8/27/2025 - Nano Banana to Gemini 2.5: Exploring Google's Image Generation

Transcript

1. 0:04[Music]
2. 0:19[Applause] [Music]
3. 0:36[Music] What is this? [Music]
4. 0:55Woohoo! [Music] Woohoo!
5. 1:14[Music] [Music]
6. 1:38Hello. Hello. [Music]
7. 1:59Is this place I can rest my forhead, gather my thoughts in sweet silence The bus is place where the feelings
8. 2:15aren't dead. [Music] From over exposed to balance is this place I can slowly face. The only one I truly can know.
9. 2:32These are tears from a long time ago. Got these tears from a long time ago. I need to cry. 30 years or so. These are tears from a long time
10. 2:51ago. [Music] Oh darling, oh darling, say unto me,
11. 3:07where have you been? All my lifetime [Music] I have been swimming seven sad seas. [Music]
12. 3:21Four women have to cost me their lifelines. I'd pull them into the waters in I'd have warned them but I didn't know
13. 3:33these are tears from a long time ago. Got these tears from a long time ago. I need to cry 30 years or so. These are tears from a long time.
14. 3:51No. [Music]
15. 4:09Uh, what's happening good people? What's shaking? What's going down? Happy Thursday night. It's Thursday night.
16. 4:17Probably going to do some more Nano Banana. Why are we focusing on this Nano Banana? Why did Google change the name from Nano
17. 4:28Banana to Gemini 2.5 Flash image generation module? Really? Everyone's talking about Nano Banana.
18. 4:42Well, yes. uh be being a a responsible uh corporate upstanding citizen. Uh we we named uh the Gemini 2.5 Flash is it's
19. 4:53consistent with the well you've got G Gemini 2.5 Pro which is the the larger model. Uh it's of course it's a
20. 4:59multimodel mole model uh which it means of course you know it's got different modes uh of media that that that it can
21. 5:07process natively. One of those of course is image. Uh and then in in the 2.5 flash which is the smaller model uh the
22. 5:14the uh the uh image generation tool is uh great greatly improved and uh the fid fidelity and promp prompt adherence and
23. 5:22and coherence uh quite frankly is uh is considerably improved. Nano banana. Yeah. No, that was the code
24. 5:31name that was never intended to be uh publicly uh disclosed or correlated with Gemini 2.5 flesh, which is the official
25. 5:39name, just to be clear. It's not it's not Nano Banana is not the I I know people seem to like it and it
26. 5:47and we we did make banana images. So, so there are banana but yeah it's called it's called Gemini 2 2.5 flash with it's it's the image
27. 5:57generation within within native natively generating within really [Music]
28. 6:12[Applause] [Music] Yeah. [Music]
29. 6:37Heat. [Music]
30. 6:51[Applause] [Music] Heat. Hey. Hey. Hey.
31. 7:06[Music]
32. 7:21Um, what else been going on? Um there was I think probably worth talking about um I don't want to give it too much
33. 7:35credence but I think it's important to note that that like one of the things that I talked about I don't know six eight
34. 7:44months ago is that I think that as as AI capabilities get stronger that the resistance to AI is is also going to get
35. 7:51stronger and there was a uh an AI I meeting in Portland. I think it was last night. And while they were in having the
36. 7:59meeting, a bunch a bunch of their cars got vandalized and there was anti- AI crap spray painted on the cars. One of
37. 8:06the things that we talk about in the AI salon all the time, like one of our values is empathy. Um, and especially
38. 8:15when people do stuff like that that's that's violent and, you know, and aggressive. Um I think it's important
39. 8:24it's important to understand that people are scared and you know the anger comes from fear and you know I think it's um
40. 8:37I think it's that fear that you can't stop this thing right and so so they're trying otherwise to stop it um and
41. 8:46that's kind of the the whole purpose of this channel is that you know technology ology has never progress has never not
42. 8:53progressed, right? Technology progresses. It always has. When the steam loom when the steam powered loom
43. 9:02showed up, you know, there was there was all sorts of riots and things like that. Tik Tok camp. Sorry about that.
44. 9:09Um Oh, that was about a week ago, Vicki. Ah, okay. I just saw I just saw the post today. All right. But anyway, um but but
45. 9:18I just want to talk about it. I mean it's you know if you run into people that are angry or fearful or or you know super resistant
46. 9:28I think one of the interesting not one of the interesting things I think one of the important things to do is
47. 9:34you know understand you know what's the fear where you know what's where's the anger coming from um a lot of times the things that people
48. 9:44are upset about I all like I agree with them Right. And a lot of times they don't know what to do with that. Um, but
49. 9:55my philosophy is I I have no interest in taking sides, good or bad, about a technology that isn't inevitably going
50. 10:05to progress forward. What I'm interested in is if it's going to progress forward and knowing what we know about it now
51. 10:12and knowing what we know about how it's improved. Hang on, I got to sneeze. Do I maybe?
52. 10:29Yes, I do. If the only thing that I can be certain about is that it's not going away and it's going to keep improving,
53. 10:45then what do we do about it? And the only thing I know how to do is learn about it, play with it, understand it,
54. 10:54understand where it's strong, understand where it's weak. I think I think one of the things that people that are outside
55. 10:59of AI that are not playing with it do is they they romant they romanticize it or demonize it, right? The romanticizing it
56. 11:07of it is it's, you know, it's this allseeing, all knowing, omnipotent being. the demonization of it is this
57. 11:14it's this evil robot that's going to kill us and take all of our jobs and and ruin everything. Neither of those are
58. 11:21true. And when you use it, you go, "Oh, I get it. It's a tool. It's really powerful. It's got these things it does
59. 11:26well. It's got these things it doesn't do well, and it's getting better and better and better and better." And so,
60. 11:32what are the implications of that? One of the implications is that as it gets better, there's going to be a demand for
61. 11:39people who know how to deal with it and know how to think about it and think critically about it and understand what
62. 11:45it does good and what it does bad and and find people that have chain of craft with AI that is fluid and flexible and
63. 11:55they can pull in lots of different tools to solve problems. Those people are going to be in demand. So, if you're here, how many people do
64. 12:02we have here? 32 and 16 on on the other channels. Um, you're way ahead of the game. And even if you [ __ ] hate AI, but you still came here,
65. 12:17welcome and cool because the stuff that's shitty about it, that is legitimately shitty about it is shitty.
