aiLL 8/14/23 - Navigating the Wild West Early Days of AI

Transcript

1. 0:00foreign
2. 0:19[Music] are we live are we live
3. 0:34has anyone joined cat dragon number one in the house welcome welcome welcome um hey hey
4. 0:48got a scratchy voice from being out in the world welcome to the AI learning lab my name is Kyle Shannon I'm
5. 0:57coming to you live from New York City New York New York New York it's a hell of a town the Bronx is up and the battery's down
6. 1:16welcome everybody this is not my normal place I am in a hotel room just a little teeny hotel room right next to Times Square
7. 1:27we're riding right in the middle of Manhattan we're right in the middle of it all everything's happening everything's happening
8. 1:35it's fantastic Bob tell him what he's won you want a new car welcome everybody Welcome we'll get started here what day is today Monday right
9. 1:52I think it's Monday a couple of meetings today one of the meetings the dude we flew out to sea didn't show up yay
10. 2:07I love coveted work commitments [Laughter] welcome everyone this yeah learning lab my name is Kyle Shannon
11. 2:17a little bit about my background I try to do these lives every night I kind of missed last night although we did a live stream uh from
12. 2:26downtown Manhattan right near Ground Zero it's beautiful gonna walk around walk by the old uh hey Dallas cooler what's happening
13. 2:36walked down by my old offices my oldagency.com offices we walked around Lower Manhattan that was cool and then I
14. 2:42was gonna do a live last night but I got home and I was just toast because I've gotten up at like three in
15. 2:48the morning that morning and it was you know by the time I got to my hotel room it was like 10 o'clock deep in the bowels of Jersey
16. 2:57so um so anyway so we're back we're live tonight it is what time is it eight something oh nine o'clock 908 here in
17. 3:07New York City New York City the way this works I'll talk about this so so these websites right here if you're new to
18. 3:15generative AI if you have not played with chat CPT go to that top one chat.openai.com I'll talk about these in a second
19. 3:23um but just go there and start playing right now if you haven't played um I'm an entrepreneur I'm the co-founder and CEO of storyvine uh we
20. 3:32are in New York visiting clients and talking about all sorts of work they want to do everyone seems to be excited about
21. 3:42generative AI our clients are asking us for their input on how they should deal with Microsoft how they should deal with
22. 3:51their agencies which is kind of funny because we're a we're a small but humble a small yet Mighty company
23. 3:58um I don't know if we're Mighty but we're we're we are small we punch above our weight um so so that's exciting I'm also I was
24. 4:07also an entrepreneur in the early days of the world wide web and one of the reasons I'm so fired up about this
25. 4:13generative AI stuff is the parallels between the early days of the web and the early days of chat GPT and kind of
26. 4:20the shift from the Geeks and scientists and mathematicians owning um generative a or owning machine learning and AI it you know chat GPT
27. 4:32sort of shifts it into now the rest of us get to play with it which is exactly what the World Wide Web did for the internet
28. 4:39so that's that sole proprietor LLC uh no we are a C Corp and we have six or seven employees and an outsourced
29. 4:49Tech team we're an 11 year old company um the uh what we do on the channel here if you got questions pop them in the comments
30. 4:59below I do my best to answer all the questions so I'll likely fall behind so let's let's jump in my voice is a little shot oh it happened
31. 5:08I just lost my I lost my dual screen hold please that's not what I wanted is this what I wanted
32. 5:23effects no that's not what I wanted hope please where did my dual camera go I knew I figured this out
33. 5:38at one point before enhance enhance oh there it is there's dual camera well not special
34. 5:55you know you want to know who uh tick tock tick tock has some help doing their little interface here the they they had some help
35. 6:03and the person that helped them do their interface what could that have been well well I know who it was it was Satan
36. 6:11Satan made it completely impossible to use well fantastic [Music] okay back to where I was going Okay so this
37. 6:25this little list of things if you're new to generative AI kind of hang on I got a different setup so I'm trying to trying
38. 6:32to figure this out on the Fly here um all right there you go so so feel free to go ahead and take a screenshot of that um
39. 6:43so so these are all sort of variations or competitors to chat GPT if you haven't played with chat EBT that's the
40. 6:50official website right there then you've got Bing and Claude two kind of major Bing is sort of a co-op Edition
41. 6:59kind of thing that's Microsoft's version of gpt4 it's connected to the internet it can generate images it can read
42. 7:06images Claude is a competitor from a company called anthropic it's got a hundred thousand token context window
43. 7:14meaning you put in really long prompts um so that's pretty good Bard is Google's entry into this perplexity.ai is a really interesting
44. 7:24kind of search engine citation GPT kind of thing but the labs.perplexity.ai they're giving us free access to the meta Lama 2 models
45. 7:34and then Pi is from a company called inflection posed from the company quora all of these are worth exploring so
46. 7:45um so that's that another thing that I wish I didn't have to talk about but I do is oops is how do you set up your light
47. 7:54in a hotel room apparently not very well thank you I'll be here all night don't forget to tip your waitress try the chili um
48. 8:04the the AI learning lab has gotten so successful that that we've been scammed we've been scammed so there is a
49. 8:19um a copycat account of the AI learning lab that's got two L's in it it's a i l l so two L's for learning that's not how it's spelled
50. 8:29so if you see that one they're [ __ ] scammers they've copied my content they've copied the name and they're
51. 8:35trying to get people to sign up for some [ __ ] scammy multi-level marketing [ __ ] I will not if I reach out to
52. 8:43you for multi-level marketing [ __ ] I will call it a pyramid scheme it'll be crystal clear but we don't do that here so if if
53. 8:52you've been reached out I apologize I don't there's no reason for me to apologize they're the [ __ ] but
54. 8:58that's going on so I'm sorry about that um if you're brand brand new to this uh this whole world and you don't really
55. 9:06know where to start and you're like I don't even know what chat EPT is go to that thing prompts.chat we'll teach you all about it what these
56. 9:13large language models are how to talk to them it's got a whole list of like all these different personas that you can
57. 9:19tell Chachi BT to act like a screenwriter or an accountant or a lawyer or whatever all that sort of stuff and it's just it'll give you lots
58. 9:26of ideas of what it makes possible if you're into the image stuff those things at the very bottom mid Journey
59. 9:33Adobe Firefly and generative fill within Photoshop Leonardo is a really cool web interface and from a stable diffusion
60. 9:41clipdrop.co is a bunch of tools from stability AI from stable diffusion and that prompt book thing right there will give you access
61. 9:52to lots of knowledge about how to use those things and then future PD is just a big old list of AI tools there's like
62. 10:004 000 in there blah blah blah blah okay so that's that the next thing tomorrow night is the the AI salon so go to the
63. 10:08salon.ai we've got a Meetup Group there you can sign up for the Meetup group and uh um you'll get information about when the
64. 10:16group is it's tomorrow night from five to seven mountain time so seven to nine East Coast time because of my business
65. 10:23travelings I don't think I'm going to be there for the beginning of the meeting but I'll probably join as a participant
66. 10:29Leah and some one of the other organizers are going to host tomorrow but we have John Knack from Adobe from
67. 10:36the AI group at Adobe is speaking tomorrow so the ones behind Adobe Firefly John Knack is is uh one of the leaders in that part
68. 10:46of the organization he's a really good guy I used to work with him um he's wicked smart he talks faster than I do and uh and he's gonna just
69. 10:55talk about what they've learned and and you know all that sort of stuff so that should be a really fascinating
70. 11:00fascinating thing so go to the salon.ai sign up for the Meetup you can also sign up for a Discord where you can come play
71. 11:07hang out ask questions make images whatever all right and my voice is going I don't know how long I'm gonna laugh tonight
72. 11:19but I'm gonna do my best I'm gonna do my best it's gonna be it's you know what it is it's gonna be special it's gonna be a
73. 11:25special night special it's gonna be special okay let me get to the questions I'm gonna scroll up a little bit here and
74. 11:38see what anyone's asked see if we got any trolls by the way if you're gonna troll at least be creative about it just don't
75. 11:45give me some boring ass [ __ ] like oh isn't it nice that the geriatrics are trying to teach people about AI if you
76. 11:53say that kind of [ __ ] you can just jump off a flipping Bridge all right hello awesome what are you
77. 12:00doing in New York I'm out here we've got a Pharma client that was like hey could you come out here and talk to us about
78. 12:06all this special AI stuff you're doing and whatever this automated video technology you have for your company
79. 12:12whatever you're doing with that we want to talk to you about that and they talk just like this I'm going to talk to you about that
80. 12:20we're like okay we can come out there do you want us to get on a plane we'll get on a plane and we'll come out there
81. 12:27and we'll stay in a hotel so we did that and then so we had one today in Jersey well no we had two meetings today in Jersey although we got
82. 12:35we got stiffed at one of our meetings [ __ ] didn't show up I mean the lovely professional uh that I'm sure had some of their commitment
83. 12:44didn't show up well the other meeting we had was really really good and then tomorrow we've got two meetings and then possibly like a
84. 12:57drive-by with a very important person and then we have uh dinner with uh with a prospective client tomorrow night and we're going to
85. 13:07um new on Joe Korean Barbecue on 32nd Street right off Fifth Avenue in Koreatown and if you're in New York and you
86. 13:17haven't been to new one Joe that's a mistake you got to go there you got to go to new one Joe it's amazing it's amazing it's amazing
87. 13:29all right I use chat CPT job interview plug-in to prep for a data analyst interview and loved it so helpful oh that's pretty cool awesome
88. 13:37which one the the job interview plug-in oh that's really cool all right oh yeah Dallas cooler the other thing I did in
89. 13:49New York I didn't come here to do this but I knew I did I knew I was coming here for a client um but I also knew that I could get my
90. 13:58world ID and I did I got my world ID so I got scanned by the orb my irises are now in the in the blockchain I don't think
91. 14:09they are but uh it was cool it was a little anti-climactic because the the place was called it wasn't called The
92. 14:16Grove what was it called the canvas I think it was called the canvas and so I'm like oh it's the canvas I
93. 14:23thought it was like some hipster work Collective or something like that no it was a clothing store just like a boutique clothing store in
94. 14:32in a in the uh the subway entrance down at uh Ground Zero at the World Trade Center um is the it's this beautifully stunning
95. 14:42building and so I went in there and it's basically just a big mall of really expensive shops right so so this store's
96. 14:48there and then you just walk up and sitting next to the cash registers the silver orb and the and the the you know
97. 14:55the women at the clothing store is like oh yeah that thing we don't know anything about that but you know make
98. 15:01your little app do what it does and just look into it that was it and I looked into it had like a little ring that you
99. 15:08know showed progress as it was scanning my irises and at some point I had to make them wider because it kind of
100. 15:13stopped she goes oh you may need to open your eyes wider so you know took my glasses off opened my eyes wipe
101. 15:19and then when it completed the ring it it went somewhere and then it it uh pinged on my phone in the app and it said um it said
102. 15:29analyzing your uniqueness and it was like this little bar going across I mean you're verified and so now I have this virtual kind of
103. 15:38nft ID card that you can flip backwards and forwards and it's Anonymous which is kind of cool so it doesn't have my name
104. 15:45on it just says Anonymous verified human with my you know I've got a private key for my identity so the idea is that there'll be a whole
105. 15:56um infrastructure built where developers can create applications where you have to have one of those IDs to participate
106. 16:04so imagine a version of Twitter that was World ID verified where you knew that everyone in there was a verified human and then imagine tying
107. 16:13that to another application layer where you could say not only is that human verified but this piece of content that
108. 16:20they created is also verified on the blockchain and we can also verify that it hasn't been modified I think that stuff's going
109. 16:27to become really really important so anyway yeah I did that too so that was kind of cool welcome sole proprietor nope we talked
110. 16:34about that Joy hail hey Joy hackers heaven have you used worm GPT I haven't I've heard about it that one and there
111. 16:43was chaos GPT for a while or wasn't it called chaos GPT the the sort of nasty ones that are going to go out and do all
112. 16:50sorts of hacking and destroy the world yeah we'll see we'll see I mean here's the good news hackers are gonna hack
113. 16:55right hackers are going to jailbreak [ __ ] hackers are gonna hack hackers are going to turn autonomous agents loose
114. 17:02just to see what they can [ __ ] up but then the good news is is that the the good guys have access to these tools
115. 17:08too so the good guys are going to go [ __ ] up some hackers so you know it'll be entertaining just hopefully they don't
116. 17:16do something so bad that it turns everyone off to this stuff because that could slow it down for a year or two
117. 17:23but I don't think it's going to slow down for a year or two I just don't here's the thing about all these tools
118. 17:28here see about all these tools see all these these are just like the chat GPT ones these are just the words
