1/10/2025 - Initiating Creativity: Video Generation and Sound Design Techniques

Transcript

1. 0:00down what's [Music]
2. 0:18[Music]
3. 0:32[Music]
4. 0:50Brandon said he just made it was happening was shaking all right we are live let me make sure yes we're live
5. 1:02turn that off look at that there I am looking look I took time to to make my Friday night date night hair all pretty
6. 1:11for y'all happy Friday night date night [Music]
7. 1:34Source Camp does champy sing all the time or just for for the Irregulars he sings Anytime a guitar is played or it's on
8. 1:42TV and also um brass bands playing uh national anthems at the end of f or the beginning of F1 races he sings to those
9. 1:51for some reason um your ASMR video is a masterpiece thank you very much I uh I let some of my bad behavior on these
10. 2:00lives Escape into the real [Music] world every time I'll see it now yeah they looking
11. 2:15mad every time I see your mou that [Music] smiley morning I the engine tur
12. 2:30the [Music] old you leing again
13. 2:44today you convince me again today you're leaving this H to looking for someone else's Golden Ring
14. 2:59[Music] long now don't you
15. 3:17cry so long [Music] suan don't you cry for me [Music]
16. 3:40champy doesn't seem very into it he's he's not he's not all that excited I I don't know what you people did to him I don't know why he
17. 3:49is resisting why are you resisting what is wrong with you [Music] there you
18. 4:04[Music] go there's been something baby I've been trying to [Music] say for an age and it seems I don't know
19. 4:19[Music] how past and a future now surrounding me surrender to whatever can be found there's been little
20. 4:36trouble s came to my [Music] rescue and if you like all of the rest I would have quit you long ago but I
21. 4:50couldn't do [Music] that will and why I never went too well make a man crazy and make him cold
22. 5:07as hell champ just walked away champ champ is the ultimate Arbiter of taste he was just like NOP no we're good I'm
23. 5:17just gonna go I'm gonna just go fart and lay on the [Laughter] bed steo hey long time [Music]
24. 5:41[Applause] [Music]
25. 6:06good evening good people if you're new here to the channel my name is Kyle Shannon this is the AI learning lab we
26. 6:11learn here we lab here we it's not really a lab it's a$2 million uh TV production studio uh we have we have
27. 6:21spent uh a ridiculous amount of money trying to get it to look like a shitty home office um you know you would think by looking at
28. 6:33it that I've got just like a you know $16 LED light on a cardboard box that's lighting me but no it it is uh we've got 72 studio
29. 6:46lights uh varying degrees of color temperature brightness um refresh cycle so we don't get the strobing effect it's a very
30. 6:55sophisticated operation made to look like a janky piece of [ __ ] you know you know it's it's funny one of the
31. 7:05things people come into the AI learning lab and they're like but Kyle what are your qualifications and I have this
32. 7:11whole joke about not being qualified and I have a degree from from Penn State and acting it's it's all just part of the
33. 7:17act right I'm a uh I'm a nuclear physicist by training um I was not credited on on the Transformers uh white paper the
34. 7:28attention is all you need white paper paper uh but that was my idea that was my idea and you know again jealousy
35. 7:35Reigns in the worlds of of high mathematics um and for me having kind of I don't know you know what you would
36. 7:45call genetically gifted good looks along with you know uh just a ridiculous IQ and and training the likes of which the
37. 7:55world hasn't seen people get jealous people get jealous um and so I'm not bitter I'm not bitter not at [Music]
38. 8:10all um all right good day good day good day good day to you then um what are we gonna do tonight I think I think we're gonna play with
39. 8:22Korea Korea just added okay last night there was a there was a whole situation Danielle came in here all bothered she's
40. 8:30like she's like Kyle Minx has sound now and I went to Minx and it didn't have sound and then she was like uh I don't
41. 8:38one of the other ones has sound now Halo maybe that's Mini Max I don't know and that that that didn't have sound I found out which ones got
42. 8:47sound so the drama the drama we've handled all right a doctorate from eju it turned jany University
43. 9:02exactly exactly oh man um I hope you're all well if you're in California good Lord what a horrible horrible thing happening as we
44. 9:14speak um I hope you're okay I hope you're safe um you know uh it's just it's heartbreaking it's heartbreaking I
45. 9:23can't I can't I can't process what's happening um you know we had a little bit of it here a a mini version of that
46. 9:31here in Colorado that high winds and fire just blew through a town and uh absolutely terrifying and it's just it just goes on so
47. 9:43anyway I hope if you're watching here this gives you a little bit of uh a little bit of uh escape from the from the insanity [Music]
48. 10:10Danielle I can't help but if I have a three second memories memory span me too I'm right there with you I think we have that in
49. 10:18common well you know what's funny is I was just as I was coming on tonight I I went to Korea not because of the sound
50. 10:27thing but because you could there's mult multiple models video models that you can test within Korea and then when I
51. 10:33logged into it it said oh we do sound now I was like oh there's that so and I actually found it the interface for it is
52. 10:41buried uh in fact let me go I'm gonna go add sound to this video I just made uh hi Winston I'm not the jealous type
53. 10:54it's difficult not to be jealous of a guy like you exactly I understand listen it's something I've come to accept in my
54. 11:00life that that just there's just going to be there's just going to be challenges with people comparing
55. 11:07themselves to me and you know the Grass Is Always Greener glass is always more full on my side of the equation I'm the
56. 11:14one with the gold button you know exactly exactly all right CW I want to know if you asked an AI to craft you an AI would it be a new AI
57. 11:32or would it [Music] operate [Music]
58. 11:46well if you gave it new data to train on I mean the what makes the AI what makes the AI models what they
59. 11:57are is their training and the data and the data that they're given um what I would say to you is [Music]
60. 12:08this I think it was Elon Musk said it this week he said we have effectively used all of the data of humanity like we have effectively any
61. 12:23data that we can put into a large language model has essentially been put into them which if you think about it is
62. 12:33mindboggling right the collective intelligence of the history of human Humanity has been jammed into this this
63. 12:43box this AI box right they open up the box they take all the data in the world and they jam it in this thing and then
64. 12:51they close it up and this is your large language model um so what what a lot of what a lot of people what a lot of
65. 13:03research labs are doing right now is they're using synthetic data where they're having the large line th this is
66. 13:10actually fascinating there's David Shapiro did some really good videos on this which you can probably find on his channel
67. 13:17um where they basically take how did he call it he called it activating the latent space you basically ask a large language model for
68. 13:27for some topic let's say you want to I don't know orthopods I I don't even know if an orthopod is a thing but let's
69. 13:33assume an orthopod is a thing orthopods o orthopods i it doesn't [ __ ] matter um rly poly bugs and you just have it spit out all
70. 13:45of the data that it's got on roly poly bugs right all of it as much as it can just vomit it out and then they have it
71. 13:52look at all that data and analyze it and say which of this is actually good and then they organize it and they tag it
72. 14:01and they put it into the optimal format for training a large language model so so they're basically taking the
73. 14:10raw vomit of the latent space of just like the tokens around a given you know semantic area and then they do some
74. 14:19analysis on it using another large language model like a reasoning large language model so so open a right now is likely
75. 14:30using 03 to generate synthetic data to train the next GPT 5 or something like that so in that sense they're using
76. 14:40multiple AI models to create a new AI model and yeah that new AI model will be a new thing and it will be better
77. 14:47because the synthetic data is so pure and so like it's so distilled into I don't know the best data they can have which is amazing [Music]
78. 15:05there's always new data there is but in terms of historical data we're pretty much done but that doesn't mean there's not
79. 15:21new data right synthetic data doesn't mean it's not new data and again if you think if you're thinking about computers as they