66. 12:28And I agree with you there. the stuff where people actually don't understand how it works and they've got it in this sort of fantasized,
67. 12:35you know, utopian or demonic extremes. Learn it, understand what's there, understand what it how it can help you,
68. 12:46right? If you just got laid off, you now have access to these tools that can amplify your ideas. I just talked to
69. 12:53someone today who was a senior executive at one of the big ad ad holding companies, really good friend of mine,
70. 13:00and he got laid off last March. And, you know, he sort of bounced around trying to find jobs for a couple of months. And
71. 13:06then he said, you know, the writing's on the wall with his AI [ __ ] I'd better get going. And he signed up for an MIT
72. 13:13course and he's he's in the middle of that and he's he's getting his ass going. So, so anyway, I just thought that was worth
73. 13:24saying saying into the universe, you can make money with Chachi Bet. Where's my button? >> You can make money with Chy. [Laughter] [Music]
74. 13:57[Applause] [Music]
75. 14:14Oh, tonight T tonight for the AI leaning lab. So, producer Brandon, uh, I've I've assigned him spellchecking and apparently
76. 14:25apparently he missed that one. That's that's on him. It definitely wasn't me. I don't make typos. I use AI to type, so
77. 14:32I never make mistakes. [Laughter] Oh, man. All right.
78. 14:46AI leaning lab. That's really funny. Genie in the bottle. I thought another channel was trying to steal your
79. 14:54thunder. No, that's just a typo. I just screwed up. Oh, man. All right. How much have you generated
80. 15:08with the sale of that button? I have never sold this button. I should sell this button. You can make money with every
81. 15:16>> um this was just an inside joke here. I listen I had a I in the in the heyday of the irregulars and all of these jokes
82. 15:24like where this joke came from in the in the early days of the of AI. >> I should have sold a bunch of merch and
83. 15:33I didn't do it. So at some point I'll probably do that. But yeah, we could sell that one. Some we we'd probably sell six or seven
84. 15:41of them. make make $17. Nice guitar strap. Thank you. I think this is you know what where I got this. I think this was just an Amazon guitar
85. 15:54strap cuz I was I had one more guitar than I had straps and I kept rotating them and one day I was just like, "Oh,
86. 16:00[ __ ] it." And I went I went to Amazon and just ordered that. That's where I think this is from. So, no sentimental value at all. [Music]
87. 16:21Did someone say early days? Lunick, what's happening? You remember the early days of >> you can make money with Jack?
88. 16:31>> I'll tell you what's funny. You know what's funny right now? It's It's not um these these douchebags have left, but they've been replaced
89. 16:41with um What are they doing now? It It's all the uh um you know, there's a new image generation model out from
90. 16:55Gemini, and I'm going to teach you how to prompt it. You can't just do simple prompting. You have to do complex prompting, JSON prompting.
91. 17:04You too can prompt with Jason. [Music]
92. 17:19You got to start selling merch. All right, we'll get on it. Actually, you know, producer Brandon, we should
93. 17:24probably talk about doing merch. I think we have We've got stickers. as long as stickers. Um, all right.
94. 17:40We were in Staples yesterday and my buddy said, "Doesn't Kyle have one of those buttons?" Oh, my hubby. That's really funny. [Music]
95. 18:02You know, [Music]
96. 18:28Oh, I want those stickers for DC. So, Daisy, uh, DM me and we'll connect with Tobias. I don't know if Tobias is here,
97. 18:35but Tobias printed a bunch of those, so we can we can, uh, we can get we can at least cover his cost on those and and get some of those
98. 18:45for DC. When are you going to DC? [Music] [Applause] [Music] [Applause] Oh, so by the way, if you're new here,
99. 18:58welcome. My name is Kyle Shannon. This is the AI learning lab. This part at the beginning where it's just like, does he
100. 19:03ever talk about anything? Sometimes I do, but this is just kind of the warm-up. It's like letting people, you
101. 19:10know, digest their food, have a little appertif. We got a very sophisticate sophisticated crowd here who has appertif.
102. 19:24I'm pretty sure that's what it is. Anyway, apparently people like them and you know they like to have their meals
103. 19:30and then they like to have their appertifs and then they like to come to the AI learning lab and a lot of people many people
104. 19:39um have given up Netflix to watch this. It's true. I don't know if it's many at least one or two. Um but we will talk about some AI stuff
105. 19:54tonight. Um, we're going to dig back into Gemini um, image generation. Um there there's some cool stuff that it does that I want to
106. 20:03experiment with where you can create you can upload three different images. But in each of those images you can like
107. 20:11give it instructions and it it can parse because okay the image generation models that are now within chat GPT and within Gemini.
108. 20:24I'm trying to think if there's another one that's got it in it right now. Maybe it's just those two. Right now,
109. 20:32it's actually native inside the large language model. It's not even I is it still called a large language model if
110. 20:38it's multimodal? I think it is. But anyway, so it can understand images and audio and video like the pictures in
111. 20:48video and text and code all of it. Um, and so you can put like written instructions on the the graphics that
112. 20:58you upload and it will follow those. So, I figure we'll experiment with that a bit tonight. I could be watching season
113. 21:056 of The Secrets of Skinwalker Ranch. You're that good, Kyle. Hey, listen. Skinwalker Ranch is one of my favorite
114. 21:12shows. Like all ghost shows, my my favorite thing about those shows is that they never find anything. The one the one treasure show that I
115. 21:22watched, it was called uh I think it was the secret or the curse of Snake Island. And and they started out this treasure
116. 21:31hunt journey on Snake Island, which is named that because it's got the highest concentration of snakes of any place on
117. 21:41Earth. It's one poisonous snake per meter, per square meter. [ __ ] idiots. And they went on there to find a treasure and then and then
118. 21:54some hint there said you should go to South America. I think they just wanted to get off Snake Island. But they go to
119. 22:00South America and then in that season they [ __ ] found the treasure. It was insane. And and you you want to know about season 3?
120. 22:10There wasn't season 3. They found the treasure and they're like, "Fuck TV. We got treasure." So, we never got to hear about the treasure.
121. 22:22Unbelievable. So, you know, if those Ghost Hunter shows, if they ever actually find a ghost, the show's over. [Laughter]
122. 22:33We just watch it like a bunch of dupes. Like, I wonder if they're going to get a ghost this week. Did you hear that?
123. 22:40Something clicked. Did you see that floater? It floated right across the camera. I'm pretty sure it was sentient.
124. 22:48I love those [ __ ] shows. My favorite one it was it's the the one with the uh what I forget what it's called.
125. 22:58It's not Ghost Hunters. Maybe it is. It's the one with the the meathead guy. The really pumped up beefy dude. He's
126. 23:06like got a big voice and he's always like getting all macho with the ghosts and then the minute one of them touches him he's like
127. 23:17But anyway, I'm glad I'm more entertaining than those shows. Um, I saw a couple of Tik Tok videos that
128. 23:26they've apparently found the treasure uh on Oak Island, which which there if they actually have found the treasure, um,
129. 23:34more power to them because they've been doing that for more like a decade now, 11 years, 12 years. It's crazy. Oh man. Okay. Um,
130. 23:49danger noodle island. Oh, the no prop. Yeah. Yeah, exactly. That that that curse of snake island or whatever the hell it was called.