119. 17:35um they ain't going away they ain't gonna be uninvented um Chachi PT saw to that when they got 100 million users in six weeks now
120. 17:47chat EPT also in the past two months the usage has gone down does that mean it's over no it just means you know hey there's other
121. 17:58options a lot of those other options are free um and B people are trying to figure out you know is this valuable to my life or
122. 18:07not this software is really new it's really thin meaning it's not very sophisticated right now it's a lot of
123. 18:13spit and duct tape so um some people if if it's not ready for prime time they'll just sort of turn off to it so
124. 18:20so that's what's going on there I think what type of consulting jobs do you see in AI um I think all sorts of them um I've been doing um
125. 18:31you know I've been doing stuff within my own company right trying to figure out what's the value proposition
126. 18:37look like what's the future state of the technology look like what's the product roadmap look like but in pitching that to my clients and
127. 18:47you know establishing some credibility like I know a little bit about what I'm talking about a lot of those clients now
128. 18:54are saying hey can you help us figure this out um so if you if you start getting AI literate if you start getting good with
129. 19:02these tools and start talking about it just put it out there talk about it at dinner parties talk about it at work
130. 19:08talk about it on LinkedIn especially um people start reaching out to you so that's so that's one thing you can do you can teach workshops
131. 19:19um there are a lot of agencies now selling automations so they're they're using tools like zapier or make or
132. 19:25bubble you know automation low code no code tools um to make automations and so there's a bunch of agencies starting up going hey
133. 19:34we can make you a Content generator or whatever it might be whatever automation um companies want to be straighter or
134. 19:42whatever it might be whatever automation um companies want to be so I'm seeing a lot of automation agencies um what types of consulting jobs
135. 19:57if you start to establish yourself out there as someone who's thinking about this stuff um just some opportunities will rear their
136. 20:07heads I was I was working with a group the other day and this guy said hey can you stay after Kyle and I said sure
137. 20:14and he's got some sort of opportunity where it's a large company they're having a 2 000 person meeting they want
138. 20:20to do some really cool [ __ ] with AI there's not much time to turn it around and I said well see if they're willing
139. 20:27to pay for it if they are let's go see if we can build something so we might have to build something in like two weeks I love it
140. 20:36have you noticed whatever you do a Google Search now gives you a generative AI response as well is that just in
141. 20:44general Google I have not noticed that I know they had didn't they have like search.google.com or some
142. 20:51some [ __ ] like that hang on let's see um what
143. 21:06I got one what what is the average speed of an unladen swallow
144. 21:28no I got nothing here I got nothing I think there's I think there's there's a uh a version of this that um yeah hang on hang on I gotta move
145. 21:43something I gotta flip something around I've got my computer on a on a pillow and it's sort of leaning toward toward me so the
146. 21:53the computer is is doing bad things it's still a bad thing it's making me cranky this is not my setup
147. 22:05I just knocked the phone off the little receiver yeah uh come on what are we doing very good are we good everybody good all right okay
148. 22:21[Music] um oh I'm gonna light where's my light got a little indirect bounce lighting over here huh see that gives me a little
149. 22:36glow it makes me a little dead it's a little cold but whatever um what was I saying oh have I noticed um so the one thing I have noticed is
150. 22:46when you go to Bard um Bard if you ask a question that that it would make sense to incorporate a link like you know what what are some shoe
151. 22:57companies that make a certain kind of running shoe um it's incorporating links into the the GPT responses I've noticed that but I
152. 23:06haven't noticed normal Google doing doing their thing that doesn't mean they're not doing it I just my default search engine now is Bing
153. 23:17which a year ago I would have bet your money my default search engine would never be Bing but you know
154. 23:27it is because it's got gpt4 in it so I'm liking it okay let's see how are we okay have you used worm GPT I have not Kyle
155. 23:43Shannon in the Big Apple what up new joke in New York it's really good here I love New York the people are amazing oh you know what
156. 23:55I did so so one thing Monique and I my business partner and I were here and um we went down to Saint Mark's Place
157. 24:05um there's a there's a Japanese restaurant there that you know you you you went into it and all the tables were
158. 24:12kind of at floor level and and you would sort of sit on the floor and your legs would go under the table and this is
159. 24:17very traditional Japanese thing so we went to that restaurant and we're all excited to have this traditional Japanese restaurant
160. 24:23experience they've completely gotten rid of all that traditional [ __ ] it's just regular tables now it's like
161. 24:32but then so we went down there and we're staying up up near Times Square and um we um we got those City bikes the
162. 24:40electric ones those were a blast there's bike Lanes everywhere in New York now if you haven't been here in a while there's
163. 24:45bike Lanes everywhere so he rode up from Lower Manhattan to Times Square on these electric bikes those things are fast
164. 24:52so it was you know it's a little sketch because you know you're flying around at like 40 you know not 40 miles an hour
165. 24:58probably 25 miles an hour but you know close proximity to people in cars with no helmet so but it was fun um your area of expertise can be
166. 25:09enhanced and augmented by AI absolutely agree Winston I think these things are ability amplifiers um if you have not played with them if
167. 25:18you have not had what I call your Kevin McAllister moment with AI which is play with it until you have until it
168. 25:28does something that you just you literally can't process and then once you have that moment then you kind of dig deeper and you realize
169. 25:39um this is you'll be like oh that's pretty impressive but you know at least it can't do what I do and then it will
170. 25:49and then you get nervous and that and that begins your crack addiction to this stuff all right tick tock just jumped me to
171. 26:00the bottom of the questions because it loves me oh let's see uh what are we talking about debugging tensorflow no I'm not that deep into it
172. 26:16I'm not that deep into it I'm talking about generative AI more think post chat gbt as companies and people try to figure
173. 26:25out how to use these tools more so than companies and you know things trying to create these tools so um
174. 26:35this is what I did in the early days of the world wide web and the World Wide Web first came out I was not you know on
175. 26:41the W3 Council coming up with HTML tags I was using HTML tags to help Fortune 500 companies build their websites and
176. 26:50so that's kind of where I am with this stuff as well so my expertise is more in using the stuff once it kind of hits that
177. 26:56Prime you know ready for prime time which this is just hitting it so we're very early if you think you're behind on
178. 27:03this [ __ ] you're not we're very early on it but it's moving fast it's moving way faster than you know it's moving way faster than you know
179. 27:14[Music] I'll see you there did you get kicked out of your office no I'm in New York I'm in a hotel room a
180. 27:25salty little hotel room just off Times Square maybe I'll go for a walk later just people watch watch the [ __ ] Insanity of Times Square
181. 27:35good to know that you're a verified human thank you please experience explain your T-shirt please it is New York New York established 1898
182. 27:47Original Big Apple I don't think this shirt was bought in New York and I actually kind of hated wearing it out because if you
183. 27:58if you wear a shirt that says New York City and you live in New York City you're a tourist so but I live in Denver now so my wife
184. 28:07bought this at some previous trip she did and so this was the only t-shirt I had in my bag so there you go so I just went out of the tourist
185. 28:19all right any way to control a device via AI like Google Assistant um I don't know what you mean by control of Voice or control a device um
186. 28:33you could I mean I mean you could certainly you could certainly I mean here's the deal chat GPT knows all the
187. 28:40programming languages so you could certainly take something like a Raspberry Pi and have it generate a
188. 28:46bunch of code for you and then you could hook an API on the front end of that so you could talk to chat gbt and then
189. 28:55if you did the automation right it would send um commands to your Raspberry Pi so yeah I would think I would think you could
190. 29:04mock something like that up pretty easily and um even if you're not super technical you need to be a little technical but if
191. 29:14you can figure out a Raspberry Pi which is that's not that tough and then um Chachi BT can even teach you how to
192. 29:22do that just tell chat gbt say Hey listen I want to get into microcontrollers and I got myself a Raspberry Pi 4.0 point x Point r
193. 29:35and it'll know what that is and then just go tell it tell it to teach you how to do that [ __ ] so yes I think you
194. 29:44probably can I don't know anything about that that's now you're getting into Robotics and things like that is your salon
195. 29:49tomorrow after party in New York City oh that's actually a good idea so the salon is tomorrow um if you want to go to the salon go to
196. 29:58the salon.ai the first link there is the Meetup link so sign up for the Meetup and you'll get the login information for
197. 30:05the Google meet tomorrow night um John Knack from Adobe is speaking um I've got a dinner with a client so I'm probably not going to make the
198. 30:15beginning of it but I'll sort of join as a participant later but Leah my co-host is co-hosting with another another one
199. 30:21of the organizers and uh yeah we'll be Off to the Races tomorrow from seven to nine Eastern five to seven Mountain four to six Pacific so
200. 30:34yes but thank you for asking Casey have you wait I asked you several weeks ago about implementing AI in a high school
201. 30:40classroom I ended up giving a two-part lesson that ended with students creating a list of do's and don'ts oh that's really cool
202. 30:48and I've been showing them each day ways to use it responsibly responsibly so far so good that's super cool and I'm curious so I love that
203. 30:59newsy floozy by the way 9.5 out of 10 on the name like the name a lot the solid it's so it's solid it makes you smile
204. 31:07you got your Rhymes you got your alliterations it's intriguing you want to know more um uh yes zero my t-shirt is New York
205. 31:18I would be curious to know if there if their list of do's and don'ts the the high school students list of do's and don'ts changes
206. 31:26once they start using it right like like if they're new to it I would imagine the list of do's and don'ts might be
207. 31:32different than if they're you know in a month so be interesting to see how that evolves But but so first off like
208. 31:39[ __ ] Kudos so rather than hiding from this you're bringing it into the classroom you're getting them engaged
209. 31:46critically thinking I like this is what I talked about when when you ask the question if I if I recall correctly is
210. 31:53the most important skill moving forward is critical thinking because these things are going to do all
211. 31:59the Tactical execution so it's not about memorizing [ __ ] anymore it's not about learning these you know specialized
212. 32:06individual skills it's about okay we've got tools that can do all the Tactical execution what do we need to do we need
213. 32:14to like conduct the symphony we need to say what do we want to create what does success look like um if these things generate a bunch of
214. 32:24content how do you know if it's any good I don't know you look at it you use your [ __ ] brain and so I think you're you know this
215. 32:33feels very right to me right you're heading down this journey of okay kids these tools are here let's start
216. 32:40exploring them let's figure out what works let's figure out what doesn't what's it good for what's it not good for
217. 32:48and and you know you come back three months from now it's going to be a completely different answer I love that I think that's awesome um so
218. 33:00Kudos man oh Casey is also a teacher awesome hello hello hello nice yeah yeah that's fantastic newsy floozy are you a teacher me too
219. 33:15Sony is a teacher oh that's right Sonya I think you're in New York Sonia aren't you I'm in New York I'm in New Jack City um
220. 33:23Hotel I am at the Hyatt House I think I'm I don't know where it is somewhere near Times Square you could cross business wait you could
221. 33:37cross business analysis with generative AI use cases and do a small pilot project yep that's it that's a really good a good thing to do
222. 33:47um with code interpreter if you've not played with chat gpt's code interpreter the data analysis you can do with it is
223. 33:54stupid powerful and stupid easy and also frustrating when you're trying to do something really specific because
224. 34:01it kind of goes off and starts prompting itself and you ask it a simple thing and it's like let me try 16 ways to do that but it's
225. 34:10it's it's it's amazingly it's staggering to watch you're just watching this machine essentially start reasoning like
226. 34:19like code interpreter from chat GPT I mean gpt4 is is just mind-blowing in and of itself but then code interpreters
227. 34:28kind of Step Above that because you kind of get a glimpse into the future of what these things are going to do and right
228. 34:34now it's early enough that you can kind of it's all slow and you can see the code that it's writing and you can see
229. 34:41it making mistakes and trying to unmake the mistakes but you can kind of see when when this [ __ ] gets sophisticated
230. 34:50you'll just ask it for [ __ ] and it'll go aye aye Captain I'm off I'm off I'm gonna go get you your tickets you're
231. 34:59gonna have a fantastic time at the Rock concert in that obscure town you want to go to and off it'll go here you go here's your tickets
232. 35:11many companies are willing to peel off a few bucks for at least exploring this I yeah more and more are um
233. 35:19are starting to do that we did a we in a workshop that I just gave we did we showed a chart of um it was a McKinsey report on marketing