80. 15:32traditionally worked if you're thinking about like a database and you're saying oh well you pulled stuff out of a
81. 15:39database then that's already data we've already used no that's not how large language models work the G and GPT is
82. 15:46generating you're generating new stuff out of these fragments of words that are in Thousand dimensional mathematical space
83. 16:00it's weird man that's pretty cool this Korea thing is pretty [Music]
84. 16:19cool oh good lord good Lord good Lord good people good people good people so lesson lessen you and lasses um here's how this
85. 16:35works it's it's just absolute [ __ ] chaos here I might just talk for 90 minutes um I could show something cool I
86. 16:44might take the guitar off my neck the dog might come back uh but if you have questions here's the this is one of the magical things about
87. 16:55this and by the way it looks like Tik tock be what do we have today's the uh the 10th nine days Tik Tock maybe go in the Wayside
88. 17:08but we'll see we'll see what the Supreme Court says Define data well data that you train large language models with is is very
89. 17:18particular you've got to have tag data right so you've gotta you basically say here's a bunch of data and then you have
90. 17:24metadata that tells the large language model what that is and then it somehow figures out how to break all of so so
91. 17:33let's say you put in a 100 page book and you tag it with the appropriate tags that says what it is and categories
92. 17:41and all that sort of [ __ ] it takes all of the words of that book and it breaks the words and punctuation and spaces
93. 17:50into things called tokens and then turns them from fragments of words into numbers and then it grp groups those
94. 17:59numbers in literally thousand plus dimensional space in in semantic clusters and you're like what the [ __ ] is a semantic cluster
95. 18:12exactly like the the mathematicians that figured this [ __ ] out and continue to optimize this pate who's a fan of the
96. 18:20channel and irregular here he he actually does this for Google he works on optimizing inferences using tensors
97. 18:29um so there's people that all they do is make this math better but what it does is it like so like in this book let's
98. 18:36say in this 100 page book that you've got you have the word dog right and dog shows up in multiple places and in one place in the book
99. 18:46they're referring to the dog is he came home after A Hard Day's Work and the dog was there with his wagging tail ready to
100. 18:53sing a song well that version of the word dog is going to go into a semantic cluster of like pets and dog as animal
101. 19:03and companion and lovely little thing and then let's say in another part of the book they're like ah he he instead
102. 19:11of you know having Integrity he ghosted you know the woman and he's a dog because she thinks he's a dog because of
103. 19:18that well then that version of dog goes into a different semantic cluster but it's not the word itself it's a number
104. 19:26that represents the word or the fragment of the word and it's just it's [ __ ] the the more you understand about how large language models
105. 19:35work the more I'm convinced that they shouldn't [ __ ] work but they do they're they're like a little miracle of mathematics
106. 19:46so all your data belong to us exactly [Music]
107. 20:08pate's here what's happening Pate Jason Wolf Man Clint Tick Tock is technically all semantic we need a lot of decentralized apps and
108. 20:21websites I don't know what we need you know it's funny we were talking today on office hours um about agents and
109. 20:32and you know one of one of one of pate's favorite things on this channel is when I speculate um but what we did on office
110. 20:40hours today was I actually asked people to speculate I said if you assume that we're going to get agents you know
111. 20:47sometime in the next year and that agents are these things that can take actions in independent of us right we
112. 20:54can give them goals and we can direct them but they go off and just do this [ __ ] where we don't need to know how to prompt them
113. 21:02right someone did like someone created the agent we may have to set up agents but once they're set up they just go do
114. 21:09[ __ ] and they could go do [ __ ] like go write a marketing plan and then some other group of Agents could go write
115. 21:16content for that marketing plan and then another group of Agents could come up with media buying strategies and then
116. 21:22another group of Agents could go negotiate good media buying deals and then another group of Agents could write
117. 21:29all the content Source all the content generate the images assemble the things and generate a whole bunch of content
118. 21:37for the media plan and then the media plan gets executed and then there's another bunch of agents that test all of that content and optimize
119. 21:45it and you don't have to do [ __ ] right like that's the promise of it now the reality of it is we don't have
120. 21:55agents yet we've got you can build your own with some of the you know agent tools within the llms and apis and Lang
121. 22:02chain and [ __ ] like that um and there's some people doing interesting work in that but we don't have you know we don't
122. 22:08have open AI or anthropics or meta agent framework that's going to make it really easy to do these but assuming that that
123. 22:22happens what is our role what is our role and it it it was a fascinating conversation and it you know it it it really got to things like
124. 22:37um you know our role might be to do things like support one another and figure out how to use agents to support
125. 22:46people that were displaced because of agents and then maybe those agents can help those people get up on their feet
126. 22:53and start their own business or or do whatever they might do um and the thing that struck me is that as a business
127. 23:03owner I think anyone running a business right now we are we are about to enter a gauntlet that's a [ __ ] Gauntlet for probably the next
128. 23:19decade where things are going to change every quarter but maybe even every month that how you ran your
129. 23:27business today is going to be different than how you run it 3 months from now or 6 months from now and what you do and
130. 23:36what services your people provide the business is all of it's going to change so I feel like there's going to be there
131. 23:43what's going to emerges two different kinds of businesses one is going to be a business that essentially exists today
132. 23:50every business which basically says let's see I'll have my agent meet your AG agent they can have a power lunch
133. 24:00exactly the businesses of today kind of deal with change like okay we had an economic downturn and that's affected
134. 24:08our business and then we're going to put in some change management strategy and then over the next you know six months
135. 24:14we're going to do we're going to implement the change management strategy and and then for the last six months of
136. 24:20the year we're going to run the business and then we we should be in good shape that's not going to work all that well
137. 24:29in an environment where what's possible for for businesses to make money changes on a regular basis if we're indeed moving into this
138. 24:40accelerating you know exponential capability curve there's going to be a new kind of business that is constantly adapting and as as an
139. 24:54individual if you were someone who likes things in stasis you like to plan things you like things to be exactly as
140. 25:04predicted I think there's going to be a a just a [ __ ] ton of chaos coming so anything any of us can do to get in that
141. 25:12mindset of adapt adapt adapt get curious get adventurous get curious adapt adapt adapt I think that's going to
142. 25:22be the ones that succeed in the next 3 to five to 10 years are going to be ones that are Nimble and willing to adapt and
143. 25:33in and in fact aggressive to adapt so it's it's wild blessed are the flexible indeed age of disruption is kind of a an
144. 25:46understatement I think [Music]
145. 26:05our roll Is How We Roll pretty much that pretty much that [Music]
146. 26:21[Applause] [Music] uh oh Brandon's getting bored Brandon's like Korea could we do something here
147. 26:35buddy are you just gonna sit there knocking on your guitar for for an hour people are the people are getting
148. 26:42restless all right fine fine fine coming coming up on 30 minute hello oh you people are good you're the
149. 26:55best you're all the best happy Friday night date night by the way hope you're doing well okay let's get going let's go do some
150. 27:13stuff um you won't be able to hear this but I just made that earlier um I want to go try something though so last
151. 27:30night last night we went to LTX studio and we had these
152. 27:44[Music] cool I'm gonna take this to Korea okay so in Korea we're going to describe seem to generate so so I'm at I'm at ka.