131. 24:01That was trippy, man. cuz they're walking through cuz there's no humans on it cuz there's that many snakes and
132. 24:07there it's just walk walking through brush everywhere and like at their feet at their waists and at their faces are snakes,
133. 24:16you know, and they're like pushing sticks away and you can see the snakes kind of scurrying away and it's like Jesus,
134. 24:25you know, TV's swell and all but really. All right, let's go. Let's go play. Let's go play. Anybody I I've just been
135. 24:35rambling. Anybody have any questions, thoughts, anything you want me to cover, talk about.
136. 24:50Should we do ASMR? Hey, hey, welcome to the AI learning lab.
137. 25:06What about cling? Um, I haven't played with cling in forever. Oh, what's great about clinging? Well, so cling.
138. 25:18So, so Cling, I think it's Cling 2.1. Cling 2.1, Luma Labs, Runway, VO3, Midjourney.
139. 25:33There's some Chinese ones. There's like there's like I don't know, seven or eight really good video models. Cling's one of them.
140. 25:42Yeah. Nano Banana. Nano Banana plus video. All right. Yeah. So, that's what we're going to do. We're going to go figure some [ __ ] out.
141. 25:50Nano banana plus video equals magic. Cling has some promos going on right now. I would think that all these video
142. 26:00companies are going to have promos going on because there's there's so many of them. I I have a feeling that the the
143. 26:05video industry is going to get really oversaturated quickly and and you'll see consolidation maybe in the next two years.
144. 26:15Facebook isn't letting me share my link. That's cuz they're [ __ ] Um, you know who's the other one that
145. 26:22that just promos the [ __ ] out of their stuff is Higsfield. And I I don't know if Higsfield has their own model or if
146. 26:29they're just using other people's models and they've got some pre pre-done special effects. It's it's it's like at
147. 26:37this point it's impossible to keep track of who's got models, who's using other people's models. I I just I can't anymore.
148. 26:47I just try to pay attention to the [ __ ] that makes the most noise. Um, did you ever upload the guy singing?
149. 26:59Oh, singing. Singing the other night. I did. Let's Let's go. Let's go look at that. Video still expensive. Video is still
150. 27:09expensive. Although right now you can Google's given everybody the option to make some videos with VO but it's like
151. 27:18they're giving you like five videos. It's not enough. Yeah. Video video is still very expensive. My guess is what's going to happen is this.
152. 27:32So, do you remember remember when I I used to play with Deco here? The real time image generation stuff. Korea's got
153. 27:40it. Um, Deco here has it. It's all based on um what the hell is it based on? Um, SDXL. I forget who made it. Anyway, it doesn't
154. 27:54matter. It's a It's a model that that generates images really, really quickly. They've got video versions of that as
155. 28:02well. So they've so so in testing in in some of these labs, they've got video models that are generating
156. 28:09video faster than 24 frames a second. And if you can generate video faster than 24 frames a second, you can start
157. 28:17to do real time stuff. and and the uh the Genie 3 that that Google just announced, which is the Speak a World
158. 28:26into Existence, you're walking through a 3D world that's actually being generated at faster than or it's being generated
159. 28:34at 24 frames a second. Um, it's not showing you my input and output. What?
160. 28:49All right, you're talking to each other. Black bar. Wait, and don't I need to show Someone wanted to know about the
161. 28:56song. So, let's let's go. Let's go find the song. So, this was the thing that I tried to upload the other
162. 29:02night. I I just couldn't. And then right before Brandon and I hung up. Oh, this was funny. This was kind of cool. So, I
163. 29:11took this image. So, this is like my corporate image, my corporate head shot from like, you know, seven or eight,
164. 29:16nine, 10 years ago. I forget when we took it. So, I uploaded that into um into Nano Banana into Gemini 2.5 Flash.
165. 29:26And normally whenever I've used that image to train someone before, my head's always in that position and it's always
166. 29:33got that dorky smile on it. And so what I wanted to do was see if like I could get it to retain sort of what I looked
167. 29:40like, age me up a little bit, um and change my expression and change my position. And so I think it took me
168. 29:48seven or eight prompts, but I ended up going from that to that, which is, you know, that's pretty cool. I look hip.
169. 29:58I look hip and like I need a haircut. Oh man. Um, but the other night, let's see, what did
170. 30:13I make? I made that dude in Nano Banana. There we go. That's the one. Tab hoarders unite.
171. 30:29I'm sharing the right tab. I think I am of a thousand favicons. [Music]
172. 30:46I bow to the altar of a thousand favic kinds. Each a tiny stained glass promise. Each tab relic an argument. An
173. 30:54unfinished prayer. We do not close, we collect. We are archaeologists of distraction. Curators of half-bun obsessions. We
174. 31:03preserve the tabs that you discard. They call us tab hoarders like it's a crime. I call us inspired keepers of the incandescent. Now
175. 31:14every tab is a shrine to a thought that refused to die. A headphone cable tangled with yesterday's revolution. A
176. 31:21recipe we swore we'd master. A thread we swore we'd answer. We keep the world in micro windows like sailors with sex
177. 31:29mapping storms of curiosity. Chrome hates us. Chrome groans under our little empires. It's memory like an
178. 31:38anxious brain on the subway. Leaking, forgetting, dropping passengers on the line. Memory leak, they say, as if sin were a
179. 31:47bug fixable with a patch. Google with its polished teeth and corporate himnels cannot fathom the sanctity of our open questions. It
180. 31:57trims, it prunes, it gaslights our tabs into sleeping bags. But we are stubborn, gentle, glorious, unruly. [Music]
181. 32:10Let the browsers call it inefficiency. We call it fidelity. We are a million tiny vows. I will return. I will return.
182. 32:19I will return. Each tab is a bookmark for an argument with the future. We keep versions of ourselves and
183. 32:26duplicate. The person we were at 2 a.m. The person who clicked that headline in a cafe. The person who loved a sentence.
184. 32:33Close one and you risk killing that small particular grief. Close one and the map of you becomes less accurate.