234. 35:31this was from back in April or May on marketing and sales that McKinsey did and 70 of the respondents said that they
235. 35:43expected in the future their companies would be using AI you know a lot or or a lot a lot right and and there was a
236. 35:51there was a companion chart that basically said what are you doing now and it was like not very much and so a
237. 35:57lot of companies right now are expecting they're moving in this direction they don't quite know how to do it yet but no one does
238. 36:06you know the fact that I the fact that I've got some Pharma Executives asking me for advice on what their agencies should be doing
239. 36:18um is telling right so it's it's we're very early very very early I heard chat GPT is going bankrupt um I don't know where you heard that is
240. 36:30that news today I doubt that's the case I'm sure they're spending daggering amounts of money um chat GPT bankrupt
241. 36:45news about chat GPT bankrupt declining users and competitors could make the is reportedly heading toward bankruptcy
242. 37:00tech.co what are the sources here I don't see any real Times of India who is this Tech talk
243. 37:22this is Yahoo news um
244. 37:37yeah I just I you know while it's possible I I don't I don't see that happening I don't see Microsoft letting that happen for one
245. 37:48thing they put in 10 billion dollars Sam Altman said he thinks they need to raise a hundred billion to do to get to AGI
246. 37:55artificial general intelligence um even even with their reduced usage like they're not making money right now on
247. 38:08the 20 bucks a month from us that they're just using that to cover their costs um they're going to make their money on
248. 38:14apis and and that sort of [ __ ] so um I don't know I wouldn't it like it's possible I'm sure at their current burn
249. 38:24rate I'm sure they're burning through staggering amounts of cash Sam Altman said when when uh when Chachi VT first
250. 38:32launched and it got way more popular than they expected it to he said the the costs of of what it was costing them he
251. 38:40said it was eye-watering or something like that it was just you know crazy turn on bard why what's oh turn on Bard in in uh
252. 38:53in Google search hang on google.com you mean go to bard.google.com yeah that's not really a search though let's see
253. 39:12I don't see a way to turn on Bard here but if you mean bard.google.com that's not really search that's just bard
254. 39:21uh or do you want me to show you Bart I'm happy to do that as well so let me just chat you but you've got lots of stuff Jason rusy
255. 39:29what's this news about it going bankrupt we just talked about that is a career in AI profitable um I don't know I think I think uh
256. 39:38we're all gonna have careers in AI like like like where we are right now is very similar to the early days of the
257. 39:46World Wide Web where you could ask you know is a career in the world wide web profitable in 1994 or five
258. 39:53I don't know if it becomes something sure um same thing with this like I I think because nobody knows what to do with these right
259. 40:04now everyone's trying to figure it out so right now if you can figure this shout [ __ ] out even a little bit people will pay you money
260. 40:13if you're a clever business person and a good salesperson they'll pay you a lot of money if you're not good at selling your
261. 40:21services and you can't figure out how to make money using Photoshop then it's you're gonna have similar problems doing this but if
262. 40:30you've figured out a way to solve problems and get people to pay you to do that then you're going to be doing great
263. 40:37with these things if you can get your head around them so um I think a career in AI is inevitable and I think the sooner you jump on
264. 40:46getting literate with this stuff and figuring out how you can take your skill set and solve other people's problems
265. 40:52using these tools to make yourself better smarter faster all that sort of stuff done great then you will and yes I think
266. 41:01it will you will be able to charge a premium for some amount of time because these things are very complicated and very impactful
267. 41:11um if I play it out even just three years into the future um people are going to be desperate to have people that understand what in the
268. 41:20[ __ ] are these things and how do we do them all right some comments in life were filtered to protect the community's experience don't
269. 41:30be jerks don't be jerks in the AI learning lab um do me a favor if you would share the live uh I just want to get some more
270. 41:37people in here if we can share the live if you're willing to and and hit the hearty heart button the the double tap
271. 41:43thing we'll get some likes going in here the other thing you should know I try to answer all the questions and I
272. 41:49can tell because I'm tired and my voice is shut I'm rambling tonight so sorry about that but you know that's that's
273. 41:55part of the charm of the channel isn't it if you know what I mean it's like you know what we do sometimes we just talk
274. 42:00we just talk we just have little conversations um and uh what was I gonna say I am now 58 messages behind so if I haven't gotten
275. 42:11to your message it's not because I'm a douche and it's not because I'm an avatar it's because I haven't gotten your message when I try
276. 42:18to read them all because I'm a conscientious Tick Tock Live host what's going on with the hair tonight I
277. 42:25got a different product in it I got some sort of mud that I carry in my little travel pack seems like it's holding
278. 42:31pretty good I mean you know it's it's still we got there's some issues don't get me wrong but it's hanging in there I
279. 42:38got a little Buster Poindexter thing going but this time of night that's to be expected all right let's see to show students how to use AI
280. 42:48as a writing assistant that will break prompts down into several smaller questions and then answers to the prompt also generate ideas newsy
281. 42:57floozy I love what you're doing I think that's really really flipping awesome Grim Reaper wrote something in Russian
282. 43:03or some sort of that kind of language is that Slavic it's not Slavic is it it's whatever Russian is but but it might not
283. 43:14be Russian it just looks in that neighborhood of languages uh I don't know how to copy and paste comments but but we can copy and paste
284. 43:23that comment if if someone wants to copy and paste that Russian comment and translate it for me assuming it's not some
285. 43:29evil nasty douchey thing um I'm happy to look at it
286. 43:45sorry I'm trying to find my where I was somebody put their phone number in there not sure what that's about
287. 43:56not sure that that's super wise but that's okay writing analysis writing analysis for GPT is far superior to Claude correct I agree gpt4
288. 44:09I think Claude and GPT 3.5 are close maybe clouds inch and up above it um gpt4 is definitely better but but
289. 44:21Claude you can put in a 75 000 Word document and it'll just do it or you can have a really long conversation with it
290. 44:29and it won't lose its memory for 75 000 words so Claude has advantages I I'm telling you I will pull [ __ ] off this list when
291. 44:40it doesn't have something that's unique to the others so chat EBT has gpt4 code interpreter custom instructions and
292. 44:50plugins so it's the leader it's the best right now thank you Luna stick welcome Moon even you Moon comedian you
293. 45:00um Claude has a hundred thousand token context something oh thank you for the Hardy heart I like the hearts hearts are sweet I like turtles
294. 45:09um Bing is connected to the internet Bing can also read images Bing can also generate images Bing is also what's
295. 45:18going to be the basis or not being Bing chat the basis for um the co-pilot product that's going to be inside all of office 365. Bard is
296. 45:29from Google it can read images it's connected to the internet and it doesn't suck as bad as it used to and is
297. 45:34probably going to get much better because they kind of [ __ ] up and are having to play catch-up to those guys
298. 45:41and so they're spending a lot of money um perplexity is a really interesting company they they've got good searches
299. 45:49with citations and they're letting us play with the Llama 2 models for free at labs.perplexity Pi just raised 1.3
300. 45:57billion dollars and Pi is an incredibly conversational bot which if you want to do work is a pain in the ass but if you
301. 46:05want to have a conversation it's actually really good it's really good at conversations and then Poe you can play with multiple
302. 46:12models and you can it's got a bot builder in it all of these have something to play with so yes I agree but
303. 46:20don't discount the others just because one's better at this than the other at some point someone's gonna roll a
304. 46:26bunch of this [ __ ] together we're going to have kind of a master GPT that does a lot of these things well and at that
305. 46:33point I'll jettison some of these things but right now worth worth and again don't think of any of these things as the final state
306. 46:44these things are all brand [ __ ] new this list used to be three like three weeks ago I only had three things on
307. 46:51this a lot of this stuff's new so just know that it's going to change it's going to change It's Gonna Change
308. 46:59Gonna Change It's Gonna Change It's Gonna Change all right oh wait by the way whoever just said that writing analysis uh MIM okay your
309. 47:11name is annoying I'm glad I don't stutter your your name is designed to make people stutter is that what's going on if it's that it's clever
310. 47:25um here's a trick I heard that I think is really good if you have a really long document use Claude to summarize it and and kind of reduce
311. 47:36the word count say say summarize this you know basically make this have less words and make it more efficient and then you
312. 47:45can make it small enough that you can take that summary to chat GPT and use chat gpt4 to do the true analysis of it
313. 47:52so use Claude as a distiller and then gpt4 to actually do the analysis that might be a cool way to do large documents best of both
314. 48:02all right our school has blocked chat GPT but not Claude ah interesting I'm preparing to push for chat GPT to be
315. 48:11unblocked good yeah do that the other thing news floozy I guarantee you they haven't blocked Pi um for your students that have uh
316. 48:19iPhones I think there might be I think Pi does Pi have a I don't know you can go check see if they have an Android app but but the
317. 48:28conversational um nature of of Pi could be really interesting for your students because it's so different like what you get from
318. 48:38chat GPT and what you get from PI are so different there might be something interesting they could do some sort of
319. 48:44analysis of what's different about it right and how are they doing what they're doing and how might they have
320. 48:49prompted pi to act like that and then you could even go to oh you you don't you can't you don't have chat gbt yet
321. 48:57well when you get chat GPT back Chachi PT has its custom instructions so you could go write custom instructions for chat GPT
322. 49:07to have it act like Pi that would be an interesting exercise does your wife pack for you uh laughing my ass off so ADHD I'm totally ADHD uh
323. 49:20yes she did help pack for me and yes and she actually had she had secret clothes hidden away in case I go to New York in
324. 49:28the summer so it's like khakis I look like I'm from [ __ ] Connecticut I have like a like a polo shirt and khakis and
325. 49:35I yeah I just look like yeah me wanting Raspberry Pi yeah exactly well I know I know that's they call it Raspberry Pi what are you gonna do
326. 49:51tis the season too the raspberries are coming pretty soon I asked mid journey to generate a chart of all edible
327. 49:58berries that produce something really nice that's super cool um if you haven't seen there's there's a there's a cool prompt in mid-journey
328. 50:06that is is one that um that makes me very happy I'll show it to you okay if you haven't oh wait I might have
329. 50:19slow crappy internet here in the hotel what's going on come on what are we doing are we working slash imagine
330. 50:33I don't know something's something something's gone South here uh let me quit Discord and see
331. 50:51if I relaunch it if it is less shitty do you know what a noling photograph is okay
332. 51:05n o l l i n g checking for updates I think my internet is poopy how's the live is this live okay give me give me some hearts if it's okay
333. 51:22uh uh or is it laggy yeah I don't think that's gonna work all right I got a couple of Hearts so it seems like it's okay all right
334. 51:32um it seems like my laptop is not happy it's it's just not happy it does not seem to like whatever internet they got going on here
335. 51:44and very disturbing it's very disturbing all right what are you gonna do um if you're in mid-journey you can also do
336. 51:57the I know there's some some lures for this for stable diffusion um type noling photograph and then a topic
337. 52:08so like Nolan photograph VW Bug noling photograph hunting supplies nulling or just noling k-n-o-l-l-i-n-g
338. 52:22um and so noling photograph is one where they take all the parts they'll take like in a like a they'll take an assembled something and
339. 52:31disassemble it and lay all the parts out in a really organized um you know uh grid layout and then they
340. 52:39photographed that that's what noling is so you can get mid-journey to do nulling on all sorts of [ __ ] like mushrooms or
341. 52:48that you know you said all the different berries or whatever it was and it did a pretty good job do do a nulling
342. 52:53photograph of all the berries and it'll lay them out really nice for you um Grim Reaper's still giving us some
343. 53:00Russian I I don't think I could can I copy I can pin this comment I can translate it oh Grim Reaper I can read your stuff
344. 53:13look at that we got I'll tell you what the AI just did the AI just I can read Russian I can read Russian now right now
345. 53:21why did chatbt invent new poisons but not a cure for cancer well okay GP chat gbt didn't invent [ __ ] if he didn't invent [ __ ]
346. 53:34human beings used these tools to do bad [ __ ] they also used these tools to create Alpha fold and created four hundred
347. 53:46thousand proteins in two days that is going to completely revolutionize drug discovery and they probably will find a cure for
348. 53:57cancer so it these tools didn't do [ __ ] people did [ __ ] if people choose to use these tools that are more powerful than
349. 54:08anything we've had in the history of humanity for bad [ __ ] well then whatever motivates them to do that that's their thing
350. 54:18part of what this channel is about is to try in this the little way we can to bend how we use this [ __ ] toward the