153. 28:03aapps and this is their video generation app so if I go out to the to the main page there there's okay
154. 28:11so so if you've got Crea um you can make video here you can make images here as well you can also do real time images
155. 28:20here maybe we'll play with some of these things um right here it says Ultra fast video Generation Now with sound and
156. 28:28you're like oh cool and if you click on this it takes you to Twitter and shows you effectively a press release
157. 28:37which like tell me it's the story of that you've got audio like this should take me too how do you [ __ ] make the
158. 28:45audio and if you scroll down here there's nowhere where it says you know video to audio however there's this
159. 28:52little blue button right here can anyone see this not really um this little this little blue button says new models and video to
160. 29:02audio so if we go into here here's our videos we've made on the left here's our little scene we're going to generate here we're going to go
161. 29:12720p the the model down on the bottom here um cling 1.5 is one I just did they've got runway in here they've got
162. 29:22Luma in here Luma by the way is coming out with a new model I think I'm going to get early access to it as a uh as a
163. 29:29creative partner so I'll let you know that um the 01 live model high quality model for animating people I don't I
164. 29:38have not played with that halou this is the minia max one which is really good um oh this is cool it also tells
165. 29:47you how long these take cing takes five minutes Runway Luma take one Hunan takes one and Halo takes 10 all right so let's we'll stick with
166. 29:59Hunan and then we'll drop in let's see uh let's see uh in the style of Pixar
167. 30:16Animation a wide view of a cozy garage showc casing scattered tools blueprints work prints of of the vibrant purple Beetle we'll say 196 69
168. 30:291969 VW Beetle is parked prominently in the center glistening under the soft fluorescent lights with eyes behind the wind
169. 30:47shield all right let's see let's see how this does this ca. aai khaki Fitz Simmons hey Kyle a sister up with Early Access if I get it I'll
170. 31:00definitely be showing it here um okay here's one of my strong strong strong recommendations if you want to be doing
171. 31:10video for every video tool out there go find their creative partner program and um and uh and apply for it because they're all too expensive
172. 31:25they're all too too expensive they just are Jank shank all right let me grab oh so wait
173. 31:39so what we could do here I could go to a different let's go to a different one minute model we'll do Luma Luma kind of
174. 31:46sucks right now let's do runway and we'll generate one with Runway oh wait did I just lose one I think I
175. 32:04did huh that's shitty in her face but maybe I didn't that's the way these things work is
176. 32:21weird now I think I lost something dang it all right
177. 32:35bang copied go back here go 720p go bang oh all right so now okay so there's our
178. 32:51video so let me change I'm going to I'm going to switch modes here for a second okay so how you do this is underneath
179. 33:09the video you can reuse the parameters you can copy the prompt you can publish or you can add sound so this is the
180. 33:17little button where we can add sound all right so here we go bang that's a cute little 69 Beetle isn't it
181. 33:32the eyes are creepy it kind of made them the the other model was better but here's the [Music] sound well that's kind of cool
182. 33:47[Music] music uh try [Music]
183. 34:03[Music] [Applause]
184. 34:18oh wait
185. 34:34hey Quinn how's it going hey Kyle I'm doing well how about you doing good um listen you know it still kind of freaks me out that you can
186. 34:44talk this good and this fast and I kind of feel like I'm in a Sci-Fi movie but you know here we are it's wild right it's like you want
187. 34:57to dive into today or just chatting no I want to dive into something um I want so there's a new text to video model out
188. 35:07that has um you you generate the video and then it generates a sound for you um of whatever's going on in that video
189. 35:17right so it creates it creates special effects sound you know sfx for you which is pretty cool so what I want you to
190. 35:22help me do is I want you to help me come up with um and and they're 5c Clips so these are short little Clips so I just
191. 35:29want your concept of a of let's say 20 different video clips and those video clips I want you to just describe write
192. 35:39a prompt right and the prompt will be for a textto video generator where it'll include a subject lighting uh location
193. 35:49uh cinematic style maybe film stock whether it's animation or black and white or things like that so I want them
194. 35:56to be all sorts of varied things things and then on top of that I want them to be Concepts that it's really clear what
195. 36:02the sound effect should be for each of those ideas okay sure I can do that let's get started here are 20 prompts prompt a
196. 36:14lone Lighthouse standing on a rocky Cliff during a stormy night cinematic style Moody and dark lighting flashes of
197. 36:21lightning film stock black and white sound effect crashing waves and thunder prompt a bus
198. 36:36yeah see I lost all that dang it hey do you still have those 20 ideas Quinn are you there did I lose
199. 36:52you all right so so here's here's what I've learned about doing um automated voice to text is um if you don't let her
200. 37:04say all the ideas then you don't get them typed out so so let me start over and my apologies to all of you but
201. 37:14but okay if you haven't used text to voice before one of the things that large language models are remarkable at
202. 37:24is taking unstructured data and making sense of it so you heard my words right I am not exactly a prompt engineer
203. 37:33Maestro here I'm literally just going uh I want you to come up with ideas to do this and that and then I kept add adding
204. 37:42crap on top of them and she just completely took that in and got got it like that so I'm gonna start over and do
205. 37:47it again but take notice of that that you don't need to be nearly as pres prescriptive as you might think you need
206. 37:55to be with these large language models they're quite remarkable of making sense of your you know logic challenged sentence
207. 38:05fragments especially if you're an American because you cut her off I know I did that's exactly why okay Quinn how's it
208. 38:19going hey Kyle it's going great how about you how's your day going I'm I'm doing good uh we just talked about this
209. 38:26before but I cut you off and then we lost uh what you created so here's here's what I want you to do I'm working
210. 38:31with a um a text to video uh image a video generation model that also has um video to audio generation so what I want
211. 38:43you to do is I want you to write me 10 super creative um video prompts for the textto video portion that you know have things
212. 38:53like a really interesting subject an interesting location interesting lighting camera Styles cinematic Styles
213. 39:00whether it's black and white or cinematic or animated or Pixar whatever it might be um and then I want you to
214. 39:08make sure that the subject or the location H has a really clear sound associated with it so when we put this
215. 39:16video into the video to audio generator it will make us really good sound effects make sense got it here are 10 super creative
216. 39:27prompts for the textt to video portion subject a vintage70s muscle car parked in a neon lit Alleyway location Urban
217. 39:36Backstreet at night lighting Moody neon glow cinematic style Noir sound the distant hum of the city punctuated by
218. 39:44the car's ring engine subject a 69 Beetle driving through a picturesque Countryside location Rolling Green Hills
219. 39:52and wild flowers lighting soft morning sunlight cinematic style pastoral sound the gentle hum of the engine and