185. 32:40There is an ethics to tap hoarding. Not greedy. Not. >> So anyway, that's that uh question. What website is this? This is just on Twitter
186. 32:49right now. So actually here's here's a request. Go to my Twitter account, Kyle Shannon. S Han N O N. and find this
187. 32:58post. It's called Tab Hoarders Unite. It's like, I don't know, six or seven posts down. And do me a favor and like
188. 33:05go like it, quote post it, do [ __ ] like that because it's kind of fun. I like it. And I I like the the rebellious tab
189. 33:12hoarder. Um, if you're wondering how that was created, um, I wrote the lyrics. I wrote the lyrics I think in chat GPT,
190. 33:25but it might have been Gemini. It was one of the large language models, Gemini or Chat GPT. And then um and then went to I was experimenting
191. 33:37with um Sunno to see if Sunno could do spoken word and it can. That's that's what it was or that's what it did.
192. 33:46Um, and then I took the audio file. Um, and I took an old picture that I had from my Kyle Shannon Dreams project,
193. 33:58which is that, you know, sort of beat poet washed up 60-year-old. And uh I took that image into um
194. 34:13into Nano Banana into Gemini 2.5 and um made this thing or maybe I did it in midjourney. I think I did it in
195. 34:21midjourney. Yeah, I did it in midjourney and then I took this image and the sun and put them together in Hedra. All right.
196. 34:35Hedra, Hedra, Hedra, Hedra does the animations pretty well. Um, Hen today came out with a new version of their
197. 34:43digital twin technology. So, for what that's worth. Oh, thanks. It looks like a bunch of people have gone and shared
198. 34:49that. So, yeah, if you go if you go to my Tik Tok uh channel or no, my Twitter, my ex channel, Kyle Shannon, and find
199. 34:58the Tab Hoarders Unite song and just boost that. That would be swell because X when whenever I post anything that I
200. 35:09think is really cool and like people are going to see this. It gets like three clicks like God social media sucks.
201. 35:20Um I guess we'll just change this tab into a Gemini tab. Okay. So where I am right now is at gemini.google. google.com and I'm on the 2.5
202. 35:36flash model. So 2.5 flash has got this new image generator tool in it or as they call it on the X it's nano banana.
203. 35:52Um, so this is here and then we can upload up to three image files. So I think what we're going to do, I'm going to change
204. 36:03how I'm sharing here for a second. I'm going to switch over to sharing my whole screen.
205. 36:18down. All right.
206. 36:32All right. Fantastic. Bob, have they won anything? Uh, no. It's It's not a game show. Um, okay. So, I'm
207. 36:42going to go there. Then I'm going to go I'm going to go to here. What's this? This is [Music] This is Keynote.
208. 36:56So, I'm going to make a new slide. All right. All right. So, here's what we're going to do.
209. 37:22Um, okay. Let's go find photos. Um, so what I'm doing, what I'm doing right now,
210. 37:36let me let me explain the madness that is about to ensue because it's not really going to be madness. it's just
211. 37:43going to be boring and and if but if you know if you know what I'm doing it might be a little less boring. Um
212. 37:52so right now I know I want to explore this image generation model in Gemini. I know that you can upload three different
213. 38:01images and tell it to combine them. But I also know that those images can be quite sophisticated. So, so what I'm
214. 38:08trying to do right now is figure out what I want to do. I don't really have a clear use case. So, for the next 15 minutes or so, I'm
215. 38:17just going to be bumbling and putting together the piece parts that we're then going to take into Nano
216. 38:24Banana and see if it can parse them. Um because again, part of the part of my philosophy, so if you're if you're new
217. 38:33here, hey Brandon, can you pop up the um can you pop up uh the AI salon banner? Yeah, there you go. So, if you're new here, um
218. 38:52[Music] All right, that's still there. Okay, great. If you're new here, go to the salon.ai. Actually, if you go to community.thesalon.ai,
219. 39:03that'll take you to the community site of the AI salon. So, this is the the AI learning lab is an extension of the AI
220. 39:11salon. It's it's a part of it. So, basically, this channel is brought to you by the AI salon. Um, it's a really amazing community of
221. 39:24AI friendly, curious, adventurous people that are trying to figure this stuff out. Um, so a lot of the a lot of the
222. 39:32work that we'll do in here that I'll play with in here, I encourage you to play along. And one of the things that
223. 39:37we talk about in the AI salon is what we call the cycle of AI readiness. And let me just I'm going to just pop over there
224. 39:45for a second and talk about it because I think it's really important. And the cycle of AI readiness looks like this.
225. 39:56Play first, mindfully create, generously lead. And so what I'm doing tonight is is play first. I'm just playing. So I don't have
226. 40:05an agenda. Mindfully create is like once you learn how to do something and once you've got some basic skills in a tool
227. 40:12um then you can mindfully create. You can go problem solve or you can go do you know self-expression.
228. 40:19Um but but the stage we're at tonight is just this play first. And so what I'm going to be doing is just kind of
229. 40:24[ __ ] around. So it it might if it seems like I don't have a clue what I'm doing, that's because I don't have a clue what I'm doing.
230. 40:34And if part of my role here on the AI learning lab is to make it okay for you to not know what you're doing, then I've done my job.
231. 40:44I'll be clueless first. You follow along. Um, and you know, don't don't shame yourself for not having your head
232. 40:51around this stuff because like nano banana, this new image model came out two days ago and part of what we're
233. 40:57trying to figure out is is it any good? And the answer is yes. But it's like, okay, how specifically is it good? So
234. 41:05maybe tonight we'll play around with some JSON prompts. Maybe tonight we'll play around with this image prompting.
235. 41:10Things like that. All right, make good sense. So go join the AI salon. Step two in the AI salon is introduce yourself.
236. 41:17So if you haven't done it yet, if you're here just lurking and you haven't introduced yourself in the AI salon, shame on you.
237. 41:26Don't be a ninny. Okay. Um, where was I going to go? I know where I was going to photos.
238. 41:40So, we're going to go find So, these images here are the part of
239. 41:56this thing I created called Kyle Shannon Dreams. So, what we're going to do is we're going to just Can I copy that? Why can I not copy that?
240. 42:09Copy. Can I go over here and paste it? Yes. Okay, great. So, we're going to get some weird weird things. Copy.
241. 42:22Oh, wait. We'll do it like this. Get that guy. We'll get that woman. and we'll get him. Okay. Copy. Okay.
242. 42:34Now, we're going to go over here. Okay. So, here's what we're going to do. Okay. Okay. We're gonna we're going to name these
243. 42:48these these people. [Laughter] I'm such an idiot. This was a fun project. This was a
244. 43:04project I did. All these images came from uh from 2022 in the fall. Um, I taught myself to install stable diffusion and and then I
245. 43:20I installed this thing called Dream Booth, which allowed you to upload some images of yourself and and make these crazy ass self-portraits.
246. 43:35Um, how are we going to do this? Oh, I know. Don't do
247. 43:54shrink. That singing he does is very annoying. All right, so let me grab all of these things. Bring them to the front.