351. 54:26good bend it toward the light because I personally think and why I'm spending so much [ __ ] time like on my
352. 54:35trip out where I'm [ __ ] exhausted spending hours a night doing this is that individual contributions that use this
353. 54:45for the good right now are gonna have a disproportionately um positive impact on the future trajectory of this technology
354. 54:58I really believe that like I really like deep down believe that are there [ __ ] that are going to do
355. 55:04bad [ __ ] with technology yeah of course we're human some of us are [ __ ] some of us have ulterior motives
356. 55:15some of us are being paid to [ __ ] [ __ ] up and the rest of us are just trying to live our lives and if we can use these tools to live
357. 55:26our lives a little bit better [ __ ] far out and I think if we bend it toward the light maybe we outweigh those other actors
358. 55:41I noticed the detail under each Berry was made up of gibberish yeah so right now um um what's it called text text not so good
359. 55:55but I did find out from someone on Tick Tock I thought what was her name I forget it but hopefully she'll be in here tonight um
360. 56:07there's a thing you can do at clipdrop.co which is stabilities thing clip drop dot Co let's see if my internet is yeah the
361. 56:18internet's just hosed on this machine let me try turning off Wi-Fi and turning it back on maybe we got a little we got a little
362. 56:27situation here we got a little situation yeah isn't that special oh you know who's running the internet here at the hotel it's
363. 56:36Satan Satan's doing it now look at all those all those unwashed people looking at pornography in their little hotel
364. 56:43rooms well and that's special maybe they're eating up all the internet with their little porno pornographic
365. 56:49pictures of the the loot and lascivious acts no not special wow wow okay create stunning visuals in seconds
366. 56:59you can make money watch at gbt wait so that worked and then um oh someone just commented in my tick tocks they said they believe
367. 57:14websites will be obsolete soon I actually kind of agree with that and I'll tell you why in a second I'll tell you why
368. 57:23when my voice is really tired I can do I can do low low low low how low can you go well let me tell you boy
369. 57:33all right the the internet just sucks here all right so we're not gonna deal with that um if you go to clipdrop.com
370. 57:42who hasn't said that the stuff under the image so ion China if you go to clipdrop.co and use um what's it called uh is it sketch
371. 57:58no it's uh it's called doodle doodle to image or something like that anyway one of the tools on clipdrop.com
372. 58:07lets you doodle an image and then and then write a prompt and it will use your doodle as kind of the composition of the
373. 58:14of the image it does text really well like like not really well it does text decently predictably so I did the other
374. 58:24night I did like I did a square and then I had a little square inside it with a little triangle above it so I I did like
375. 58:31a sign hanging in a window and I put Help Wanted sign in a shop window and it generated three images and one of
376. 58:40them absolutely said help wanted in clear text the one was total gibberish and one was mostly help wanted
377. 58:46so I think with the sketch tool at clip drop you can start to do some predictive text it's not going to be that long
378. 58:53until they crack that nut they'll start layering in some of the stuff adobe's doing with their with in Firefly they've got a type
379. 59:01tool there where they're taken full-on fonts and like re-rendering [ __ ] on top of them at some point they're going to
380. 59:06start layering that stuff together in a prompt and you know sort of dynamically compositing it or or something there's also technology
381. 59:14that's they can make stuff more predictable now so there's some cool stuff coming but right now text sucks so you still
382. 59:23gotta take [ __ ] into Photoshop and fix it I he that that's the thing about these tools oh there's lots of people here
383. 59:29welcome everybody my name is Kyle Shannon this is the AI learning lab I'm not my normal uh thing I'm in a hotel in
384. 59:36New York because I'm here on business I'm tired as [ __ ] my voice sounds like [ __ ] well I don't know if it sounds like
385. 59:41[ __ ] but it just sounds like it sounds like I've been chewing on gravel yeah you know Tom Waits uh yeah let's go
386. 59:53why do I do this I don't know yeah anyway that okay welcome everybody she's my raspberry pie that's nice can
387. 1:00:08you explain why it's so obsessed with obfuscating text yeah I can actually so it's not obsessed with obfuscating
388. 1:00:16text it is not aware of anything do you want to see how images are created I have a presentation for that
389. 1:00:33let's see
390. 1:00:49yeah I had a feeling so I've got my my problem with my little mac is I've got um too much [ __ ] on it so my big files get
391. 1:01:04dumped up to iCloud do I have anything I have it anywhere no I don't I can't tell you how to do it well so so
392. 1:01:17here's the deal I'll I'll talk to you how the image generation things work so the way image generation works
393. 1:01:29there's two models that are involved there's the large language model right and the way a large language model works
394. 1:01:35is you put in your prompt and then it understands those words and it it doesn't it it actually doesn't understand those words it turns those
395. 1:01:46words into a mathematical probability of the very next word that should be the response or the first word of the response and
396. 1:01:54then it takes your prompt plus that word and it creates the probability for the next word so it is if you distill down what it's doing it's
397. 1:02:06actually a quite it's quite a simple mathematical function your prompt turns into a mathematical probability it creates a word creates a
398. 1:02:16word creates a word creates a word and it writes all that [ __ ] and it looks it looks like it's sentient and it looks like it's
399. 1:02:23reasoning and it looks like it's smart and it looks like it says it's a probability engine it's got the awareness of what it did the same
400. 1:02:32awareness that your calculator has when you type two plus two and it gives you four well how did it do that it must be conscious no it's a
401. 1:02:41[ __ ] calculator so so it's literally like that it's a fancy [ __ ] ass version of that but but that's what it is
402. 1:02:51so that's that's the the language part of of things like mid-journey and stable diffusion the image part at least for
403. 1:02:57the the diffusion model works like this Gans work a different way there's there's different ways to do it but for
404. 1:03:03the most part what we're playing with right now are diffusion models what diffusion models do is they take a bunch
405. 1:03:09of images that have been tagged with words okay so there's a cat in that one there's a dog in that one there's a desk
406. 1:03:15in that one there's right and then they they run every single image that's been tagged through a noise
407. 1:03:23filter they they noise it so it starts as an image and then it turns it into random noise through a particular noise algorithm okay
408. 1:03:33then when you type in your prompt it basically goes and says go find me all of the um the the images that have that were tagged with any
409. 1:03:48words that are relevant to that prompt so it's using the large language model as a filter to filter out right so maybe
410. 1:03:54you go from a billion images that are in the training set and you filter that down to you know a hundred thousand
411. 1:04:03it's not seeing the images the images are these noised random noise patterns and so when you put in your prompt it
412. 1:04:11filters it down and then it averages those hundred thousand noisy images into a single noisy image and then it uses that same noise
413. 1:04:19algorithm to denoise it toward toward whatever it looks like it's not aware of all of the pieces of individual grains of sugar if you think
414. 1:04:31of every picture in the training set as a grain of sugar but you dump it all into cake batter mix and Stir It Up with
415. 1:04:39eggs and butter and [ __ ] and put that in the oven you don't know which grain of sand generated to that so all it's doing is
416. 1:04:47denoising some random pixels into something and so the reason text is garbled is because it might know that it
417. 1:04:57needs text but it's meshed together you know many many images and is you know kind of just denoising them it doesn't know what's in there
418. 1:05:08it's got no awareness of what it's doing it's just a mathematical calculator so which I like once you understand how
419. 1:05:17they work I think it's even more remarkable that they do what they do but but that's why I can't do text so in
420. 1:05:23order for it to do text they're gonna have to do some more specialized algorithms that maybe say okay you know
421. 1:05:30start denoising the image and then go look at the text on it and maybe you have a separate layer that's doing text
422. 1:05:36in the context of that image so they'll be more sophisticated ways to generate images in the future but
423. 1:05:43I will leave that to the mathematicians I mean whoever figured this [ __ ] out is it's amazing it's amazing it's just amazing it's
424. 1:05:54amazing what time is it it's getting late here in New York City it's getting late late my voice is going
425. 1:06:03sorry to hear that it's banned there oh I know that sucks the chat GPT police that really sucks hi I need to improve my English go to
426. 1:06:11these things they're really good at it put in your language oh you know what you can do uh Lati latiz Latisse here's what the Doula tease
427. 1:06:24in your language in your native language go to these things and have it create custom um tutorials for you
428. 1:06:35the way you like to learn like if you like to learn with flashcards have it give you flash cards if you like to
429. 1:06:40learn with you know something like Babel or you know one of those things um have it give you lessons and it will
430. 1:06:48and I would even imagine you could go to Pi and just talk to pie [Music] Translate all right Grim Reaper medical
431. 1:07:00corporations won't let Chachi PT come up with a new cure for diseases you know why well yeah that capitalism right they you know like
432. 1:07:08Pharma companies don't want to cure for cancer again medical corporations are not going to not let chat GPT come up with it
433. 1:07:24what big corporations will do if they want to squash something is they'll let again chatgpt is not coming up with this
434. 1:07:30[ __ ] human beings prompting chat you're coming up with this [ __ ] if one of those people comes up with a cure of cancer a cure for cancer
435. 1:07:39but those corporations might do is buy that technology and then bury it but they're not they're not stopping
436. 1:07:47this [ __ ] from doing it and the fact that meta open sourced llama 2 means that those big ass corporations
437. 1:07:56can't do that like let's say that the big Pharma companies could strong arm open Ai and Claude and Bing and barred
438. 1:08:09and Pie and Poe and everyone else that's coming and they can never do anything that would cure cancer well you still got
439. 1:08:20llama 2 out there in the public domain without any safety guard rails on it they can't they can't control that those companies control it
440. 1:08:29after the invention not before I think yes exactly I told them it's a living document and we will revisit it
441. 1:08:38throughout the year that's cool any good AI tools for short form video editing other than video or Opus I don't know I I haven't
442. 1:08:48I've been doing a lot of that I I would just go to futurepedia.io go to the video section there and you know learn
443. 1:08:56just start playing and just know that most of them are going to be a waste of your time there's probably a lot of
444. 1:09:03video tools but there might be something good in there um videos the video is the one that I've used
445. 1:09:12but even that I mean the problem with it is they aren't cheap they're really [ __ ] expensive which I I suppose it's expensive in compute
446. 1:09:21but I have a feeling it's just expensive because they can right now there's not a lot of competition [Music]
447. 1:09:29did your meeting go well one went really well really new big opportunity and one the [ __ ] dude didn't show up we flew
448. 1:09:40out here to meet with two people one of the guys today didn't [ __ ] show that like if there was some sort of emergency great
449. 1:09:53if there wasn't that is [ __ ] classless there will be manipulation with AI Grim Reaper yeah I don't know what you mean you mean with
450. 1:10:09content yeah there's gonna be a lot of manipulation thought you were wearing a Trekkie shirt now it's New York City
451. 1:10:16the Indian time says they may go bankrupt by the end of the year is the end is the Indian times um respectable I don't I mean I don't know
452. 1:10:24I don't know that sounds it sounds respectable um it's possible that that open AI goes bankrupt by the end of the year
453. 1:10:33I can't imagine that Microsoft or anyone else is going to let them go bankrupt unless Microsoft has some poison pill
454. 1:10:43that says if they go bankrupt they own it or some [ __ ] like that which is possible depending on their investment rights
455. 1:10:52I just don't see it happening investors will come in it's it's too important a technology all all of this competition would not
456. 1:10:59exist had chat CPT not did what it done what it did did what it done did what it did I'm tired uh 10 billion over how long 10 billion
457. 1:11:14see I think most of the 10 billion from Microsoft was in compute credits so I would imagine that they burnt through
458. 1:11:23a lot of that 10 billion dollars already cost 700 Grand daily yeah yeah so so so they're they're blowing through the cash
459. 1:11:33right every time we do one of our stupid [ __ ] write a poem that sounds like a cowboy got a rock in his shoe every time we do
460. 1:11:43that it costs them a couple of bucks like it ain't cheap to run these things um they're too important they're they're
461. 1:11:50too important to let it die in my opinion I like that just seems like a doomsday kind of sensationalist thing like if if listen if I haven't
462. 1:12:01read the article if there's a quote from Sam Altman saying we're in trouble great but if this is just one of those
463. 1:12:07speculative things that's probably just someone trying to buy some time to build a competitor to chat gbt uh
464. 1:12:16uh yeah 700 Grand a day is a lot to operate something they'll cut back first quarter spending don't expect much from AI yet
465. 1:12:26um this is a great intro to wait this is great for introduction and profit taking wait most tickers wait what this is