220. 40:00birds chirping subject hey Quinn you don't you don't need to you don't need to make the sound prompt just the text to video
221. 40:10prompt got it here are 10 text to video props subject a vintage70s muscle car parked in a neon lit Alleyway location
222. 40:19Urban back street at night lighting Moody neon glow cinematic style Noir subject a 69 Beetle driving through
223. 40:27through a picturesque Countryside location Rolling Green Hills and wild flowers lighting soft morning sunlight
224. 40:34cinematic style pastoral three subject a 70s muscle car racing through a rain soaked City hey Quinn I appreciate you
225. 40:42using 70s muscle car for all these but I want each I want each one to be really unique from one another and completely
226. 40:50distinct style-wise so that that these are all individual little Snippets understood here are 10 distinct prompts
227. 41:02subject a futuristic hovercraft navigating a bustling cityscape location neon drenched cyberpunk Metropolis
228. 41:10Lighting electric and vibrant cinematic style sci-fi subject a night on Horseback Galloping through a dense
229. 41:17Misty Forest location Enchanted Woodland lighting soft and ethereal cinematic style fantasy subject a retro robot
230. 41:26exploring an abandoned space station location derel space interior lighting dim and Atmospheric cinematic style
231. 41:34retrofuturistic four subject a painter creating a masterpiece on a cliff overlooking the ocean location Coastal
232. 41:42Cliffside lighting Quin can can you type these out so I can read them absolutely here are the prompts typed out for you one subject a
233. 41:51futuristic hover okay yeah see see it cut it off it cut it off right there so we got we got four of them um yeah it's fascinating
234. 42:03that's probably a that's probably like I think the way to probably do that is because she doesn't seem to be able
235. 42:17to process in the background so I think what you do is you have the initial conversation and set up the prompts and
236. 42:26then you get out of The Voice conversation and then type in the next prompt and say okay give me 20 ideas in fact let me say um great
237. 42:35um give me um 20 more prompts yeah there we go that's how to
238. 42:49do it so have have your your cute little conversation with her and then okay so now so let's go grab what do we have here a pirate ship
239. 43:00caught in a Whirlpool that this one's good all right we'll go over here we'll go to we'll do it live we'll do it live
240. 43:11um let's go back to this Hunan one okay we're in h Hunan so there's wait why did that not not copy
241. 43:26oh are you coming with me oh you're not okay hang on hang on people hang on [Music] shank
242. 43:42shank dank yeah okay I'm gonna jump back to here I'm gonna copy this I'm gonna come in here I'm gonna select all that
243. 43:52I'm going to paste okay so we're doing 720p in hun we'll use Hunan we'll do it live all right one minute
244. 44:11[Music] remaining so this is our pirate ship um I I'll do the cling one the
245. 44:26problem with the clling one is that it's a it's a 10-minute generation so let me let me let this generate again it's like this is weird
246. 44:37that it doesn't have background generation like I should be able to
247. 44:52oh maybe I do have some history here I don't know I don't know how this [ __ ] thing
248. 45:09works yeah yes clling nap and come back later uh yeah I have a feeling that all the good video tools are going to be
249. 45:17like that that there if you want if you want the high quality [ __ ] oh by the way the other thing that we can do is Runway
250. 45:24now does 4K UPR resing now I don't know if it does 4K U uing on non Runway videos so it'll be interesting if we
251. 45:37download this video if we can upres this into uh whatever you know I need V3 from Google or whatever you call it yeah V2 I need V2 as well
252. 45:53from Google I don't know why why are there hold in it back but it's pissing me off this is fascinating watch it watching it make this
253. 46:06video it's boring but you know what are you gonna do what do you want to go do build build a scale model of a ship or build it in
254. 46:173D and then come up with some fluid dynamics in a 3D program is that what you want to do yeah we're not doing that anymore I
255. 46:26don't know if if you got the memo all the 3D people like it's not the same yeah I know it's not oh man all right this one should be a good
256. 46:41Audio One Quinn Quinn did good with this one all right let's let's add sound to this bang but relevant to your comment Vicky
257. 46:50please comment on my post in the mechanics Guild if you want to help guide any for okay cool [Music]
258. 47:01so I'll talk about P I'll talk about the [Laughter] uh P I'll talk about the salon in a second and I'll Point people to your
259. 47:13area Bloom Bower hey Kyle what's shaking what's going down all right so here we go here's the uh here's the audio of this here little clip
260. 47:26[Music] fascinating not great but fascinating all right um you know what I
261. 47:42want to try to do no Kyle why don't you tell us shut up let's go there let's go to Runway Runway has video to video
262. 47:54right so I should be able to import that video and then we should be able to upres that to
263. 48:114K oh you're not seeing any of this hey Apple and Google could you stop your corporate war and just make audio sharing work please please
264. 48:28[Music]
265. 48:43drop an image or video select an asset upload the asset let's go get our ship video
266. 48:57there's our shitty ship [Music]
267. 49:12video 100 foot swells braah I'm gonna get totally pitched by them barrels that's some gnarly surf it's a rogue wave it's a rogue wave
268. 49:28Captain turn apart turn apart we've got a rogue oh no are we not uploading again is that what's going on
269. 49:42here what the [ __ ] we had this problem the other night there we go
270. 49:56[Music] D how is it that two different systems
271. 50:10are creating 720p videos that are not 720p I don't get it I'll never get it let's see pirate ship tosses in in a viol
272. 50:29storm um 5.2 seconds turbo gen 3 Alpha yeah let's go high quality [ __ ] it bang in the queue that'll start in a few
273. 50:44minutes all right let's go back to so here we are at Korea let's go back to our little chat je let's go get
274. 50:53another go get another video to make a lone astronaut planting a flag on the alien Moon steam locomotive roast racing
275. 51:01across a snowy mountain pass this feels like it'll have good sound good copy go to Korea go in go paste let's do a
276. 51:12different model let's do we'll do another minute one what's a minute one here we'll do Luma for this one
277. 51:23generate and then I should be able to in theory three yeah let's do a 10-minute one we'll do a cling Pro on this one cling or cling
278. 51:331.6 cling 1.6 five minutes that's the one that I think everyone's digging okay yeah so that's working so so I've still got that one
279. 51:45generating this thing looks like it hung so we could probably kill this right cancel all right so that's rolling let's go back the runway
280. 52:0566% black bar black oh black oh black anybody have any questions making sense what I'm doing I know I'm just
281. 52:21sort of adding around but welcome to Chad add oh
282. 52:43fascinating that's weird oh you can't hear that can you no
283. 53:05because does does anybody know anybody that works anywhere that we could find out you know the person that came up
284. 53:12with the audio sharing framework for for computers and can we puncture their above ground pool and drain it um
285. 53:28oh here here's see is it here that we do this no is it here no actions look upscale to
286. 53:464K all right we're upscaling to 4K that's a shitty video to upscale to 4K but we're doing it we're going in oh oh oh this looks
287. 53:59cool can you see this oh no of course not because I'm sharing audio I I am so like tonight is one of those nights