248. 44:11Let's lock that bad boy. Okay, now I've got more room to play with. Now this makes it easier. Okay, so
249. 44:19we got those three and then we got these three. Right. Okay. Put that there. We'll put that down there. All right. Now we'll add some
250. 44:34text. And now we're going to name these things. So, this one we're going to name Willie, as in Willy Wonka. [Laughter]
251. 44:45We're going to name this guy Tom. We're going to name this guy Benson. Benson.
252. 44:59Uh, ideas. Jared. Jared. That's Jared. And and you're like, Kyle, why are you naming them? I just cuz I'm playing. I'm
253. 45:13playing. People, don't you understand play? Uh, this is the mad hatter. What should his name be? Um, um, Haddington. [Laughter]
254. 45:28That's original. Um, we're going to call her, uh, Bessie. And then we're going to call this guy Tony as in obnoxious award recipient.
255. 45:44Tony. Tony award winner. Okay. Willie, Tom, Jared, Haddington, and Bessie. Okay. So, let's duplicate this.
256. 46:00Okay. I got an idea. Okay. Got it. Got it. Got it. Going back to my picture things. Now I'm going to go Kfax and Wazi exhibit. Okay,
257. 46:17we're going to grab this sunset. Copy. Get rid of all these things.
258. 46:34going to put in that sunset that was this was on the uh the mountaintop right above Boulder, Colorado. This was I was driving to
259. 46:44Boulder one night and that sky was there was amazing. All right. So, we're gonna I don't think we need to
260. 46:54label this one, but we're gonna have Okay, we're gonna have that. And then we're going to have We're going to find a place.
261. 47:08Ah, this this is a good one. Okay. Copy. Go here. This was Oh, this is also really cool. So, so when I took this
262. 47:16picture, this motel picture, so you really can't see it, but right in between these two cars, there's this
263. 47:27little fold up chair. And so, I was just in the parking lot. I had my camera out. I was taking pictures cuz the sky was really cool. And
264. 47:35um and I saw this sign and I wanted a picture of the sign. And as I pulled in, there was there was this trashy woman
265. 47:43sitting in that chair smoking a cigarette. And I tried to discreetly pull up my camera and get a shot of her,
266. 47:49and she just immediately took a beline into her room. It's still a cool shot, but it would have been cooler with the
267. 47:56trashy woman smoking a cigarette there. But you know, what are you gonna do? All right. All right. So, okay. So, here's what
268. 48:06we're gonna do. We're going to we're going to we're going to take screenshots of these things, save them to the desktop.
269. 48:25Okay. And then I got to remember these names. Hey, do me a favor, Brandon. Can you just write down these names? Um, Willie, Tom, Jared,
270. 48:41Haddington, Bessie, and Tony. I love, by the way, in case you didn't know it, one of the things that gives me
271. 48:55great joy in life is just being an idiot. Okay. All right. So, that's saved. Okay. So now we have some input images. You're
272. 49:06like, "But Kyle, where is this going? This is so fascinating. H what are you doing? I don't know." Okay, so we're going to add files.
273. 49:20So I got to go find screenshot. There should be two new ones. Boom. Right. Yep. There's that. There's that. Okay. So, what we're going to do
274. 49:36is we're going to say I want um a group photo of
275. 49:53Willie Tom Wait. Willie, Tom, and Haddington
276. 50:13right in front of the camera. Arm in arm.
277. 50:27I want Willie, Tom, and Haddington. So Jared, Bessie, and Tony. I want Jared, Bessie, and Tony
278. 50:44to be spread out across the second story balcony of the hotel.
279. 51:04motel looking down on the threesome taking their picture and I want
280. 51:19the sky to be that crazy sunset. Okay, so in theory, we're going to have three people standing in front of the camera,
281. 51:35the hotel in the background, three people up on the second floor balcony, and the crazy and the crazy sunset. Oh, and then we'll do this.
282. 51:48Make the image 16x9 wide aspect ratio. Make sure.
283. 52:04No, I'm not going to say make sure anything. [ __ ] it. Let's see what it does. Right. Okay. We're at gemini.google.com google.com
284. 52:14playing with 2.5 flash or nano banana and we're going to see if it can actually understand and parse those images. All right, here we go.
285. 52:28[Music] Uh uh uh uh uh uh uh. Now, what's really powerful about this if it works is here we go. Whoa.
286. 52:47Okay, we didn't quite It sort of [ __ ] up a little bit. It put them on the third floor. Okay. So, so, so, so it
287. 53:06it kind of it kind of worked. I mean, I didn't give it the best instructions, did I? Um, all right. That's kind of cool. So, so
288. 53:21it made Tom How do I zoom in on these? So it made it made Tom. So Tom, this guy, the guy in the middle has Willy's outfit
289. 53:33on, but that's definitely Willie. That's definitely Haddington. And that's definitely Tom. That's cool. And then up here you have you have
290. 53:46it put Haddington in place of the woman. Did I put Did I put Bessie? Did I put that right? Hang on.
291. 54:04Yeah. Jared, Bessie, and Tony. Okay. So, now we're going to do this. Now, we're going to say, um, I want a closeup
292. 54:19of Bessie on the roof near the motel sign with the
293. 54:37Um, sunset in the background. This is pretty cool, isn't it? All right, let's see what it does with
294. 54:51that. We'll see if it remembers Bessie. All right. No, it didn't. It didn't
295. 55:06understand that. Okay. It just gave me that image again. That's weird. Okay. So, we'll try this. We'll try
296. 55:20Oh, you know what I'll do? I know what I'll do. No, Kyle. What? We'll re-upload those two images. I'm going to say, "Okay,
297. 55:32totally new image idea. I want Bessie by herself on the roof.
298. 55:47The camera, the the photographer is on the roof with her and
299. 56:01the motel signs are behind and above her. The sunset sky is still
300. 56:19dramatic. All right, let's see if it'll just take one of those characters and give us a new thing without having to start a new chat.
301. 56:44Hey Ryan, let's cool it down, please. Thank you. So, did it do Bessie?
302. 57:03Yeah, it did. Well, sort of. Yeah, that sort of looks like her. Wow, that's pretty remarkable.
303. 57:20So, this [ __ ] works. All right, so let's let's go get a different sky. [Music]
304. 57:42Uh
305. 57:59black bar. Black bar. Oh, blackberry bamb. Oh, blackberry bama jam. All right, there we go. There we go.
306. 58:13Mine didn't work. It gave me the code for the image. Yeah. What I'm noticing about nano banana is like sometimes if I say
307. 58:23um you know have these three standing near each other, it doesn't understand that it wants you to make a picture. So,
308. 58:31I'm like I'm finding myself having to go make a photograph of like you have to really expl be really explicit with it.