466. 1:12:35great for introduction and profit taking most tickers 10 from highs I don't quite know what you mean but I
467. 1:12:46kind of know what you mean yeah so so yeah user just dropped off with Chachi BT things are down a bit wait listen
468. 1:13:00if what we get to play with today is this powerful do you know the [ __ ] that they're playing with that we don't
469. 1:13:10get to see I mean Sam Altman said there's something they developed at chat GPT they will never release it's that powerful what is that
470. 1:13:22the CEO of stability AI says the image generation tool that they're playing with internally it's stability AI is
471. 1:13:30generating images at 60 frames a second so it's doing 60 frame a second video that's the tool they're playing with
472. 1:13:39inside open AI right now or in inside stability AI right now the [ __ ] that that we haven't seen yet is
473. 1:13:48like okay this stuff's not ready for prime time okay great it's not but you wanted to write some social media copy it'll [ __ ] knock it out
474. 1:13:58it'll get you to 80 of what you need in five minutes and then you can spend the rest of the hour refining it
475. 1:14:07I don't know man I don't know I would I would not piss on the grave of generative AI just yet just because there's a little
476. 1:14:16there's a little Summertime Blues and and there's not as many users as there were in the in the frothy aftermath of it launching
477. 1:14:28um have you tried out goblins.tools yet goblin.tools yet yeah I have so I would show [ __ ] here but I can't
478. 1:14:36unless my internet has gotten better now it really hasn't look see it working let me let me try one more time
479. 1:14:44maybe they reset The Rounders because people were complaining no no it's still shitty um yeah goblin.tools is a to-do list
480. 1:14:55maker that you just put in a to do and then you hit the magic wand button and it will break it down like a full-on
481. 1:15:03project manager it's pretty amazing um it with my ADHD to-do lists give me hives and goblins to-do lists really
482. 1:15:13give me hives because it's like you know I just want to accomplish this it's like here's the 73 steps you need to do to
483. 1:15:20accomplish that yeah I'm good I'll just not get [ __ ] done I'm going to write a book not get [ __ ] done
484. 1:15:31the way of the the way of the add wizard let's say DD people we get [ __ ] done when we [ __ ] want to get it done and
485. 1:15:41it's always at the last minute prefrontal cortex block of cheese from here forward nothing but net
486. 1:15:56I'll take a cross-section here a little brain scan it's just like great Bondo hairdo yeah I know it's the
487. 1:16:08it's it's that mud [ __ ] like like it's a little teeny tin and you don't think it's doing anything but then you put it
488. 1:16:14in and it is it's like [ __ ] Bondo it's pretty funny thank you oh just jumped me to the bottom hey dang it dang
489. 1:16:21it tick Jack why are you doing that to me it's just makeup it's just rigm friggin man let's rock them suck them you son of a [ __ ]
490. 1:16:32I want to meet Mr Tick Tock and I want to say when I report someone for copying my [ __ ] account and trying to scam my
491. 1:16:41followers and I report that four times why is that okay why are you just saying yeah they're fine what are they doing to you
492. 1:16:50yeah and then why did you jump me to the bottom of the comments when I'm just trying to scroll through them why why why
493. 1:16:58someone programmed that it should be fired right now [Music] all right I think I found my place oh there was the Bondo in the hair comment
494. 1:17:08the Russian translation is you two can make the Russian translation you can make money watch every day that's pretty
495. 1:17:16good Tobias was shaking man all right Luna stick good evening if you hold down a comment there's an option to translate he said AI will
496. 1:17:28bring manipulation yep got it I figured it out old guy figured out tick tock stop the presses want to stop the presses mean well see
497. 1:17:39in my day there were a thing called newspapers and they had these things called presses where they pressed ink into paper
498. 1:17:49and then if there was breaking news someone would yell stop the presses and then they would literally stop the presses that's what it means
499. 1:17:59yeah [ __ ] Jen's ears with your God damn it everything's there instantly all the time it's not how it used to be dang it
500. 1:18:15sup buddy chatgpt4 welcome welcome welcome Emilio's wife many dinosaurs awesome there thank you welcome everybody let's just joined
501. 1:18:25um I am definitely many questions behind down below down in the below times more than 100 questions behind so if it seems like
502. 1:18:36I am just talking about random [ __ ] that's why that's why because I'm trying to answer all the questions
503. 1:18:45isn't that good is mid-journey not good anymore it's not on the list it's on the list it's just not on this list
504. 1:18:52oh wait when was the last time you were here it's that Buffalo okay so so here's what here's what happened it's that
505. 1:18:59Buffalo mid Journey used to be on this list but this list only had three chat GPT things on it and now it's got seven
506. 1:19:09so so I made one page for chechi PT [ __ ] and then I made another page for prompts.chat the prompt book for images
507. 1:19:20and future pedia and then I put all the image generating tools down at the bottom here so I got mid Journey Adobe
508. 1:19:26Firefly Leonardo and clips uh clipdrop.co which is stable diffusion stability ai's little experimental portal um so yeah a good eye
509. 1:19:39uh no it did not go away mid journey is still the [ __ ] and no one's caught it yet and I don't think they will for a
510. 1:19:46while but stable diffusion XL is close but it ain't there um but yeah it's it's we we have a situation where there's there's just too
511. 1:19:57much [ __ ] here but but we need it there until further notice thank God I can change my name back tomorrow what how
512. 1:20:04did you get stuck changing your name to chat gpt4 and why can you not change it back is this just a special tick tock
513. 1:20:10tick tock being [ __ ] special kind of thing Chad gbt4 changing to Claude Dr J coming in with a solid one there Grim Reaper is
514. 1:20:21saying a bunch of random stuff in Russian like why does chat GPT share poison info yeah I saw that Winston thanks
515. 1:20:29um just watch your verified video too bad you didn't get the free crypto thanks Gary yeah I know yeah Gary
516. 1:20:35ginsler thanks Gary thanks yeah you know what I feel so much safer that I didn't get that free money that Sam Altman
517. 1:20:41wanted to give me I I just it warms my heart to know that I'm not going to be ripped off by someone who wants to give
518. 1:20:48me money thank you you're the best bitter you betcha all right the chat GPT doesn't cure cancer not sure why he's using Russian I don't
519. 1:21:04know why he's using Russian maybe he's just trying to see if I can figure out how to translate [ __ ] but I figured it
520. 1:21:08out there I figured it out because I'm like an old guy with tech chops but not real technical chops but like I can
521. 1:21:17point at [ __ ] and click on it you don't think that takes skill it doesn't it's not the point that's not the point all right
522. 1:21:30good day Christian Martinez yes the shrewd one heart Dr Jake heart heart heart Joy heart heart thanks for the hearts this live is great
523. 1:21:40no lag when you switched cameras oh really weird huh fine life is good many hearts no lagging for All of Me
524. 1:21:53better here than in Colorado isn't that [ __ ] awful I'm in a hotel hotel Wi-Fi is notoriously shitty and the [ __ ] I buy
525. 1:22:01at home that I spend way too much money on is crappier than this the hell why am I chewing my cuticle because it's [ __ ] tasty
526. 1:22:13stop picking on me weirdos Zen AI oh I can't look at the internet because the in what okay you know what I'm gonna do I'm gonna restart my
527. 1:22:22machine that's what I'm gonna do because that's gonna fix it that's gonna fix it there is more than one way to get the internet I could also
528. 1:22:30tether it to my phone couldn't I let's let's restart it and see what what happens all right show me how to make money
529. 1:22:40online with AI you can make money with chat so Giovanni Hendrickson Henriksen um I have a point of view on this
530. 1:22:54and my point of view is that making money with AI or making money with chat gbt is not it ain't about the tool it ain't about the tool
531. 1:23:10what do you mean by that Kyle what do you mean it ain't about tool here's what I mean about that um what did you say there show me how to
532. 1:23:19make money online with AI the answer to that question Photoshop beta failed to quit why okay we force quit it screw it we're out of there
533. 1:23:31[Music] make money online with AI you can replace AI with any other tool and the answer is exactly the same make
534. 1:23:42money online with Photoshop make money online with HTML make money online with CSS make money online with project
535. 1:23:49management make money online with shovels they're all tools how you make money is you figure out a problem someone has
536. 1:24:07why is it not using my fingerprint what just happened to my machine did one of you [ __ ] hack me and screw up my machine
537. 1:24:15I'll be pissed you're in here speaking Russian someone's probably hacking me taking me down all right that thing's back let's see if
538. 1:24:27let's see if uh where's here's clip drop let's see if we get some some internet back um so so how you make money with any tool
539. 1:24:40is you figure out a problem someone has you figure out what your skills are you identify that problem you're like I can fix that problem
540. 1:24:50and many times it's like I can fix that problem in ten thousand dollars and if you fix that problem good enough and
541. 1:24:57it's worth ten thousand dollars people give you money they give you the ten thousand dollars and then you go fix that problem
542. 1:25:04and with this generative AI [ __ ] maybe you can fix that problem faster for five thousand dollars so you can go into the market and you
543. 1:25:12can say hey they're fixing that problem for ten thousand dollars I'll give you double the results for half the money but you
544. 1:25:21gotta sign today and then you just made five thousand dollars and you use chat GPT to do it faster than those people doing it by hand
545. 1:25:30and then you're laughing all the way to the bank and you're going I made money with her but you didn't start with Chacha Peppa
546. 1:25:39did you you start with the problem so I know that's a smart ass answer but that's actually the real answer and the
547. 1:25:46reason I make fun of people that say You're Gonna attach if you take is that how they're making their money is with fear
548. 1:25:57wait a minute someone has a secret for how to make money with chat gbt I can't [ __ ] figure it out they're gonna give me a course
549. 1:26:07and it's only 299 dollars per session amortized over 18 months and if I sell that course I can make money no you're making money selling
550. 1:26:20courses that's the that's the oldest trick in the book They're not making money with Chachi Pate other than the fact that
551. 1:26:28they wrote the course for chatgpt about how to sell chat gbt courses to make money for Chachi see how circular it is
552. 1:26:35it's circular there's pyramid schemes and there's circular marketing schemes circular marketing and if they got really clever they'd
553. 1:26:46make it a ah the bear gave me the dangly lights I love the dangly lights um if they were really clever they would
554. 1:26:53turn their circular marketing logic into a multi-level marketing scheme and they could have a pyramid scheme on top of
555. 1:27:00circular marketing logic and they would really make money watch activity if you're at the top of the pyramid you see how bitter I am
556. 1:27:14you know why because people like that make me [ __ ] crazy go generate some [ __ ] value in the world and people give you money for it
557. 1:27:32there's no easy answer none none you've gotta do the [ __ ] work you know what I'm saying you come in here you expect the easy Road out
558. 1:27:47what are you gonna do you gotta [ __ ] walk a mile then you gotta walk another mile then you gotta walk another mile what have you done then
559. 1:27:58you walk three miles you're [ __ ] thirsty and your feet hurt you still haven't made a [ __ ] dollar walk a thousand
560. 1:28:09figure out problems solve them get literate with this [ __ ] because here's the deal let's say what's your name oh Giovanni Hendrickson
561. 1:28:19let's say you're really good at project management learn how to use these tools to do project management better than your competition
562. 1:28:28and you probably can these things are ability amplifiers what's the [ __ ] that you're good at that you enjoy doing that you can talk about
563. 1:28:39and then figure out how to do it with this cheaper better faster you know that old adage you can have a
564. 1:28:46good faster cheap you could only have two of the three I think with these tools you can have three of the three and right now
565. 1:28:54in today's world it you can use these tools to make money and not even tell anyone you're using AI you just walk in and go hey you know how
566. 1:29:06your competitors say you can have two of the three good fast and cheap I'll give you all three what's that mean
567. 1:29:15it means I'll deliver what they deliver twice as fast for half the money try me try me for a month see if I can't do it
568. 1:29:29and then do it and if you can do it then do it again and then do it again and then do it again and when they pick their jaw up off the
569. 1:29:37floor going how the [ __ ] did he do that you'll be on to the next thing and you will have made money which at
570. 1:29:45gbt my light just went dim it went dead do I have a I think I have a thing I think I have a thing it doesn't really matter
571. 1:29:55I gotta wrap this up anyway I'm tired tired [Music]
572. 1:30:14all right Green Acres is the place to be hi living is uh something I see anyone Green Acres
573. 1:30:34[Music] there see that I'm not shiny again I'm all shiny and blue I'm like I'm like corpse over here and something over there okay
574. 1:30:49cures for cancer already exists but I'm not a healthcare professional yeah there's that new there's LK 99 and then
575. 1:30:54there's some other numbered like G gr something or G something R 1066 or I don't know what it is there's supposed
576. 1:31:03to be some pill that cures all the cancers yeah listen we're gonna get free energy we're gonna have levitation we're gonna have flying
577. 1:31:11cars we're gonna have cures for cancer we're gonna solve climate change we're gonna just just figure people
578. 1:31:18the world can't be made in a day but it can be made in a year with these things or not all right let's see what time is it 10 30.