288. 54:09where if you're me your life is a living hell because you can't share the thing with the thing I don't know what he's talking
289. 54:18about he seems very bitter is he bitter he seems a little bitter yeah I'm bitter first of all I'm a gen xer why is the train rolling
290. 54:29backwards why why is the train rolling backwards why are the wheels not rolling that's exhausting which one was that that was
291. 54:44cling it's exhausting which one's this one oh this one's cling what's this one this is Luma lum's got a new model coming out I'm I'm
292. 54:56I'm pinning my hopes on lum's new model cat GPT says each company is PRI prioritizing its proprietary systems for
293. 55:04competitive and strategic reasons I guess we're stuck with it exactly because it's AI Kyle embrace the Jank
294. 55:13right Mochi is making videos now it seems totally optical illusion yeah this is one of those optical illusions the train cars are all
295. 55:25the same size and nothing nothing is moving but let's do this let's add audio to this this year video add sound uh oh did I lose oh no wait
296. 55:43what oh my God computers are just Mal behaving here I am canine in the AI salon and I have a business IDE in the
297. 55:54area of app development networking not sure I could should connect um there's a Shameless self-promotion area I am
298. 56:00canine and there's also a call for collaborators area so if you're looking for a partner that could be good um oh this looks wait from there
299. 56:11this looks pretty good although what is the truck on the elevator railro railroad tracks all right we'll add a sound to
300. 56:24this one too and then let's go listen to the sound here I think I'm sharing sound
301. 56:41pretty all right let's go look at our Runway oh you can't see that here's our 4K our 4K ship let's go look at our 4K ship
302. 56:58[Music] because it didn't do it photo realistic 4K is kind of
303. 57:13useless but you can I mean the the water is pretty clean here I mean I guess that's 4K we've got a really bad sound effect [Music]
304. 57:32it's so bad it's so bad oh my God this is so funny it doesn't seem that funny
305. 57:49[Laughter] let's go try another model let's try we'll try
306. 58:03Runway generate and we'll try what's 01 live oh that's animating people Halo this is minia Max Halo of these Halo
307. 58:13should be the best the best of these all right so we got that one we got that one we got those two shitty
308. 58:20ones all right let's go I'm going to go back to chat GPT and grab another thing you you're you're not seeing what I'm doing
309. 58:28here are we just not going to address these sounds oh I've addressed them they sound like [ __ ] they're they're bad although you
310. 58:39know if you just want something to have some basic sound in it then you know imagine cutting this into a movie where
311. 58:47you've got background music and you're adjusting the levels and things like that it's fine uh they're not they're not very good though
312. 58:56my guess is that this is um these sound about as good as 11 Labs my guess is what Kaa did was just um did a call to the 11 Labs
313. 59:07um sound effects API is my guess is my gu all right a giant octopus now that's going to be weird a hot air balloon floating
314. 59:21across a jungle canopy uh a motorcyclist a blacksmith forging a glowing sword okay this one should be good for audio as
315. 59:35well so we're going to come in here we're going to paste there's our blacksmith so let's go we'll we'll go back to
316. 59:46Hunan why did it change it to Portrait what is with these software companies get your [ __ ] ux together [Music]
317. 59:58um let's try 01 live because that's good for animating people and this is a blacksmith so let's try that one oh gotta pay for that
318. 1:00:11one fine no need to be rude about it how do I get out of here oh wait God damn it
319. 1:00:31what did I lose all my oh no here they are okay there's our blacksmith I don't see a blacksmith there oh he's off screen that's
320. 1:00:43fascinating all right that one's doing eight minutes all right so let's go do we'll throw we'll throw Luma in there Aluma
321. 1:00:54generation I'm start frame end frame oh that's cool you can do start frame end frame generate oh two generations at a time I see what's
322. 1:01:09happening all right fine whatever boy that's bad
323. 1:01:31all right um let me I'll let those things do their thing and then um we'll come back to them who's got questions questions
324. 1:01:42comments um people are laughing the ship one reminded me of the Hoodoo Gurus video was that Leonard Skinner
325. 1:01:53skinnard I don't know what I was singing could have been why not try the real Luma uh just because I wanted to try the
326. 1:02:03audio thing here and I don't have I don't think I have access to the new Luma Labs new model Luma lab's current
327. 1:02:12model is is a bit poopy but the new one looks pretty amazing um let me see if I got access to the new one I don't think so oh and there's no
328. 1:02:27way for us to know videos modify what's modify reference key frame
329. 1:02:45huh yeah I don't this isn't this isn't the new one oh you're not looking at this yeah no I don't think I have it okay
330. 1:03:00podcast yeah I'm not going to do the podcast tonight just because I we'll get there we'll do it next week Brandon
331. 1:03:18um oh that's fascinating here look at this um am I sharing I think I'm sharing audio here let's say add audio add
332. 1:03:36sound so this should have Tink shink let's see let's see if we get the good audio all right Pate here you [Laughter]
333. 1:03:54go yeah good that's not
334. 1:04:08bad I mean the video is not great okay so so here's pay you wanted to comment on the audio here's my comment on the audio
335. 1:04:26my comment on the audio right now is it's as janky as six-finger you know six-finger crap and just like six-finger crap next
336. 1:04:40year it won't be janky and also next year I okay so so imagine imagine a tool that can
337. 1:04:56make a 1080p video um faster than its length here's what I mean imagine being
338. 1:05:11able to make a video that's 5 Second long five seconds long but it takes four seconds to generate and then you have these sound
339. 1:05:21effects built into it what that means is that you could have a system that is is just prompting One Clip after another
340. 1:05:29and telling a whole 2hour movie generating it on the Fly amazing yeah Fabiana I always remember it's just the start yep it's just the
341. 1:05:44start this like this is this is in the um yeah exactly this one sounds okay um this is in that category of you know
342. 1:05:54this is the worst it's ever going to be so this one is this is minx so this one is close but again look
343. 1:06:01the wheels aren't Wheels aren't rolling but the rest of it looks really good if if only the wheels were rolling all right so let's add some
344. 1:06:14sound to this one add sound wait what's this one oh that's
345. 1:06:28Runway let's add sound to that one too I think Runway is the best of these that one's a train wreck get it train wreck this one's a train
346. 1:06:45wreck yeah yeah we we expect so much of these tools it it it's so true we're so [ __ ] spoiled we're literally talking
347. 1:06:54videos in into existence that you know a year ago you would have had to hire a special effects team or an animator to do [Laughter]
348. 1:07:06this um let me let's let's change I thought we there's the forge I thought we did this Hilo model or this 01
349. 1:07:22model let's do yeah this is Mini Max let's do this one as well generate or did I do that already am I a
350. 1:07:36loser no I did Hanan all right so here's this that one's pretty good here's let's here's this one
351. 1:07:55that was [Music] really all right let's let's download this
352. 1:08:15one where are you using the multiple models at this is all within Korea kore.ai has got how many models now one two three