309. 58:38But, let me I'm going to go grab a different um location. I think this one's cool. This barber shop. Oh, we'll do Union
310. 58:54Station. Okay. Copy. Okay. So, we got Union Station. I think we got room for one more location, don't we? We sure do.
311. 59:14Sorry. Sorry. Carrying on rather Cheerio Pip. Okay, we'll get the barberh shop.
312. 59:29When I retry, it says something went wrong. It showed me the thinking then the code. That's weird. Um,
313. 59:36I would Oh, hang on. Physical dexterity on tik tok. Oh, I missed it. Sorry. Hang on. All right. Um,
314. 59:51okay. So, we got three locations, two skies. Let's get a third sky. Three locations, three skies. Oh, we'll see if it can do logic. That
315. 1:00:02would be cool. That would be cool. Um, let's get a golden sunset. Okay, here we go. this one. Copy paste.
316. 1:00:18By the way, if you want to see these photographs, most of these photographs are at um if you go to kyleshannon.com,
317. 1:00:27that's my photography website. These are actually made with a camera and my eyeballs. And and I gotta tell you, you know, it's
318. 1:00:38funny when I got when I got hate blasted on LinkedIn for saying that you could possibly be creative with AI. Um,
319. 1:00:49you know, people were just saying, you know, you can't possibly be creative because it's just pushing pushing a
320. 1:00:54button. So is [ __ ] photography. Like the reason I was talking to a friend once because I couldn't figure out what what the
321. 1:01:06through line was of my photography because it all felt very different. And he said to me, he goes, "Oh, no. You
322. 1:01:13you've absolutely got a a photographic voice." And I said, "What is it?" And he said, "You take pictures of the things
323. 1:01:21that most of us walk by and don't notice." And I was like, "Oh, that's so cool." And and I find when I'm when I'm doing
324. 1:01:29image generation in AI, when I'm just sort of [ __ ] around with it, it has a very similar feeling as to when
325. 1:01:36I'm out with a camera that I'm just like sk always scanning and I'm like, is that a picture? Is that a picture? I'm like,
326. 1:01:44oh, that's a picture. And then I'll stop and I'll take it and I just take a bunch of them. And then when I get back, I
327. 1:01:49look and see if anything was actually a picture. And I feel like that's how it is with AI, too. That you just generate
328. 1:01:56all this [ __ ] It's like walking around the world and just like there's a piece of gum on a bench. There's a dude that
329. 1:02:02looks like he's a lion. There's right there's two pe people arguing. There's that looks like a Renaissance painting.
330. 1:02:09You know, look at that sunset. And when you're doing AI image generation, it's the same thing. It's just like massive
331. 1:02:16amounts of input and then you can pick and choose just like you do with photography. So anyway, for what that's worth. Okay, let's
332. 1:02:28screenshot this bad boy. So now we got three locations, three skies back to here. Okay, so we're going to go upload files.
333. 1:02:41There's that. There's our people. Okay, I want a black and white
334. 1:02:56portrait of Tom standing in front of the
335. 1:03:12is it barber shop? Hang on a second. Yeah. Hair and nail. G&G up down up down. G&G uptown cuts.
336. 1:03:27Yeah. Okay. in front of the barberh shop with the
337. 1:03:41blue sky above it and then in parenthesis everything should
338. 1:03:56let's see Um, I just want the shapes of the clouds in black and white. Black bar. Got it.
339. 1:04:12Black bar. Black bar. Black bar. Black bar. All right. Let's see if we get a portrait in black and white of Tom,
340. 1:04:22who's the little comedian dude. So, wait. This This is Tom. [Music]
341. 1:04:41[Music] Oh, interesting. [Laughter] It sort of did it. It definitely gave us Tom.
342. 1:04:58It definitely gave us Tom in black and white with the blue sky behind it and black bar. Um, that's pretty amazing. But it left all
343. 1:05:11this other stuff. It It redid some of them. That's crazy. Oh, look. And it took the It took the people out of this one. Okay.
344. 1:05:27Okay. I got I got an idea. Um Okay. New picture. Have all six people hanging out
345. 1:05:48near the guy on the bench. at Union Station. I don't want any other
346. 1:06:02locations or skies. Just a new version of that of
347. 1:06:16that picture with The six people added,
348. 1:06:47I can't pass something cool and not take a picture. I know. Me, too. It's It drives my family [ __ ] crazy because
349. 1:06:53I'm like, "Wait, wait, wait." All right. It completely screwed this up. But it's kind of fun how it screwed it up.
350. 1:07:13Who's this guy? That guy's That guy doesn't exist. Put Willie up there. This is wild. It's just [ __ ] wild, man. Crazy.
351. 1:07:31All right, let's try a new thing. Let's try a new thing. Okay, we're going to take these two. I'm going to duplicate this.
352. 1:08:13All right. We're not going to label anything. We're going to do this as a new a new chat.
353. 1:08:33We're going to upload the thing we just made. So, we're going to say I want a Oh, sorry. Tik Tok pin.
354. 1:08:46Kyle, thank you for always being willing to take one for the team and put yourself what's that say out
355. 1:08:55there. No, listen. I appreciate the kind words, Source Camp. I This stuff is [ __ ] intimidating. If you don't think AI is intimidating at
356. 1:09:10this point, listen, here's So, there's a couple of things going on. When chat JPT first came out and when
357. 1:09:18there was just basically a handful of image generation tools and some animation tools and there was kind of
358. 1:09:23chat GPT and nothing else, it was really easy to pay attention to what was going on. But even back then, people were
359. 1:09:30intimidated by AI because they thought you had to be an engineer to be able to do it. Still no black bar. I know it's cuz I'm a bad man. Um,
360. 1:09:43and that wasn't true. And a big part of what this channel was in the early days was this shit's easy. Just use your
361. 1:09:50words. You don't need you don't need math and science and engineering to do this. What's happened in the in the two and a
362. 1:09:58half years since is that there's so much progress and there's so much advancement and the these tools have evolved so much
363. 1:10:06that you don't have to be a programmer or an engineer still, but you do have to [ __ ] put some hours in. I I heard um I saw a
364. 1:10:16a a post on LinkedIn today that that I thought was really good because because I talked um I don't know a year or so
365. 1:10:24ago about the 10,000 hour rule is dead. Meaning that as AI tools become more and more capable, you don't need to put in
366. 1:10:3310,000 hours, you know, drawing or taking pictures to get good at photography or drawing, right? You can just do it with a prompt. and and
367. 1:10:42someone on LinkedIn today said, um, 10,000 prompts is the new 10,000 hours. And I was like, ah, that's really good.