579. 1:31:30what am I doing with my life laughing my ass off church lady well I'm not special hmm
580. 1:31:45isn't that terrific isn't that terrific so the machines are going to do your work for you whatever happened to good
581. 1:31:52day's work for a good day's pay everybody wants to get rich with chat JPay yeah okay oh wow wow [Laughter]
582. 1:32:03Satan runs the internet everywhere oh [Music] stable doodle that's what it was called Emilio's life yes clip drop stable doodle
583. 1:32:16clipdrop.co that's very good info thanks yeah I wish I could show it to you because it's cool oh wait what's going on over at the yeah salon
584. 1:32:24everybody figuring it out oh there's my little doggy um reload do I have internet oh I might have internet wait it's really slow
585. 1:32:41no no they still sucks it still sucks still bad it's still bad am I on the right Wi-Fi yep [Music] hello from Bulgaria yuhi
586. 1:32:58yuhi I think that's right hi the Bears here awesome Greta gloves the prompts.chat link you shared has been very helpful thanks
587. 1:33:10enjoying and learning from the lives that's awesome thank you oh look clip drop just showed up here's stable doodle all right oh
588. 1:33:21the internet's working let's go play hahaha oh but I I actually granted thank you so much for acknowledging that that helps because
589. 1:33:34the whole point of me showing all that [ __ ] and repeating it over and over on these lives is that I think this stuff
590. 1:33:40helps so by you letting me know that it helps that's actually great so thank you for that okay so someone earlier asked
591. 1:33:47about why doesn't mid Journey make text well and I explained why it doesn't do it but last night no two nights ago someone on
592. 1:33:55the live said but stable doodle does good text and I'm like that can't be true but Auntie Jane I didn't think image
593. 1:34:04diffusion models could do text well well Jimmy Okay so here's what I did I went you're gonna watch my
594. 1:34:15lovely illustration illustration skills one of the reasons that I'm so [ __ ] excited about mid-journey and things
595. 1:34:21like that is this is about the level of my illustration skills yeah all right so we have a
596. 1:34:35we have a sign in a window okay and now we're going to type in a prompt um let's see um um buy one get
597. 1:34:50one free sign in shop window all right so hopefully this will say buy one get one free so let's see let's see what happens that
598. 1:35:07actually seemed like pretty peppy response I think I got full on internet here how exciting is that now watch it'll just sit here for hours
599. 1:35:18I'm also thirsty and I ran out of water what am I gonna do what am I gonna do people what's happening I'm right near Broadway
600. 1:35:31I could go out and just sing at the top of my lungs right now because you're in New York City and people wouldn't even bat a [ __ ] eye
601. 1:35:39here we come to save the day I'd probably get Applause for that one all right I don't know what's happening with this
602. 1:35:48we'll move on to another question you speak like a voice actor nearly Punk well thank you very much oh who did who deal was I loved it
603. 1:35:59hey everybody hi Kyle in New York City hey bear what's happening um I appreciate the the uh the voice comment
604. 1:36:08Barry was Barry's last name my voice teacher at Penn State would would appreciate that comment he taught me to took talk good words
605. 1:36:19with the good voice talk with words in a good way use your words Luke I am your father we speak in tongues why do we speak in
606. 1:36:34tongues why not um if it seems like I'm getting Punchy that's because I'm exhausted [Laughter] and my vocal cords feel like someone ran
607. 1:36:47a cheese grater over them but when that happens I can talk nice and low like this welcome to the AI learning lab how can I help you
608. 1:36:59well isn't this fantastic aren't people that talk like that all the time wild I used to play uh poker
609. 1:37:06with Patrick page who's a Broadway actor and he just talks like that he's like hey could you could you hand me the
610. 1:37:14chips you're like Jesus it's like every time he talks it scared the [ __ ] out of me he's an amazing guy though
611. 1:37:23apparently clip drop is just not gonna [ __ ] cooperate I'll come back to it but whatever what are you gonna do what are you gonna
612. 1:37:31do it's not my setup and even though my setup at home is a little janky it works it works my hotel set up here I got my
613. 1:37:40computer up on a pillow it's very sad oh all right stable doodle is fantastic
614. 1:37:55everyone's welcoming the bear the bear came in got some hugs nice concache PT think of something new is it creative um it okay well hmm
615. 1:38:07thinks of something new in a similar way that humans think of something new in that we're all large language models so the
616. 1:38:18the G in GPT stands for generative so when you give Chachi BT a prompt the thing it's giving to you is original content it's generating word
617. 1:38:32by word a response to your prompt that is just by math different than what other people are going to get um
618. 1:38:42if you prompted in a clever way it will give you new ideas and you know like like if you think about how when you listen when you listen to
619. 1:38:56rappers talk about how they think about language and how they [Music] um put phrases together and Rhymes together
620. 1:39:05and they figure out how to break up language in ways that they can they can make Rhymes out of words that don't
621. 1:39:11actually rhyme but if they you know stack them in certain ways and then you also hear them talk about well you know
622. 1:39:18one technique I is you know I'll combine this concept with a completely you know nonsensical concept and then I'll see
623. 1:39:25what comes out the other side you can do that with these large language models you can say you know um take jungian archetypes and combine
624. 1:39:39those with Latin names of fungi and you know write me a story where all of the characters are named after mushrooms and you know the jungian
625. 1:39:55archetypes are put in there and it will [ __ ] do that has anyone done that before I don't know probably not
626. 1:40:07is that creating something new well I would say that you have created something new using a tool that if you had enough time to research
627. 1:40:21all of the inputs that you just described take all of the young Ian archetypes and all of the Latin names
628. 1:40:27for fungi and turn that into a story that matches the Star Wars plot it will do that you could go research
629. 1:40:37all that [ __ ] you could do that for yourself and it would probably take you a year and you can just say to these things go [ __ ] do that
630. 1:40:48that's how creativity works right but you as the human are the ones still prompting it and goading it and figuring
631. 1:40:57out thinking creatively again I think creative thinking is the skill moving forward that is the valuable skill tactical execution
632. 1:41:08not as valuable creative thinking strategy curation taste so yeah I think it can be creative when it's paired up with a creative human
633. 1:41:23ask it a shitty question it will give you shitty answers ask it a really creative question structured in a really creative way
634. 1:41:31it'll give you stuff that no one's ever [ __ ] seen before and it'll do it like that so yes I think it can do something new
635. 1:41:41we are large language models that is a large language model in fact architecturally how the neural net Nets work
636. 1:41:53why they are the way they are they tried to mimic our neurons now our neurons do stuff in a much more sophisticated way than these things do
637. 1:42:07and there's many more of them but I know what's going on you know is anything a human does truly creative
638. 1:42:19isn't it all derivative from [ __ ] you've learned in your life yeah but that's human okay it's language it's words words words words words
639. 1:42:31words words clip drop you disappoint me my hotel internet has disappointed me
640. 1:42:46bah humbug I'm gonna try again I'm doing a window oh no oh let me do it look I just did creative all right it's a room now it's a room
641. 1:43:04I don't know what I'm doing oh yeah this could still work [Music] help wanted
642. 1:43:20sign in shop window all right we'll see if it does its thing I have a new car oh not me law huh what what about when it was asking me which
643. 1:43:35boxes have traffic lights I don't know what you mean is that some sort of inside reference to Schrodinger's cat is that a quantum
644. 1:43:44physics joke or am I just a dum-dum I might be a dum-dum it's possible I'm a dum dum am I a dumb dumb there's 100 people in here and I'm
645. 1:43:58whacking my stupid [ __ ] head with a bottle all right um thanks that was helpful explanation on image generation oh by the way I'm
646. 1:44:06very far behind on my questions here very far behind because I was talking about image generation like 45 minutes ago so
647. 1:44:14if I'm not answering your question that's because now I'm an old man I'm just an old man I do my best
648. 1:44:25I get on The Tick Tock and I do my best I click on interface elements and I've read the manual the manual is it's dry
649. 1:44:37reading but it's very informative so I've read The Tick Tock manual and uh it's a it's uh you can get it at the
650. 1:44:43library uh I mine's slightly out of date my mine was from 1982 was the was The Tick Tock manual that I'm I'm it was it
651. 1:44:52was the Tic Tac manual that was that was how to make Tic Tac candy um right anyway
652. 1:45:06does NASA runs their AI on d-wave quantum computers maybe I don't know I ain't been to NASA I don't have access to their Quantum
653. 1:45:17Computing lab and and and I ain't said to them hey NASA are you running your Quantum compute are you making a few
654. 1:45:26reversed engineered alien technology with your the rocket Parts you got off the moon from the moon base that you
655. 1:45:35didn't go to the moon but there was rocket Moon based alien people on the moon that you didn't go to I'm confused
656. 1:45:43now could you use your quantum computer to help me figure out which conspiracy theory is real or not because it's the
657. 1:45:51polar ice caps have the hollow Earth and but it's it's it's a flat Hollow Earth so wait if it's a flat Earth with a
658. 1:45:59Hollow Middle is is Earth a donut I think or it's a donut in the Moon is cheese with alien technology that we
659. 1:46:08didn't go to from the Mars with the face and the pyramid on Mars I'm so confused I I try to believe it all and but then
660. 1:46:24um I think the Earth's donut I think I read that I think I read that on the internet it was a quantum computer that figured
661. 1:46:31out the Earth was a donut yeah and but all the pilots are like the pilots that flew over it that don't see
662. 1:46:38a donut that that's a lie they trained them they took them to the base in the four they call it do you think it's Area
663. 1:46:4751 but it's not it's S4 and they took the pilots there and they said listen when you fly um we're gonna give you video to put on
664. 1:46:56your phone just don't tell them it's a donut because it's the hollow Flat Earth it's the hollow Flat Earth with the waterfall
665. 1:47:04in the ice ring yes so I don't know if NASA is using d-wave quantum computers for their calculations to return to the Moon that
666. 1:47:19they didn't go to in the first place to get the aliens that are the reason they left and never went back because they're
667. 1:47:25afraid yes I don't know I don't know I don't know I don't know we just talk about AI here we don't talk about all
668. 1:47:32that stuff I don't talk about NASA I'm talking about Neville Kai is teaching me Spanish at the moment oh that's awesome Chachi BT well
669. 1:47:41businesses eventually create their new local they're now local chat GPT that focuses on content that they train yes I own China
670. 1:47:50that to me feels like the killer app for all these things so right now these things are remarkable because they're kind of reflections of
671. 1:47:58humanity right but they're not Reflections of My Humanity they're reflections of all humanity and they're
672. 1:48:05not reflections of my business they're reflections of all businesses so when I can just point it at my data and say
673. 1:48:12tell me more about this thing I created tell me more about the person that I am and what I said out in the world I think
674. 1:48:19that becomes the killer app I think I really do I think that's that combined with Automation and autonomous agents
675. 1:48:28that's when it starts to get really fun and probably skip up the hair just the glue just broke look at that that's a
676. 1:48:36[ __ ] firework display on top of my giant aging bald [ __ ] head all right [Music] um so yes Kyle I how do I get barred to
677. 1:48:47review and analyze a PowerPoint which I've uploaded on Google Drive you don't write now you wait basically um
678. 1:48:58if you put the PowerPoint in PDF so Bard can't read Bard can only read images so so so this is [ __ ] up okay
679. 1:49:14all right hang on okay I learned I learned this stupid and
680. 1:49:29ancient so you don't have to okay here's what you can do so how do I get bored okay so you've got a PowerPoint presentation
681. 1:49:41right yeah okay you got a PowerPoint presentation what can you do with that if you can get the your uh you turn
682. 1:49:57first you turn it into a PDF so you download it as a PDF you could upload it to code interpreter and have it extract the text out of it
683. 1:50:08at Chachi BT you can upload the PDF wholesale to Claude cloud.ai and it will probably read it it won't be able to read the images being
684. 1:50:21I don't think you can do [ __ ] with it barred if you took screenshots of the PowerPoint and put those how would you do that
685. 1:50:40like so so Bard can read images um Bing can read images Claude can read files poke and read files
686. 1:50:54code interpreter and some plugins can read files at chat.openai.com I don't think I answered your question
687. 1:51:02well basically turn it into a PDF and upload it to Claude that's your answer don't worry about all the [ __ ] I said just
688. 1:51:11turn it into a PDF upload it to cloud and start interacting with it and Bard will when when they roll barred into the
689. 1:51:21office suite of Google then then it'll work but it just doesn't now and I feel like but Kyle isn't that stupid yes
690. 1:51:31I I think it's stupid that I have to have seven things on this screen right now I think that's really stupid but that's where we are so
691. 1:51:41welcome to the early days this is I mean this is what it was like in the early days of the web it was like you know
692. 1:51:48nobody can figure this [ __ ] out because all the technology sucked you know we were like wait is it this is bad right this is bad
693. 1:51:57yeah this is bad but companies are still willing to pay us to make these stupid websites yeah a lot of money
694. 1:52:04more than we ever imagined all right same [ __ ] here and then five years from now the world will be transformed because of this [ __ ]
695. 1:52:15and if you're one of the ones that took some time right now to get your [ __ ] head around this stuff and got a little
696. 1:52:21AI literate you made good shape you're gonna be in good shape you'll be like yeah man I'll tell you what in the early days man
697. 1:52:29there was this tick tocker his name is it doesn't matter what his name was but he was old as [ __ ] man and he was just
698. 1:52:35talking like you just like the world wide web and then I was like ah [ __ ] you man and he's like you're just a [ __ ]
699. 1:52:41old man it's like Elder talk you're just [ __ ] like talking to people and anyway he was like oh man and he didn't