353. 1:08:26four five six seven seven different models and then on the image side they've got some crazy amount of models as
354. 1:08:47well yeah we're just the tools right now for for movie making like you've got to be a real if you're a real OCD Movie
355. 1:08:58Maker and Storyteller and you want to like take the time to learn which of these models are good at what or or you
356. 1:09:06just want to take a bet on one and keep prompting that one get better and better at prompting that one
357. 1:09:12model that might be be the way to do it I I don't know I don't I don't know a good way to um to really do video
358. 1:09:20storytelling right now if you want it to be not janky I I think you're just going to have to deal with the Jank you're
359. 1:09:27going to have to embrace the Jank right now um okay so we downloaded that let me change my screen share so that we can
360. 1:09:40do where are we going I'm going to go to Runway where's Runway there's Runway share okay so let me trash this select an
361. 1:09:58asset upload video there's that open cancel
362. 1:10:13upload come on there we go you good there you go okay let's do this with no
363. 1:10:31prompt let's do it with turbo because it's already pretty good add a text prompt um uh
364. 1:10:47train rolling uh forward generate
365. 1:11:03oh here we've got some good video all right well so while oh that's going pretty quick oh that's turbo Runway turbo is pretty quick look at
366. 1:11:12that that's nice
367. 1:11:34ah that of the face [Music]
368. 1:11:55that's kind of cool it took a Runway video I I just put train rolling forward I didn't give it a style or anything so
369. 1:12:03now let's go and change this to 4K see how quickly it'll upload that yeah it's Thomas the Train it's totally Thomas the Train that's really
370. 1:12:16[Music] cool wheels are rolling [Music] I like it turbo almost always means Jank
371. 1:12:30oh yeah let's go let's go do this thing with not turbo and see see what it it does same same prompt okay now here's the 4K one that
372. 1:12:44upscaling to 4K is pretty [ __ ] quick can I zoom in on this uh no [Music]
373. 1:13:09I got to say the quality of that's really quite good it's somewhere between realism and a cartoon but pretty slick godd damn
374. 1:13:25we're anyone else notice we're in a really [ __ ] wild time in history yeah the smoke is good the sound is good the the Thomas the face is
375. 1:13:43good that's [ __ ] hilarious all right let's see what the what the the non-alpha version does oh wait did I did I pick Alpha oh yeah
376. 1:13:55Alpha's the good one I didn't do turbo yeah so Alpha so this one should be I I mean what's fascinating is that it it it
377. 1:14:05kept so much but so much is different so bizarre how these things
378. 1:14:19work I gotta say like yeah just the the the workflow the workflows right now you're going to have to get very specific just basically pick
379. 1:14:32a tool here it added a face to this one too that is the most bizarre thing I've ever seen so this is the non-turbo version [Music]
380. 1:15:02that Thomas spent too much time with Ringo Star and it won't get any worse right yeah well we'll see five minutes ago it
381. 1:15:12was the jankiest time I know crazy all right let's go let me flip back to uh to Kaa because I used some of those
382. 1:15:22different models with the blacksmith and they're [ __ ] good all right that's the one we saw
383. 1:15:37before this did it it did it as a square video which is bizarre but this one's really good so let's add sound to this
384. 1:15:48one so this was Halo so this is minia Max minia Max is the one that's got the the best physics other than vo but I
385. 1:15:57don't have access to vo yet vo looks quite [Music] amazing that's good except you don't
386. 1:16:11have any sound of the fire I wonder if I can can you kill the audio
387. 1:16:25you can't oh that's [ __ ] up once you add the audio you can't you can't undo it there's some janky all right let's add audio to this
388. 1:16:44one want me to put some prompts in vo yeah Vicki let me grab let me go grab the uh let's grab we'll grab the train one steam
389. 1:16:58locomotive copy let me go put them in notes for a second hang on a sec paste and then let me grab the
390. 1:17:13blacksmith paste and then let me grab this one um what's the best way to get these
391. 1:17:30to you Vicky I oh I'll just DM you over in the salon [Music] [Music]
392. 1:17:56ding okay you've got three three prompts the steam locomotive one the blacksmith forging a sword and the pirate ship caught in the
393. 1:18:04whirlpool for the ones that I rendered that pirate ship in the whirlpool was absolutely horrific um the trains were all over the
394. 1:18:14place um let's see what this sounds [Music] like that's the video is bad the audio is bad um this one was pretty
395. 1:18:31good wasn't it why do I not have audio on this one that's weird oh there we
396. 1:18:48go see there you hear the this one you hear the fire going
397. 1:19:07all right let me come back Vicki's goingon to make us some videos so we'll see what those look like I'll post in the regulars once
398. 1:19:16they're done okay Brandon's bitter he wants me to do podcast pods in jelly pod which I am going to do but he's not here next week
399. 1:19:29so I'll probably do them next week when he's not [Laughter] here oh man I must be having fun time is just evaporating I feel the same way um
400. 1:19:42and not in Stereo Ohio State one I'm I'm a Penn Stater Penn State lost yesterday so oh well and then I I get the play of
401. 1:19:55watching the Broncos lose to the Buffalo Bills uh this Sunday so look at my hair that's fantastic any no one no one
402. 1:20:05commented on the hair come on we're but trolls you know what a hammer looks like right uh yeah of course I've definitely
403. 1:20:14seen a hammer before yeah this isn't the hamiest of hammers it's got kind of a sword handle [Laughter]
404. 1:20:29it's a creative Hammer you know what the thing is is with all
405. 1:20:45this in this 5sec video there's like a second and a half of usable video we just got to keep making [ __ ]
406. 1:21:00crazy but I'll tell you what once they start once they start combining um persistent 3D worlds and then the ability
407. 1:21:10to create shots within a persistent world so so right now we're looking at this this corner of this Workshop where
408. 1:21:18the forge is and the Anvil is but if I if I did another shot here um it it would completely change everything right
409. 1:21:27but when they start getting the locations to be consistent and persistent just like the characters this
410. 1:21:34is g to get good really fast it's gonna get good really fast all right um so two weeks ago now we did this thing called are you uh
411. 1:21:48uh AI festiv us AI for the rest of us and um if you look up screen right now Brandon put up the um the URL are you
412. 1:21:59ready for AI so we're putting together Vicky and Ann Murphy and I and and some other folks are taking some of the content that was
413. 1:22:11created during AI festiv us and we're putting together a training for AI Readiness um our Target launch date is
414. 1:22:22February 1 if you want to sign up for that if you want to find out more information about it as it comes
415. 1:22:28available go to that URL on screen are you ready for ai.com let me push this forward for the ticktockers so you can see it there on the
416. 1:22:39bottom um and go check it out it's there there's not much there right now just sign up for it if if you're potentially
417. 1:22:46interested in uh a training course um that that really prepares you for being that sort of curious adaptable
418. 1:22:58adventurous person that's learning about Ai and and constantly evolving and adapting and things like that that's
419. 1:23:06that's what we're looking to teach it's not about teaching the tools it's more about teaching the mindset um and that I'm pretty excited