368. 1:10:53And and so that's that's what we're playing with here. Like, there are probably an infinite number of ways that
369. 1:10:59I could be prompting these images. Um, what we're just experimenting here using image prompts and trying to get it to do something.
370. 1:11:08This is just an experiment. the there are there are people are going to find all sorts of ways to exploit this. So,
371. 1:11:16I hope you're playing along tonight. Just just play along. And if you're not using these tools, just play with other
372. 1:11:21tools. Doesn't really matter. Put in your 10,000 prompts. When you say Tom, do you mean Tom Lake? Looks a little like him. Uh, I don't
373. 1:11:31know. I just he looked like a comedian, so Tom seemed like a good comedian name. Okay. I want a classy vertical photo
374. 1:11:49graph of the man and the woman standing in front of
375. 1:12:08Union Square station with the sky. The golden color
376. 1:12:22which completely changes the color of Union Station. Not Union Square Station. That's New
377. 1:12:39York. Union Station. All right. Um, this should be a single coherent photograph.
378. 1:13:00[Music] Where I know I'm going to do a sec
379. 1:13:35Okay, let's see. Did we get it? Oh, we did. How cool is that? Union Station
380. 1:13:50Traven and Train instead of Travel and Train. [Laughter] There you go. All right, people.
381. 1:14:06We're This is This is kind of a thing of another, you know, another realm. So So now you can start to do stuff like this.
382. 1:14:13So now you can just start to to art direct this. You can say, "Um, that's a little too golden. Let's Add
383. 1:14:29some reds. Black bar. Yeah. Blackberry. Buy my jam.
384. 1:14:45Nice. Beautiful. Um, put her in a blue dress and hang on. Let's go look at um
385. 1:15:04we're going to go Google images and we're going to go uh 2015 Jaguar XF maybe.
386. 1:15:21All right, that's what it looks like. 2015 Jaguar XF. Okay, back to here. Back to here. No, back to here. Yes. Okay. Put her in a blue dress
387. 1:15:37that matches the 2015 Jaguar. What was it called?
388. 1:15:51XF. They
389. 1:16:07are standing in front of and lighten up the
390. 1:16:23exposure. I know that's a cool sky, right? Amazing.
391. 1:16:40There you go. There's the Jaguar XF. She's now in a blue dress. Dang, man. This is something. This really is something.
392. 1:17:00Um, put the car on a proper 3/4 angle farther behind them.
393. 1:17:26Trying out Google's new photo images, huh? Yeah, exactly. Oh, it didn't it didn't understand the threequarter shot of the car. Um, I want
394. 1:17:36to see the front of the car. on an angle. So
395. 1:17:51it adds depth behind them. Make her hair straight
396. 1:18:06and his hair gray. Uh, I think that I like the dress matching the sky, though. His suit could match
397. 1:18:20the car, maybe. All right, let's we'll go back to that after this one. Everyone's an art director. No, but I
398. 1:18:29think you're right. It's It's still okay. So th this is a thing that I noticed that Gemini seems to get like this
399. 1:18:39composition stuck in its head and it just can't not do that. Let me try this. I'll try um let's make his suit
400. 1:18:55dark blue. Return her dress to gold and put the car
401. 1:19:09farther in the distance at an angle. See if that if if it does that. [Music]
402. 1:19:25If you give the photo back to it, it seems to jump start it. Yeah, it might. Can you just say it car a 45 degree angle? Yeah, it could.
403. 1:19:37Yeah. So, he's not in a blue tux. She's sort of in a gold dress. And the car the car is now stuck. All right, we're gonna do Okay, hang
404. 1:19:51hang on, people. I'm putting in my 10,000 prompts. New chat. Upload files.
405. 1:20:07Um, the two of them are standing for an elegant
406. 1:20:22portrait. Behind them at a 45 degree angle is a 1969 GTO Judge.
407. 1:20:45Um, the sky is golden and they are in front of Union Station.
408. 1:21:01All right, let's see what we get here. Got to go with a 69 GTO Judge. What are we doing with a 2015 piece of snot Jaguar?
409. 1:21:18Yeah, that's what I'm talking about, baby. Yeah. See what I'm saying?
410. 1:21:33Got that tricked out. Yep, this works. Union Station. Even the N is is uh is out just like the real that real photograph.
411. 1:21:49Pretty amazing. You want to see something? You want to know what's what's most annoying to me about this?
412. 1:21:56You [ __ ] kidding me? You're going to watermark all these images? For as much as I pay you, Google, you're
413. 1:22:03going to stick a [ __ ] watermark on my image? What are we 1997? We're going to put watermarks on everything so everybody knows it's us.
414. 1:22:14It's called branding. We have a branding specialist. And what the branding specialist did was he said, "Why don't
415. 1:22:19you take your little AI symbol, the little star, and you just put that in the corner? No one will even notice it."
416. 1:22:25No, no, we notice it. It's a dick move. Get rid of it. Yeah, I know Photoshop takes care of that, but I shouldn't have to [ __ ] do that.
417. 1:22:39That is not the Dukes of Hazard car. This is not the Dukes of Hazard car. But we can say um replace the car with the Dukes of Hazard car.
418. 1:22:56[Laughter] Hey, Google needs brand recognition. They might be forgotten. I'll tell you what, Google's catching up. I mean, this
419. 1:23:09image gen like this this makes um this makes Open AI's image gen right now look like [ __ ] because um Open AI can't
420. 1:23:18retain characters across generations. Um it's got that piss filter where everything's that that bizarre antique
421. 1:23:25yellow. Um it's slow. It's OpenAI's image gen is really slow. Um, it is good at text text I would say better than
422. 1:23:37this this Google image gen. But but these guys they're they're getting ahead. Oh, the green dress with Daisy Dukes.
423. 1:23:47Oh, that's good. That's really good. Uh, wait. Oh, that goes against my guidelines. Um, oh crap. Uh, make it the uh what was it? The 69.
424. 1:24:00No, it was the it was the 1972 Trans Am with uh uh black with eagle on the
425. 1:24:14hood. The smoky and the bandit one. Just a sec. Generating image. All right, we got past that.
426. 1:24:33There we go. That's a Smokeoky and the Bandit car. I mean, not for nothing, but good lord, we're in just an amazing
427. 1:24:48time.
428. 1:25:03Okay, I'm gonna I'm gonna uh I'm gonna go post this on Twitter because I'm an idiot. Oh my god, I love being an idiot.
429. 1:25:21Why does he always do that? It just doesn't make any sense. Oh, I want to get rid of that stupid [ __ ] thing. Okay. Well, we we we've
430. 1:25:31got we've got ways to get rid of your stupid [ __ ] watermark. Yeah. Tools. Um new. Um
431. 1:25:47where am I? Is this it? No. This it? No. Okay, that's it.