700. 1:52:47know how to use tick tock or nothing he like read all the comments he was like dumb anyway so I started looking at chat
701. 1:52:53gbt it's like let's chat to EPT it doesn't matter right now but I started playing with this thing I got pretty
702. 1:52:58good at it man and it and then that's why that's how I became a billionaire you know why because all right
703. 1:53:12that was a lot of a lot of wind up to get to a horrible punch line wasn't it all right I've created some new and
704. 1:53:18interesting math by mining information so by Leading chat GPT on that's interesting on a path you can do some
705. 1:53:25stuff with known information okay Winston yeah so what Winston is talking about here is really true one of the things that I'm increasingly
706. 1:53:32increasingly impressed with with these tools is if you give it something it's really good at analyzing it or learning
707. 1:53:39it like give Claude like documentation for some piece of software you wanna or like some programming language you want to
708. 1:53:50start programming in it'll learn the whole [ __ ] thing and then just start programming for you um it'll analyze the screenplay and tell
709. 1:53:58you what sucks about it it'll take whatever math stuff Winston put in there and combine it together and give you some new math
710. 1:54:06so yes it can be creative how's it creative you stick a creative human being next to it who who is conceptually
711. 1:54:14figured out oh these things can ingest anything and they know everything and if I'm clever about how I combine that data
712. 1:54:22and ask the question I can get some remarkable [ __ ] that's the art that's really cool Arcanum are there any open source ones
713. 1:54:35you can run locally lots of them um I don't know them uh check out a project called GPT for all and it allows you to install it on Mac
714. 1:54:46windows and Linux and then you can pop in whatever open source large language model you want it gives you a basically
715. 1:54:54a running environment where you can run large language models inside it and uh that's pretty good but I I'm not
716. 1:55:01following the open source world because it's moving too fast and it requires too much of my brain to make work so I just
717. 1:55:07I go like I think open source is really good but you got to be really geeky and the interface is always really bad and
718. 1:55:14you got to be good at command line and then you got to run terminal and then you got to understand what Jupiter
719. 1:55:19notebooks are and then you got to run Google collab and then you got to probably get good at Python and then
720. 1:55:23when it breaks you got to realize oh you don't have the library installed and then you don't have the library installed you're like [ __ ]
721. 1:55:31that's my experience with open source didn't show up did he call no the guy that didn't show up today didn't call
722. 1:55:42didn't respond to the email from the woman that was hosting the meeting that invited us on his behalf um didn't email that response so
723. 1:55:57um I don't I don't know about any of you like I got [ __ ] anxiety when it comes to like I show up early for meetings I I like
724. 1:56:08I have I have whiffed on two meetings in my 11-year career at storyline and I can remember both of them like the [ __ ]
725. 1:56:18like I apologized 40 [ __ ] times so I don't get it but whatever why would they need to discover cancer
726. 1:56:27cure when we have found many of them but doctors still ignore them doctors don't ignore them the giant corporations do not fund them
727. 1:56:39you can't make money on a cure it's America it's about capitalism it's about dollars we like dollars you gotta you it's about you've got
728. 1:56:52shareholder value I don't think you understand this you've got shareholder value if you cure cancer you can't provide shareholder value if
729. 1:57:03they're not needing drugs year after year I don't silly silly why would you cure cancer instead dollars
730. 1:57:17and the ordinary people's run wait the ordinary people's Run for the AI is actually the AI learning about the mass people of the world
731. 1:57:27uh uh okay okay yeah yeah okay what you said that's right sure all right do you head back to Denver in the morning new
732. 1:57:37we have another meeting tomorrow that I'm fairly confident that person's gonna show up same host person um
733. 1:57:48and that's a long meeting and that one I'm really excited about and then we're having dinner with both of them
734. 1:57:58at new one Joe in Koreatown I'm very excited about and then we go to Betty buy and we get up bright and early we leave at like six
735. 1:58:08in the morning on Wednesday um he wants to know what God we wish for I don't know who he is but yes everyone wants to know what God we wish
736. 1:58:18for except government will step in the worst case they won't let China get too far a jump on us
737. 1:58:34um [Music] I don't know possibly that's way above my pay grade I don't understand geopolitical currents very well so I'm not going to
738. 1:58:47comment on that in 1998 I rooted Korea Telecom and flooded Google Yahoo and lycos offline using a Venom script oh cool
739. 1:59:01in uh in 1980 when would this have been like 80 80 or 81 my buddy Scott and I was in we were in high school
740. 1:59:16and he got a modem and you could get lists of computers you could call up that that didn't have any password protection
741. 1:59:26on them because that's how it was back then because not many people had mowed them so you know the hackers would would you
742. 1:59:33know they would write scripts to go find all the computers that didn't have any controls on them and so I remember we we
743. 1:59:40got this list of of all these phone numbers and we found a phone number of a the control system of a mall in North
744. 1:59:48Carolina and so we log into this thing and we turned the heat up to 99 and we reversed the escalators and we were like we're hackers
745. 1:59:57[Laughter] oh man this is some good old days that's awesome uh 99 Plus messages by the way if you're new here I read all the [ __ ]
746. 2:00:12questions I get very far behind I've been doing this for like two hours now my voice is shot I'm exhausted as [ __ ]
747. 2:00:18I'm Punchy as [ __ ] and I am behind on your questions so I'm probably not going to get to your questions but you can
748. 2:00:24keep asking them because I do try to get to all of them um but anyway you should follow me and then that way when I do these lives I
749. 2:00:30try to do these every night um you'll get notified and then you know yeah whatever that was Google NASA was so mad for taking a huge
750. 2:00:42chunk of the world's neural network no that just seems exhausting me to the bottom oh who can tick tock why don't you do that to me amazing
751. 2:00:57cranky I feel like I'm talking like muscles now that's all yeah we're Doug and Wendy Winer we've got diverticulitis yeah
752. 2:01:13gen xers will get that the rest of you like who's Doug and Wendy weiner what's diverticulitis you're stupid old man
753. 2:01:21why are you old are you old were you on Tick Tock if you're old you big fatty you big fat Rosie O'Donnell looking
754. 2:01:31fat Alec Baldwin looking [ __ ] why why are you old you shouldn't be old shouldn't old people just stop [Laughter]
755. 2:01:42wait be careful the website looks like a joke but it's not which one which one was the one that looked like a joke but it's not
756. 2:01:51hello Uncle Kyle hello Mr drecker or Mr Decker Troy Decker Troy Decker good to meet you good day good day good day to
757. 2:02:01you sir good day brother right there rather to you to you too and do you choose that oh well um fancy a game snooker yeah
758. 2:02:17I really think big chat is terrible compared to GPT 3.5 fuzz I couldn't agree more Microsoft lobotomized Sydney which was
759. 2:02:27the project name for gpt4 inside Bing after the New York Times revealed that Sydney was batshit crazy so Microsoft
760. 2:02:37made Bing chat a little bit of a dum-dum and they fixed it yet so yes I would agree with you but I do like being being connected to
761. 2:02:47the internet is convenient and I do use it sometimes and it makes images which sometimes if I need something in a pinch
762. 2:02:53I'll just knock it out my CIS ADHD gave me her cover letter to edit that was 3 000 words she was on a roll I guess wow yeah sorry about that
763. 2:03:04that's that here that's one of the downsides of these things is I've done this before where I'll start writing
764. 2:03:10something and I'm like oh that's a good paragraph and I'll dump that into a word doc and then I'm like wait rewrite that
765. 2:03:15paragraph and then I'll dump the the new improved version of it over there and before I know it I've got a document
766. 2:03:21that's got like 17 versions of the same paragraph and I'm trying to figure out which one's best and it ends up making a
767. 2:03:27nightmare of Hell of yeah editing so gotta be careful with this long time to wait for my jokes to hit in real time
768. 2:03:35thanks Kyle and uh sorry about that AI learn and get used to it tick tock is about 10 cent app Chinese don't give in a [ __ ]
769. 2:03:47about IP or copyright yeah we knew that all right I don't I yes I don't what what am I supposed to do
770. 2:04:03about that I can't do anything about that except make money controversial thought the internet is already alive and AI is reshaping the
771. 2:04:13world bit by bit I don't know if that's that controversial I think the internet is alive I mean the internet's not alive the
772. 2:04:22internet's a protocol but it's at the center of all of our lives or a lot of our lives and then this stuff's going to amplify it
773. 2:04:30dramatically so yeah so I think this I think what this does is it accelerates all of that [ __ ] and it it hides all the complexity
774. 2:04:42and reveals all of the all of what's underneath the complexity um and it does it really fast so um it used to take half an hour to
775. 2:04:53download one song you kids are lucky exactly Ace dogs got it Ace dog 90. Ace dog 90 knows what's going down
776. 2:05:03I remember when we were using dial-up modems to call into Mall Control Systems you could watch the the [ __ ] type across
777. 2:05:13your screen which is funny when you look at gpt4 right now it's not quite that slow but like if the server is busy you
778. 2:05:19watch it typing across your screen we're gonna look back at that and go remember when chat GPT you know wasn't
779. 2:05:26just a million words in a third of a second remember when you actually had to wait for it to write a paragraph man that was torture
780. 2:05:38that's coming um oh the the the Brontosaurus is the AI large language model mascot is it that's cool those are not tiny dinos they're
781. 2:05:51tortures or the tortures why do I not know this Amelia's lifetime check 10 26 thank you oh it's now it's 11 11. oh that's why
782. 2:06:01you put that in there so I know how far I'm behind I am um 40 minutes wait more than 40 minutes 45 minutes behind
783. 2:06:11all right I think I'm gonna scroll to the Bottom now I know Russian didn't see it come into my feed that was earlier is
784. 2:06:17that a new shirt Mr Kyle it's yeah I just I haven't worn it because it's got a lot of [ __ ] on it but it was the
785. 2:06:23only thing I had in my bag I think Gary Gensler was a decent guy when he was teaching me about cryptocurrencies at
786. 2:06:30MIT he oh I used to think Gary Gensler was a decent guy yeah I mean I'm sure he's a smart guy but you know he's trying to protect the
787. 2:06:40old yield Treasury and you know not compete with um technology he can't compete with so I don't know but it pisses me off
788. 2:06:51because I didn't get my free goddamn crypto why don't you get 5G internet I have 5G internet because I don't know because
789. 2:06:59I'm a little lazy it doesn't I I just it's I can't be bothered you have no proof that I have your MacBook all right I'm scrolling to the
790. 2:07:10bottom I'm jumping a bunch of questions I know I just jumped a bunch of people's questions I'm going all the way to the
791. 2:07:15bottom holy crap was I behind okay now I'm at the bottom and I'm going to scroll up just a little bit and then I'm
792. 2:07:20gonna do a handful of more questions and I gotta go to bed look at my eyes they're all puffy and tired they're tired I'm like [Music]
793. 2:07:35I need water the Moon is a satellite made out of cheese for upper Dimensions the Earth is a toroid and also flat appearing spherical
794. 2:07:48from here with a donut in the middle a hole a hollow Earth if you will that's where all the aliens came from
795. 2:07:57except for the ones that came from interdimensional space but the other ones came from the inside of the donut or the what we would call
796. 2:08:04The Hollow Earth if the Earth was fear but we no it's not it's flat and so it is a toroid and it is a donut shaped a donut Guyana of the
797. 2:08:14name would taste is sweet Cinnabon would be sweeter I think you ATR too many dollars
798. 2:08:30huh I think you ATR to many dollars anyone know what beer maker is talking about I
799. 2:08:45don't know what you mean beer maker that does not seem like words anyone who just signed in welcome to the man behind the curtain
800. 2:08:55exactly NASA is Disney Disney is NASA and Coca-Cola you think it's an enjoyable soft drink oh no oh no hmm
801. 2:09:10all right moving on well I'm an ex-evil hacker I own NASA servers they are all the world's leaders in AI oh they are
802. 2:09:18the world's leaders in AI all right that's cool listen I maybe I like I said I'm I'm not so know so much about the
803. 2:09:28quantum computers other than they got qubits and I used to play Qbert anybody used to play Qbert kubert was run on a super computer one day
804. 2:09:39and you put a quarter in it and then you jumped no anyone I like your bald head thank you it's lovely so I put this part of the toupee
805. 2:09:53on first and then I lift it up and I have another toupee that I put on the sides and this I try I got a toupee that
806. 2:10:00I tried to make it look like I have wisdom this is what they would call a wisdom patch this is a wisdom to pay
807. 2:10:07and then I put on well this is three components of a toupee I put them on in random order and I tried to make it so
808. 2:10:14the hairs don't align so they all go in different ways and then I lift that up and I put on this thing and I'm
809. 2:10:20completely bald in the back of my head it's a look it's a look I like I've I've developed it over years so thank you I I
810. 2:10:29take great pride in my artificial encumbrance on what is considered by some ahead yes all right I don't understand code
811. 2:10:41interpreter I know if if I could show it to you if I could but I don't think I can let's see if we can if I can get to chat gpk
812. 2:10:58no I can't it's not working I'm trying to think of the best way to get your head around code interpreter um
813. 2:11:12actually here's here's is that a way hmm go to so if you're trying to understand code interpreter if you have access to it
814. 2:11:28there's a site called kaggle k-a-g-g-l-e and at kaggle you can get data sets I don't know [ __ ] about data analysis but
815. 2:11:37you can go there and you can get like Spotify top 200 song data or Healthcare in Cincinnati or crimes in your city or whatever there's just
816. 2:11:49it's just a site full of data sets so go there and just sort of browse the data sets and find something that's personally interesting to you
817. 2:11:59and then you download the what it'll download as a zip file and like you don't know what's in the zip file you could read
818. 2:12:07you could read on kaggle what's in it but I don't read I don't read the manuals that's a Gen X thing and an add