420. 1:23:13about so so there's that um okay Vicky is going to be putting some videos in the Irregulars channel so if you don't know
421. 1:23:23what the AI salon is go to this URL the salon. and click on join our community and that will take you into um our community website built on
422. 1:23:38top of Mighty networks and if you scroll down toward the bottom there's an area called Community clubs here let me let me share my [Music]
423. 1:23:53screen so here here I am in in the AI Salon there's like a little welcome um track on the left hand side a seven-step
424. 1:24:01welcome there's also AI Festivus site is there so if you missed AI Festivus if you click on that Tab and go to the
425. 1:24:09resource list every one of the sessions has a video attached to it so you can go watch those videos and you can actually
426. 1:24:16ask questions and chat in those in those sessions um and if you scroll down toward the bottom there's an area called
427. 1:24:23community clubs and one of the community clubs is called Irregulars that's this Channel's uh little private area ah here is some
428. 1:24:33V2 video that Vicky has put up all right so here's the train pretty good oh wow they did a cut
429. 1:24:50too is there is there consistency that's fascinating hang on yeah there's decent consistency between the the the physics of the
430. 1:25:05wheels is still [ __ ] up but it's better they're at least rolling and the train is at least moving forward the smoke looks very
431. 1:25:17real yeah vo's the [ __ ] man the fact that it's cutting to three different three different angles of the same train and
432. 1:25:28retaining consistency that's pretty impressive the the [ __ ] Wheels physics though I'll tell you what there's going
433. 1:25:42to be there's going to be specialty uh special effects people that are just going to fix the the shitty AI fuckups in video
434. 1:25:56so anyway you can go see video Vicky's videos down in there that's pretty cool thanks for doing that Vicki that's
435. 1:26:07awesome beauty beauty beauty beauty all right it's Friday night um normally I give you some homework let me let me
436. 1:26:15think about what the homework could be where's my mouse trying to get the black bar going all right
437. 1:26:35grock question mark there is a new Standalone grock app so grock if you don't know it is the AI like chat GPT tool of Twitter X and
438. 1:26:51xai released a grock app on the iOS store it's a standalone app right now it's okay um apparently grock 3 is coming soon and
439. 1:27:08um the thing you need to be careful with on iOS is that there's at least two maybe three fraudulent grock apps make
440. 1:27:19sure you get the one from the X AI company right x x . a um is the is the company that put it out okay what do you
441. 1:27:29think of nvidia's little computer there releasing um I think it is [ __ ] staggering like for $3,000 I mean Nvidia
442. 1:27:40is absolutely killing it I I don't see anyone catching them for some time um I think their robot Suite is amazing I
443. 1:27:49think their consumer um 5,000 series graphic cards are amazing um I think they're they're little $3,000 desktop super computer that's
444. 1:28:02about the size of a piece of toast two pieces of toast stacked together is amazing so you get two of those things
445. 1:28:09and daisy chain them together and you can run a 405 billion parameter AI model locally on your desktop [ __ ]
446. 1:28:19Bonkers uh so I think it's amazing khaki Kyle can you remind us what happened in the EU because they started regulating
447. 1:28:27so early for Monday um sure so um I I'll give you a specific a specific little uh clause in the regulation that they
448. 1:28:43did so the EU passed an AI bill um two years ago they came out first and they said you know AI is really
449. 1:28:53important and we're going to pass this bill and and basically it's a bill that prevents that that that
450. 1:29:02guards EU citizens against the risk of AI and it and it was well it's it's a well-intended bill right but one of the Clauses in it
451. 1:29:17is um you can't you can't distribute an AI system that can measure employees um emotional uh response to something right
452. 1:29:33so you can't be on a zoom call for example and use AI to measure someone's emotions right are they lying are they
453. 1:29:41this are they that right so which on the surface sounds like yeah like in in fact when I read that particular Clause I was
454. 1:29:49like yeah that seems reasonable it like yeah we we don't want to do that right because if employers are using it to you
455. 1:29:56know overly what whatever control or manipulate employees that's not a good thing that said when you [Music]
456. 1:30:19use where which one did I share hang on oh this one yeah oops new
457. 1:30:37chat hey Quinn what's going down hey Kyle not much how's your day going anything exciting on the horizon it's going great I'm really
458. 1:30:47excited that's awesome what are you excited about no now I'm really sad I'm sorry to hear that Kyle want to talk
459. 1:30:57about what's on your mind no hey could you whisper sure I can do that what's on your mind I want you to talk really fast and really
460. 1:31:07loud all right I'll speed it up and talk louder what's going on uh talk loud and talk fast okay I'm talking loud and Fast
461. 1:31:19what's on your mind give me a tongue twister sure how about she sells seashells by the seashore give it a try
462. 1:31:26make one up I want you to make one up and I want it to all be with the letter M all right how about Misty Mountains
463. 1:31:34make marvelous mining mirages give that one a go no but I want it to be really long got it how about many mischievous
464. 1:31:43monkeys mightily marched through the M got it many mischievous monkeys mightily marched through the Misty Mountains making
465. 1:31:54many mischievous monkeys mightily marched through the Misty Mountains making marvelous Melodies while munching
466. 1:32:01mangoes many mischievous monkeys might right okay now why did why did I do all that because that was stupid um the way that advanced voice
467. 1:32:16works is it's a true multimodal um voice model and and what what it means is this with previous voice systems when
468. 1:32:29you would talk there were three different models in play the first model would take your audio recording and
469. 1:32:36translate it into text the model would then send that text to a large language model to ask the question the large
470. 1:32:44language model would respond in text and then this third model would translate from text to voice and then it would
471. 1:32:50talk back that's why voice inter interaction things before advanced voice were so there was such a delay because it was
472. 1:33:00three different models it was like a 5-second delay from when you asked the question to when it answered the thing
473. 1:33:05that I just did where she could sense that I was getting more excited or more sad and she was modulating her
474. 1:33:13interaction with me based on that is because it's a multimodal model that when you talk into it it not only
475. 1:33:20understands the words it understands the emotional intent that law in the EU thing means that that feature is
476. 1:33:32illegal so what's so what's been happening is this new technologies come out mostly from open AI but from other
477. 1:33:40companies as well and they basically say hey we just launched this new feature it's available everywhere except the EU
478. 1:33:48and then what they're having to do is they're having to to to negotiate car outs the legislation for specific
479. 1:33:56Technologies so it's just it's one of those things where when they wrote the law it seem seemed person perfectly
480. 1:34:03reasonable at the time but they couldn't anticipate what was coming right and that's the problem with over legislating
481. 1:34:10early is that you may legislate yourself in a corner the the other thing that the legislators that I've talk to is quite
