Sep 12, 2924 - Oprah Got AI WRONG! (My Trip to Lobby Congress)

Transcript

1. 0:03champ you want to come sing you want to come sing buddy [Music]
2. 0:43you're loudy you're loudy that dog is loudy all right here's the deal tell all your friends to come it says there's one
3. 0:53person in here but I see more of you in here you're tapping away all right [Music] so it is possible we can we can stream
4. 1:03the Oprah thing but uh apparently ABC's got some weird [ __ ] up thing I thought we could watch it on Hulu and it said
5. 1:11streaming on Hulu tomorrow um so I just signed up for a trial for some [ __ ] TV network that's ridiculously expensive
6. 1:20so hopefully this works I'm on ABC right now I'm apparently on the channel that it's going to be on but it's possible it
7. 1:28switches to my local news in instead of the Oprah special so this could be a bust welcome welcome to janky [Laughter]
8. 1:42TV all right I'm back from Washington my back is still [ __ ] up it's hot I'm cranky I'm stressed out Kyle sign up on YouTube TV free trial
9. 1:55okay yeah maybe I should do that too um hang on yeah it's a good idea
10. 2:10let's see subscriptions let's see YouTube
11. 2:26TV sign up and try YouTube TV 10 days for free now you guys got to help me unsubscribe to all this
12. 2:40[Laughter] [ __ ] start free trial base plan probably in the base plan
13. 2:57no more info of course they're not alphabetical of course they're not
14. 3:25alphabetical all plans include local channels $65 a month what happened to everything streaming being
15. 3:39cheaper unbelievable next check out
16. 4:01did you watch the debate okay Taylor Swift did kamla hara's begging for money let's
17. 4:15see can we search let me go back to fubo TV
18. 4:33choose your favorites and we'll Rec oh skip dismiss all right ABC World News Tonight okay so there's
19. 4:44ABC Denver 7 popular now okay I think we have it I still have your card number you can skip this here we go what time is it 58
20. 5:00good Lord it doesn't need to be this stressful people look my hair's all tall what's happening what's happening all
21. 5:16right let's go over to fubo TV that says it's got three minutes left hair lock on the spacecraft so all
22. 5:29four members of crew were exposed to the vacuum of space while that hatch was open closed vertical translation the
23. 5:37crew is expected to come home early Sunday splashing down off the coast of Florida and tonight they say they are
24. 5:43now one step closer to eventual bases on the moon hey everybody what's happening an Murphy history made and what a
25. 5:51beautiful view of Earth thank you for watching here on a Thursday night I'm David M we'll see you right back here
26. 5:55tomorrow from all of us here good night thank you for World News Tonight with David mure America's most watched
27. 6:06newscast op right now at 6 a brush fire in Northern Colorado shuts down a busy Interstate for hours dry air and Gusty
28. 6:14winds fuel fire danger on this weather action day hot temperatures and Gusty winds a fire weather warning continues
29. 6:21until 8:00 p.m. plus disturbing testimony in the boulder King sup trial I stood there all right what's that let's
30. 6:31see live now on Fox wait what do I do Brandon on YouTube t TV click live and find your local ABC affiliate
31. 6:43live local ABC affiliate see this says Denver 7 News this Mike Nelson joining us with the first of the forecast we hit 93
32. 6:55today that's the 61st day of 90 or hotter so far this summer and way above the average
33. 7:11so I thought it was on 8:00 p.m. change your time zone oh my God how do I change my time [Laughter]
34. 7:27zone change your location Kyle great how do I do that let's see manage your account location Denver area current playback area let's see
35. 7:44update update home area allow us to access your location so we can serve the right verify your current area
36. 8:02started I know is there any truth to the Aurora thing I don't [ __ ] know no Central Time how do I change it
37. 8:17area Let's see we go here CC oh this is going to be a [ __ ] wamp
38. 8:32wamp lie Kyle lie uh I don't know if I can it's saying let's see update current playback area update location see it's just getting it from
39. 8:46my GPS go back and click on AI future of Everything YouTube TV no
40. 9:04thanks can I search search AI future got
41. 9:22it here's today at 7 so this an hour from now looks like it's going to be in an hour go back and click on AI future of
42. 9:35everything I just had an interview today I'm scared of interactions like that okay I got it now on fubo okay someone got it on fubo
43. 9:46Jesus here we go play channel it's not it's not playing it damn it it's on now I believe oh I believe
44. 10:01that too then you must be in EST maybe you're delayed in time zones in Denver Live on YouTube TV where
45. 10:17where um oh ra AI future of us says airs today at
46. 10:337 this is why I can't live in Denver anymore this is unbelievable the hell happened to TV where can I watch Opa Oprah live well
47. 10:50apparently you're going to watch it here are you going to stream it I can't get it to [ __ ] work invite inv Brandon to share his
48. 11:00screen and join your live all right we can do that or we can watch it in an hour we I finally found it on my local channel oh
49. 11:11my God this is so frustrating well I guess we'll do it in an hour it's on cable TV no it's well it's on yeah it's on cable TV if you
50. 11:24have networks and you're not in Denver apparently yeah I know it's on now Source Camp I'm very well aware of that this is
51. 11:37a yeah Hulu's streaming it tomorrow that's what I found out at half an hour ago it's on ABC I know that but you