432. 1:26:14Yeah. Let's we'll we'll make this a little more cinematic. Beauty. Good people.
433. 1:26:30And let's let's go in and let's let's do some some manual color correction here, shall we?
434. 1:26:46[Music] do [Music]
435. 1:27:08d do d do d do d do d do d do d do d do d do d do [Music] Swinging ding ding ding ding ding ding ding ding ding ding [Music]
436. 1:27:22ding ding ding dang dang ding ding ding ding. All right. Uh, save. Now we're going to go to the Twitter.
437. 1:27:35We're going to put that image there as if it really happened. Oh, wait. Why did not not save? Come on. Come on, computer.
438. 1:27:45There we go. All right. There's the image. Um, one of the reasons I was late for late for my AI learning
439. 1:28:04lab live tonight was the Cannonball fun
440. 1:28:20gala fundraising spectacular that I founded
441. 1:28:35and was the sole donor for It was a
442. 1:28:51fantastic event for the six people that showed up. I invited
443. 1:29:06[Laughter] Bert Reynolds, but he never got back to me.
444. 1:29:21Um, the the comedy there, if you don't know, is uh Bert's no longer with us. All right, there you go. All right,
445. 1:29:36idiocy. Idiocy. Idiocy shared. So, do me a favor. If you would go to Kyle Shannon onx and uh
446. 1:29:53big comments or or retweet it or or not. Oh man alive. All right, people.
447. 1:30:08I take the train. Oh, ran friaking chastic people. All right, yep. All right. Um,
448. 1:30:23okay. So, the lesson for tonight is like like I really don't know what I'm doing yet. I really don't have a sense of
449. 1:30:37where the boundaries of of this new image model are. Um, and that feels kind of icky. It feels, you know,
450. 1:30:49like I think I have it in my head that like, well, you're on here all the time. You should know all this [ __ ] You just
451. 1:30:55can't know all this [ __ ] So, if you're feeling clueless, welcome to the club. Um, but just keep playing, just keep pushing
452. 1:31:07it. Just keep And I don't give a [ __ ] if it's this tool or some other tool you're working on. It could be vibe coding. Um,
453. 1:31:17just understand that the the more you prompt, the more that you learn, the the better you're going to get. And what I
454. 1:31:23would encourage you to do is share what you're learning in the AI salon. Share your images, share your prompts, share
455. 1:31:30your victories, share your defeats. Um because we can all help each other figure out, you know, how to get there faster. All right.
456. 1:31:42Beautiful. Fantastic. Where am I? I don't know. Sleepy Joe, you're at the AI learning lab and we were talking about uh learning
457. 1:31:57N AI and we were uh we just made this here image out of two photographs that were uh part of an art project I did three years ago. A
458. 1:32:13photograph I took of Union Station, a photograph I took of a Colorado sky, and a Smokeoky and the Bandit Trans Am.
459. 1:32:21And we just told And we just told Gemini's new image model, "Put all that [ __ ] together." And it did. So, that's what we're doing.
460. 1:32:31Um, and if you're new here, I go live five nights a week. So, tomorrow night is tomorrow's Friday. Tomorrow night is Friday night, date
461. 1:32:42night. Uh, you can bring a date. You don't have to. I can be your date. We'll just hang out. And then, uh,
462. 1:32:49>> uh, Kyle. Oh yeah, producer Brandon >> for Steo tomorrow is Friday, but for the rest of us here states side it's still Thursday,
463. 1:32:59>> right? Because today's Wednesday and I knew that and that was >> because you had a podcast with Murphy today.
464. 1:33:04>> That was that was really obvious to a lot of people and one of the things I like to do on here is see if anyone's
465. 1:33:12paying attention and apparently at least at least one person is. Thank you, producer Brandon. Tomorrow's not Friday.
466. 1:33:21That's so weird. I thought today was Thursday for sure, but no, this was the day I took my kids to South Denver for
467. 1:33:27treatment. I had the podcast with Dan Murphy. Yeah. No, today's Wednesday. Solidly Wednesday, except for Steo who's
468. 1:33:35in Australia. It's Thursday, so for him, but it's that's a whole different thing. It's geography thing. Anyway,
469. 1:33:50oh man. Um, thanks for reminding me what day it is. I can't keep up anymore. He's in the future. He is. Steo's in the future.
470. 1:34:01Steo prompts from the future. He He's got He's got 12 hours on us all. Um, I'm trying to think what else. So, never
471. 1:34:10mind about that. So, Thursday, tomorrow night, normal normal night tomorrow night. So, I'll see you at 8:00. If
472. 1:34:16you're new here, five nights a week, I do these things starting at 8:00 p.m. Mountain time for the most part. Um,
473. 1:34:24and sometimes it's a bit more structured. Sometimes, like tonight, it's like when there's a new tool that's got some depth to it,
474. 1:34:32um, I I try to dig into it a little bit to understand it. This thing's pretty pretty flipping capable. So,
475. 1:34:40um, okay. So, a couple of things. I want you to join the AI salon. So, go to the salon.ai and, uh, click on join our community and
476. 1:34:50do that. But I also want you to check out, can you pop up are you readyforai.com? As if, as if he read my mind. Um,
477. 1:35:00so we just launched the AI readiness training program, which if you go to are you readyforai.com, you can learn all about it. It's an
478. 1:35:10amazing program. Um, it's comprehensive. It covers five different areas. It's it's not specific to any tool like
479. 1:35:18midjourney or like if we did training that was just about midjourney and then this tool came out, you'd be like, wait,
480. 1:35:25why am I learning midjourney? Shouldn't I be playing with nano banana? The answer is we're going to we're going
481. 1:35:31to play with different tools all the time. What the AI readiness training program is about is about shifting your
482. 1:35:36mindset. Thinking about how you think creatively, how you solve problems in business, how you use AI, you know,
483. 1:35:44ethically and p with privacy and security in mind. Um, all sorts of things like that. So, go check that out.
484. 1:35:51And uh if you're trying to if you're trying to up your game with AI, it's it's actually really important you get a
485. 1:35:57certificate that you can, you know, brag about on your LinkedIns uh and tell everyone you got your [ __ ] together with
486. 1:36:03AI because I think that's going to be increasingly important. All right. All right, people. Fantastic. Bob, tell him what he's won.
487. 1:36:12He hasn't won anything. Okay. All right. Um [Music] The answer is always it depends.
488. 1:36:34All right, cool. Groovy, peace out, everyone. Uh, be nice to each other, please, in the comments. Thank you. All
489. 1:36:41right, I'll see you tomorrow. [Music]