819. 2:12:12thing and I just I'm a rebel man I don't [ __ ] read manuals [ __ ] you with your manuals my son reads manuals I yell at
820. 2:12:21him all the time I'm like no we don't read manuals in this family have I not taught you better and obviously I have failed as a father
821. 2:12:31anyway so don't read the manuals just download the archive you got a zip file now so you go to code interpreter
822. 2:12:37and upload that zip file and then just say to it please examine the data inside this file um you know understand the nature of the
823. 2:12:51data analyze it for anything that you think is interesting and tell me what you discover and then just just watch it
824. 2:12:59and it will just start prompting itself it will give itself a list of instructions and it will reveal something about that
825. 2:13:07data and it might at some point ask you questions what would you like to do with this data and then you can say things
826. 2:13:12like I don't know I want you to find what you think is interesting in it and tell me and just let it go and then when it
827. 2:13:19makes those little boxes it says working you can open those and you can like watch the code that it's writing it's
828. 2:13:24actually writing and executing python code so it's remarkable so just go get a data set and just ask it to do [ __ ] and then
829. 2:13:33the once you kind of understand what's in the data then you can say things like now I want you to visualize the most
830. 2:13:39interesting Trends in the data and it will write you charts and graphs and and here's the deal code interpreter if if
831. 2:13:47you're in business and you want to understand some data you don't need a data analyst now you can just go ask it to do that [ __ ] and
832. 2:13:58but but the way I think about code interpreter is this right now it only does Python and it only does like a handful of file types and right
833. 2:14:08but it's kind of taking on its own life like you give it kind of a goal and it's like well let me go open up this ZIP
834. 2:14:15file and it writes code to do that and it opens it up and says well it appears there's two csvs in here and a Json file
835. 2:14:21would you like me to examine those sure and it goes and examine those look what I found in this CSV look what I found in
836. 2:14:28this one it appears that these two are about crime statistics and the Json is labels about but and let's just go do
837. 2:14:35that so if I just sort of play that out three years in the future now it can do all programming languages it can take in
838. 2:14:43all sorts of files and it can accomplish anything and it'll be connected to the internet and it can write its own code
839. 2:14:49and it'll know every API out there and it'll be able to get any data source at once so you won't have to go to kaggle
840. 2:14:55and find a stupid [ __ ] CSV file it'll just go find all the data so so think of code interpreter right now is just kind
841. 2:15:04of a Looking Glass into what's coming that that's kind of how I'm about it right now but there I I guarantee there
842. 2:15:12are people using it in incredibly powerful ways I'm just not one of them I have a Blog can I use chat gbt to
843. 2:15:20write blogs oh [ __ ] yes it's really good at that go to uh chat GPT pay for pay the 20 bucks for gpt4 because it's just so much better and
844. 2:15:31here's one thing you can do especially if you have a particular writing style take one of your favorite blogs that you
845. 2:15:38think is structured well and written well and say based on this blog paste write me 10 blogs on these 10 no I have a Blog about X
846. 2:15:51here's an example of my word paste give me 10 blog ideas um that are non-obvious and I may not have thought of and it'll give you the 10
847. 2:16:05ideas and then you can just say okay I like one three and seven write me a blog post copying the tone and structure of the
848. 2:16:14example I pasted I showed above for one three and seven and it'll write you those three blogs and they will probably be at 85 or 90 percent
849. 2:16:25and then you'll have your Kevin McAllister moment [Music] I had no idea but it can't do it again can it it can do it over and over and over
850. 2:16:42and if you get clever at giving it good prompts it will get really really good at [Music] getting better and better
851. 2:16:52first you take BART across the river on your boat and then you leave him on the other side then you go get pie
852. 2:16:59it's a journey we've all been on Dr J of course we started with Bard and of course we took it across the river
853. 2:17:07what else are we going to use but a boat you need a boat what are we going to swim with bud no put Bart in a boat
854. 2:17:15you put part in the boat you cross the river then of course you leave them on the other side he's Google after all silly Google
855. 2:17:25you invented this technology and then screwed it up somehow you stay over there I'm going to be playing with pie
856. 2:17:33we come back and we talk to pie oh the conversations we had she talked all night yeah I agree solid solid I like it
857. 2:17:42Alicia I love you thank you appreciate the kind thoughts same Kyle basically I don't track time well I'll show up early always
858. 2:17:52all right let's see I'm going to do a couple more questions I'm leaving can llms generate their own training data
859. 2:17:59yes and no sort of maybe we don't know it's very possible so llms can create synthetic data and they can do it really really well like I had
860. 2:18:12it generate a bunch of personas for me that were [ __ ] brilliant and so you could theoretically take those personas
861. 2:18:18and you could say hey let's train an llm on all these personas that this thing generated some people are saying that's good and
862. 2:18:26it accelerates these things dramatically and they're going to get stronger and stronger and stronger and then there's
863. 2:18:30other researches that say when you do that you're gonna have this thing called Model collapse and it's what it sounds like it's good
864. 2:18:38it's good it's good and then it's [ __ ] and they're thinking that synthetic data training these things on synthetic data
865. 2:18:46by these things is going to lead to model collapse and they're all going to suck I don't know I don't know I don't know
866. 2:18:55the math I don't know how this [ __ ] works I just don't know why am I hitting myself in the head with a remote
867. 2:19:04because someone's got to change the channel okay it's very simple the Earth is a donut and you change
868. 2:19:12channels like this okay everybody knows that if you think that's a conspiracy theory then you're talking to the wrong guy okay
869. 2:19:22everybody knows that bail Count Me In all right am I done I think I'm done I think I got to the bottom oh wait here's
870. 2:19:33Winston Kyle you make me enjoy The Tick Tock so much I'll start talking in your funny voices too oh God it's bad it's bad it's not
871. 2:19:48quite Adam Sandler bad when he was doing all his bad voices but it's it's it's rough it's but thank you I appreciate
872. 2:19:55that what do I think will happen when Google crashes um hmm well I think that Google has a choice to make right
873. 2:20:12the orb um Google needs to decide are they going to be Kodak right Kodak invented the digital camera
874. 2:20:26and Senior Management said but that's not our business model get the [ __ ] out of my office Google invented the Transformer
875. 2:20:37which all of these things are using and then they realized oh that's an existential threat to our business model
876. 2:20:45get it the [ __ ] out of my office Unfortunately they released the paper so all these data scientists went off
877. 2:20:54and created [ __ ] like open AI and chat GPT and Claude and pie and perplexity and llama
878. 2:21:11so now they're playing catch-up they have tons of money um I think their business model has to change I do like it just I think it has
879. 2:21:21to change Microsoft has all of their products to fall back on like Microsoft can can make Bing chat free and they can they can lose money on
880. 2:21:32it because they can sell Office 365 and they can sell windows and they can sell Microsoft azure what's Google got they've got the ad
881. 2:21:41business they had a monopoly for 12 years something like that 92 93 [ __ ] market share for a decade and then overnight this thing comes
882. 2:21:55along that just [ __ ] obliterates the need to go to websites you can just get the answers oh [ __ ] what do we do
883. 2:22:07so I don't know they're gonna have to reinvent themselves I don't know I don't know I don't know they got smart people
884. 2:22:15and they got a lot of money but they got a lot of hubris and their engineer heavy and artist light you know you look at Apple
885. 2:22:29[ __ ] best engineers in the world and the best creative minds in the world working together in sweet little Harmony like Kumbaya
886. 2:22:40here's the church here's the Steep will open the doors where are the people they're the people that's Apple that's Google so
887. 2:22:52I don't know so I think they got some structural challenges because I think that balance is necessary these are
888. 2:22:58large language models they're not large data models so the humanities the Revenge of the liberal arts major
889. 2:23:07the Revenge of Gen X it's all coming to pass it's all coming to pass those that think with a horizontal mind in a broad world view
890. 2:23:18will be rewarded in this new world those with myopic Focus will be penalized so I don't know I don't know
891. 2:23:34I think it's got to be rough there right now like like it's got to be like It's gotta be rough there 'll get it like apparently Gemini the
892. 2:23:47the thing that's coming after bard is is supposedly quite impressive but it but you know you don't think GPT 5 is
893. 2:23:55going to be [ __ ] impressive you know you don't you don't think what [ __ ] Pie's gonna do with 1.3 billion
894. 2:24:03dollars is impressive these are the guys that invented anthropic those are the guys that invented gpt3 they're now competing with open AI so
895. 2:24:14this does this list doesn't stay static so Google's gonna come out with Gemini and do they come out with that before GPT 5 or after
896. 2:24:25right does Gemini jump them ahead of gpt4 or are they constantly playing catch-up because they're five years behind I don't know
897. 2:24:36and then do they have the do they have the cajones to throw their own business model out if they need to
898. 2:24:43I don't know like what is the business model for all these things nobody [ __ ] knows that right now nobody
899. 2:24:52no matter what you build in software right now just using these Technologies today you can replicate anyone's business anyone's
900. 2:25:02technology probably over a weekend and it's gonna get faster you're gonna be able to look at a competitor and go um AI could you please go
901. 2:25:16grab all the marketing positioning identify the target audience identify the tech stack and replicate that and
902. 2:25:24then point it toward this target market and have that thing launched by um four how do we compete in that world I don't
903. 2:25:34know do you because I don't I don't know why I'm laughing uh all right two more questions
904. 2:25:48I can't register for the salon tomorrow what's wait what's going on why can't you register for the salon is is Meetup broken
905. 2:25:58because I was trying to send what's the best way to assist it three Innovations should be released should it no I don't think so RJ
906. 2:26:14I can't register for the salon tomorrow Link's not clicking through properly hang on uh Salon dot AI
907. 2:26:31oh come on I don't have no internet hang on let me there's one way to once again the cat here we'll flip over to connect
908. 2:26:39to the internet what what the [ __ ] is going on here the connection was interrupted what is happening
909. 2:26:49oh here we go wait something's happening oh [Music]
910. 2:27:07um [Music] it's driving me crazy people why is the internet's not working let me
911. 2:27:21try another browser Chrome enhanced privacy and Chrome all right it jumped back to that my machine is not happy
912. 2:27:36[Music] let's see [Music]
913. 2:27:59[Music] I think Chrome's working it looks like Edge just edged not so happy with this hotel internet I think we're gonna have to
914. 2:28:09we're gonna have to explore some other opportunities we got some opportunities to improve the experience bi-weekly Meetup join this group
915. 2:28:26is this what you're getting to so the if you go to the group and then you say join the group and then the featured event is
916. 2:28:41this one here oh wait but here's what I did I went to the salon.ai which is a link tree and it's the second link right here is
917. 2:28:51our Meetup is it the Discord that's broken the Discord community this might be broken now that seems to be working maybe not
918. 2:29:08but anyway if if it's if the discord's broken that I can go that I can go fix but the Meetup seems to be working the Meetup seems to be
919. 2:29:17go to a salon and go to the second one bi-weekly meet up and that's our Meetup and then you can do that there
920. 2:29:26you can go there and that should work oh the link Tree Salon it's yeah it's the the it's link tree l-i-n-k-t-r dot e e well no just go to
921. 2:29:43the salon.ai that's a website the salon.ai wait hang on hang on people we're gonna wait wait that was not the right thing this is the
922. 2:29:56right thing go to that website address the salon.ai go there that's a link tree that looks like this see my stupid little cyborg dude and
923. 2:30:13then the bi-weekly Meetup the second one down that's the one to sign up for the salon tomorrow see and we got pretty pictures see we
924. 2:30:22got pretty pictures this was generated mid-journey in AI salon that was by Ali Ali did that one see we got 614 members that's a lot of
925. 2:30:35members not as many as here on Tick Tock we got thousands we got thousands here on Tick Tock all right listen
926. 2:30:47I'm tired it's way past my bedtime I got important meetings tomorrow but they don't start till later in the day so I can sleep in
927. 2:30:54I can sleep in sleep sleep in all right um got it ship hopefully that got you in there oh got it thank you okay good all right
928. 2:31:03people are in oh it looks like the AI learning lab of Kyle Shannon on tonight's show on tonight's show we have um
929. 2:31:12welcome Clint uh by the way I won't be I'm not co-hosting tomorrow because I have a business dinner here in New York
930. 2:31:22when the salon starts but Leah and one of our other organizers are going to be hosting and John Knack from Adobe is going to be speaking
931. 2:31:30he's worth listening to because he's funny and energetic and smart as [ __ ] and he's got you think I have ADD John
932. 2:31:40Knox like [ __ ] world class man um but he's he's senior in adobe's AI group so he's he's one of the dudes
933. 2:31:48behind Firefly and generative Phil and all that sort of [ __ ] so he's the man so he's this is the second time speaking at
934. 2:31:54the salon so it's worth going there for that I'll be there tomorrow I just don't know when I'll join um all right everybody hey peace out it
935. 2:32:03was good joining you from New York City I'll see you tomorrow night from New York City after the salon I'll come do a
936. 2:32:08little bit of a little bit of gaggling gaggling around but uh yeah all right peace everybody Fannie Willis is talking oh she is all
937. 2:32:21right I gotta go watch Fanny Willis peace out everybody there's some [ __ ] going on in Georgia The Devil Went Down
938. 2:32:26to Georgia he was looking for a soul to steal he was in a bag because he was way behind and he was looking wait man well
939. 2:32:33anyway that song all right peace yeah that's what I'm talking about bye everybody