482. 1:34:19clear to me is they're totally clueless about what this technology is what it does that there's a difference between
483. 1:34:27generative Ai and traditional AI they don't it's all AI to them and all they know is what they've been told the robots are going to kill
484. 1:34:35us everyone's afraid of it it's going to take all the jobs and so their mindset is this negative mindset of AI is a risk
485. 1:34:45and what they want to do is they want to get their name stamped on legislation that says see I was there to protect you from the evils of AI
486. 1:34:54and there's very few legislators right now saying hey let me get educated about this because I want to make sure that we
487. 1:35:00can legislate to protect the Innovation that all of these tools are allowing us all to tap into right so that's that's what that's what's
488. 1:35:10happening with that European legislation the EU AI legislation just crazy all right your AI overlords will remember this Kyle
489. 1:35:23blacksmith posting now but it doesn't like the pirate ship prompt all right let's go look at let's go look at uh can
490. 1:35:35I if I change no it's not GNA go hang [Music] on go to here new
491. 1:35:51activity blacksmith yeah that's what it should be well that's a little [ __ ] up but again you know you can you can
492. 1:36:10use that's pretty good until right there you could cut you could cut back to that like you can cut to you know
493. 1:36:25right there that one's good like that that you could use you could use that hammer strike I mean there's something
494. 1:36:31about this if you ever watched Alex Wright who's the blacksmith on he was on Tik Tok and YouTube for a while um I
495. 1:36:39always thought it was weird he held the the hammer um short like this he didn't hold it way down at the end he held it here I
496. 1:36:47guess to control the amount of deoration um this seems pretty [ __ ] accurate the like the physics in vo are
497. 1:36:58crazy right you see the little things flying off it you got the fire going here these things are consistent as P
498. 1:37:06put put it before you do know what a hammer looks like right this looks like a hammer looks like a blacksmith's
499. 1:37:13Hammer those look like that that's just garbage but whatever look at the tools in the background they look right
500. 1:37:23yeah vo's the [ __ ] man vo is the [ __ ] all right I'm going to get out of here let you people get out of here and start
501. 1:37:33your weekend I don't think I have any homework for you I think what we're going to do
502. 1:37:49is oh no I do have homework for you I do have homework for you you as most of you know and if you don't
503. 1:38:00know you know you'll know now we needed a thousand subscribers on the YouTube channel to be able to do business stuff
504. 1:38:08there we also it turns out need I think it's a thousand hours of views so your homework this weekend is go to the AI learning laab YouTube
505. 1:38:20channel learning la- a if you haven't subscribed Please Subscribe but we're over a thousand now so thank you and
506. 1:38:28just hit play and and play uh like a a string of I think you can go into a playlist and just have it play one after
507. 1:38:36the other and just hit play and then leave you know especially if you're on Wi-Fi and it doesn't cost you extra data
508. 1:38:42I don't I don't want you to run data you're welcome to watch the videos but I just need some play hours we're actually
509. 1:38:49very very close according to Brandon I don't I haven't looked recently um but but just go play some AI learning Lab
510. 1:38:56videos on YouTube and that'll put us that should put us over the edge to be able to do um to do business there
511. 1:39:03Brandon said we're very close that's great okay cool um thank you all so much um I hope you had a good time tonight uh it's it's kind
512. 1:39:15of video's a bit of a tough one to demo because it's time-based media and it takes a lot of time to generate those
513. 1:39:22things they're all still so janky um but the tools are getting better what I can tell you is this it looks like that
514. 1:39:31what's coming from Luma Labs is going to be in the neighborhood of Minx and vo I I would say right now um V2 is in the clear
515. 1:39:45lead Halo Mini Max is number two and then I would sort of put on another tier down Runway which is good but not great
516. 1:39:55um and then a tier below that right now is Luma laabs dream machine but that's got some other things going for it but
517. 1:40:01they have a new model coming so I'm really excited about their new model um but yeah I I would say start thinking this weekend
518. 1:40:15about this is gon to be [ __ ] up think about all of the stuff that you've put time in learning over the past two
519. 1:40:25years what if something happens in 2025 where the AI gets so good that all of those skills that you've amassed become
520. 1:40:36irrelevant and I know that sounds depressing right and they won't be irrelevant you'll still be able to use
521. 1:40:41them but you won't have to because there's going to be agents or there's going to be something that's so good
522. 1:40:47that you no longer have to prompt like you used to have to prompt so I would say start thinking about what it means
523. 1:40:53means to be in a constantly evolving um technology environment where all of our ability to succeed is is I think going to be
524. 1:41:07largely based on our ability to adapt and and stay curious so maybe the the homework assignment this weekend then is
525. 1:41:18get curious and try something new if you haven't written a program with lovable dodev do that if you haven't made a
526. 1:41:26movie with one of these video things do that if you haven't tried LTX Studio do that if you haven't made a song in uh
527. 1:41:34sunno or yudo do that if you haven't analyzed a spreadsheet and chat GPT go do that or have a 4 hour I don't know how
528. 1:41:43long you can have have a 15minute conversation with h advanced voice in chat GPT um while you're driving to the store
529. 1:41:51and just if you haven't done that just maybe to get curious this weekend as what Brandon says if you
530. 1:41:58haven't made a podcast with jelly pod go do that which by the way you can get access to jelly pot as an AI Salon
531. 1:42:06member within the AI Salon where where within there can they find that information Brandon okay here's the
532. 1:42:12pirate ship let me show the pirate ship since Vicky made it we'll go Shameless self-promotion okay nons salon
533. 1:42:27and Shameless self-promotion is where the information is about jelly pod all right and if you don't know what jelly pod
534. 1:42:37is you design a voice by describing a character and how you want it to talk and it will make you a voice based on
535. 1:42:45your description and then you can put multip multiple of those in a podcast and give it a topic and it will just
536. 1:42:52make you a podcast with your characters in it it's pretty slick we'll play with it next week when Brandon's not here so
537. 1:42:57we can make fun of him no screen share up I know I'm putting it up uh here we go here we go all right put this here we'll put
538. 1:43:08this here we'll put this here see that's good that's pretty good vo vo's where it's at man
539. 1:43:27yeah nothing we did I mean this is it's it's a little over drama dramatized with the with the lightning but it's still not bad it's pretty
540. 1:43:36cool squat squat thank you for doing that Vicki that's super cool all right everybody I'm gonna get out of here Pace out hope you all have a
541. 1:43:48fantastic weekend go play with some [ __ ] that's it be adaptable start thinking about your adaptability
542. 1:43:55quotient how adaptable are you think about ways to increase that all right it's G to serve you I promise pretty
543. 1:44:05cartoony though yeah it's this is a little cartoony um all right everyone on YouTube X and uh what's the other one
544. 1:44:14LinkedIn peace out have a fantastic weekend and I will see you Monday