all
536. 1:43:32that white smoke. The next day we had a new pope. I'm not saying Anne had anything to do with it. I'm just saying she blew
537. 1:43:44white smoke from her cigarette, got a new pope the next day. All right, new pope. All right, let's see if we have a
538. 1:43:55video yet. 61%. We're almost there. It's going to be so good. Did Lovable make our game yet? our game thing. Nope. Oh, wait. No. Ah, damn
539. 1:44:12it. Got nothing, [Music] people. Why are they sitting so far apart?
540. 1:44:28Oh, lordy, lordy, lordy. I think I'm going to text her [Laughter]
541. 1:44:44this. Oh, it's so good. I'm a bad man.
542. 1:45:03Uh oh, I'm not charging my phone and my phone's about to die. That would be awkward. He didn't even say
543. 1:45:19goodbye. All right, here we go. Here we go, people. Do we have Do we have comedy gold AI weirdness? What is
544. 1:45:36it? It's perfect. It's perfect.
545. 1:45:55Oh my god. Oh my god. Okay, we got to download this. Oh god. Oh, it's so good. It's so
546. 1:46:13good. All right. Can we Does it allow Oh, restyle. That's interesting. Use frame. Oh, interesting. All right, let's go here.
547. 1:46:28We'll go here. Oh, yeah. That's Oh, you can't see any of that. Oh, could you not see any of that? Hang on. Billy has a
548. 1:46:37comment. Yeah, I know. Billy said something at the end there, didn't he? Um, hang on a sec. Let's see. Maybe Ann. Okay.
549. 1:46:56Okay, this is good. I'm just what you can't see right now is I am texting our [Laughter]
550. 1:47:16friend here and I'll say I'll put the message. Oops, it happened again. I didn't know you and William were so
551. 1:47:35close. Ah, that's comedy right there, people. I am telling you that is comedy gold. Comedy go. I know. Oh my god. So
552. 1:47:45funny. We still can't see. She may not be your friend for long if you keep this up. It's so good. It's so
553. 1:48:00good. Okay, so listen. I said this before. We'll just keep watching this because it's awesome. Um, I said this before when I was
554. 1:48:12talking about the AI salon. If you're new here, I said play first. Remember when I said play first? No, I don't
555. 1:48:19remember that. When did you say play first? I wasn't here. So, how could I have heard that if I wasn't here?
556. 1:48:26Um, as stupid as this was, as silly as this was, we learned something really important tonight. Like, Joy Perie is a full-on filmmaker
557. 1:48:39and now knows that this is possible. Like, this is a level of control that we haven't had an AI before. And it's kind of crazy, right?
558. 1:48:51Um, I wouldn't have gotten to this if I wasn't just [ __ ] around. Like, I wouldn't I wouldn't know how to do this
559. 1:48:59if it was any good, right? You know, there's something about having someone that you know, like I I know enough what
560. 1:49:06she looks like. She looks She at least starts out looking like her. She kind of morphs into someone else by the end. But that's bonkers.
561. 1:49:19This is the stage we're at. I could probably you could you could probably put this into another
562. 1:49:37um lip syncing tool and give him dialogue. I've had students wanting consistent characters all year. There
563. 1:49:47you go. Well, they exist. It It exists now. The guy following them has has some rhythm. Oh, yeah. He does, doesn't
564. 1:49:57he? I think he he looks like he recognizes William Defo. He's paparazzi. This guy back here. That's [Laughter]
565. 1:50:06great. Billy cheers her up at the end. Yeah, she does. She gets a little happier at the end. She can't help him. He's
566. 1:50:13lovable. I'm going to do this, Kyle. Yeah. Like like this is crazy. This is crazy. Zoom in on the guy following. Can
567. 1:50:21I zoom in? I don't think I can. Hang on. Maybe I can here. Um, hang on.
568. 1:50:36There's the guy following upscale enhance. Exactly. Oh, man. All right, good
569. 1:50:51people. Uh, it is late. Oh, 9:49. Now, we're we're we're sort of right on time. You could totally out add a soundtrack. Yeah, you could do
570. 1:51:02soundtrack, you can do music, you can put dialogue in there. Again, again, people, people, people. If you
571. 1:51:16think that any job is safe from transformation, you're wrong. Like, this stuff has gotten so good in the past six months. like all of
572. 1:51:31it, large language models, coding tools, video tools, image tools, like the image coherence within Chat GPT at this point is
573. 1:51:40insane. But this is image generation within Runway ML, which is a video model, but this image coherence, we've
574. 1:51:47never had this before where you could put in um like layouts of how you want to arrange things. It's just bonkers.
575. 1:52:03just bonkers. Anyway, all right. Maybe that's Billy's
576. 1:52:18agent. Billy, I prefer William. Yeah, whatever. Billy, listen. What did you What did you do to piss her off so [Laughter]
577. 1:52:33much? All right, Kak's off to bed. Good night. I'm gonna get off to bed, too. Um, tomorrow's Tuesday. I don't think we
578. 1:52:41have anything tomorrow, do we? Tomorrow night? No. Um, so we'll be here tomorrow, same time, same bat time, 8:00 PM.
579. 1:52:55Um, if you're on YouTube, like and subscribe. We're trying to build our user base over there, our subscription
580. 1:53:02base over there in case Tik Tok flies the coupe, which it is, and it isn't, and then it is, and then it isn't. So,
581. 1:53:10um, go subscribe over there so you can not miss You wouldn't want to miss a minute of one of these, would you? No.
582. 1:53:16All right, cool, Groovy. All right, peace out, everyone. Have a good Monday night and I will see you