52. 11:44can't watch ABC anywhere that's my problem um the summon overs we'll be here in an hour all right I don't have enough
53. 11:55subscriptions my TV provider is blocking it due to Disney this is is unbelievable this whole [ __ ] fragmentation of of
54. 12:02the media is just it's going to be convenient you can watch whatever you want unless you want to watch something oh my God all right
55. 12:21well all right I'm in California and we'll be sure we be watching it from you my TV provider is blocking it due to Disney
56. 12:31yeah I just subscribed to $150 worth of subscriptions so I could watch it but they're not letting me watch it for an
57. 12:38hour so we can watch it we'll watch it at the top of the hour um yeah there you go all right fantastic
58. 12:48um so I'm trying to think what I want to do oh this sucks I I'm I I am so tired of capitalism I really am deregulate it it'll be great for
59. 13:04everyone uh-huh search a BC uh ABC Oprah play channel
60. 13:20news I got to see this [Laughter] numbnuts she's in interviewing Sam right now thanks Dr
61. 13:35J uh tell us about Washington all right I guess I guess we'll call that a punt um we'll get the dog singing all right
62. 13:44so let me tell you about Washington so I went to Washington Tuesday night had a nice dinner at a Georgian
63. 13:53restaurant like Georgia the country I don't know how to describe the food it's kind of like you know Mediterranean
64. 14:03food where you got all the fresh veggies and olives and hummus and [ __ ] remove the olives and hummus add in more pickled
65. 14:15cabbage that's pretty much it a lot of vegetables um and then so the
66. 14:29so what we did was on Wednesday there were 12 of us there 12 AI entrepreneurs and they basically had schedules I got paired up with a guy
67. 14:42named George who was awesome he's lives in St Louis and he's got a company that does AI robotics lawnmowers for like big grassy
68. 14:53areas like airports and you know Municipal parks and [ __ ] like that where you've got many many acres that need to
69. 15:01be mowed and basically no one no one wants to do the jobs he found out that there's like basically a labor
70. 15:08shortage um of people that don't just want to mow the lawn for eight hours you know in the blasting Sun uh who who knew and
71. 15:20uh so that's his business he's cool he had a PhD in robotics and mechanical engineering he's like three years into that business
72. 15:31sounds like it's going pretty good um and so we went to the first the first office we went to was Senator Bud out of North
73. 15:44Carolina and we met with with with all of these we met with a staffer so far all political okay interesting yeah actually you know what
74. 15:57don't tell me what's going on in the Oprah thing I'll just watch it and pretend like pretend like it's [Laughter]
75. 16:07live um but what we could do so um open AI did launch today strawberry um it's called 01 so you have
76. 16:22GPT 4.0 gp40 and this no one new one is called 01 I I don't know what they're [ __ ] doing anyway back to Washington um so we
77. 16:35can play with I played with it a little bit I got it about an hour before I left work um and played with it a little bit
78. 16:42and it's interesting what it does um so I have some thoughts already um okay so with all of these people we met with staffers and meeting
79. 16:54with staffers is really interesting because sometimes you get staffers that are educated about your topic so so the the
80. 17:02the group that helped organize this they set up all these meetings this is what they do um so they know everyone and they try
81. 17:11to get sort of the highest most relevant staffer you know for for these meetings um and the first one we went to was was uh a North
82. 17:22Carolina thing um it was George and I and Kathy from the rxn group and then this woman I forget her name who was
83. 17:32from North Carolina who was also had an AI startup and so we just introduced ourselves so the constituent the one
84. 17:38from North Carolina went first and she said you know here's my here's my thing and then George talked about his thing
85. 17:44and then I talked about my thing and we all kind of had our message um for for the uh you know for the staffer and the
86. 17:51basic idea is you're just trying to educate them on what message you want to get out so the message the message that I wanted to get out was
87. 18:01um there's a couple couple of major ones one is a lot of the legislation right now is legislating the
88. 18:08technology as opposed to the use of the technology and so Kathy actually had a really good analogy to this so if you talk about cloud
89. 18:19computing we don't regulate cloud computing but if you're storing Health Care information in the cloud that's
90. 18:27covered under HPP regulations if you're doing Financial Services um data that's covered under Financial Services regulation but you're
91. 18:37not actually regulating the tech what's happening with AI is they're regulating the tech and it's sort of sweeping all the all
92. 18:46the use cases up into it that might not be relevant for that piece of Regulation that's kind of what happened in Colorado so part of it was
93. 18:55just be mindful of how your writing the regulations because it's sweeping up small businesses and it's sweeping up
94. 19:03use cases that might might not be relevant so the protection is good but the execution is really bad um I also
95. 19:14wanted to talk about the difference between generative Ai and AI as we've known it before right so if you think
96. 19:20about ai's been around for decades and a lot of the perception is just AI is just this one thing and what I talked about was you know
97. 19:31this community so I was there we were we were the the salon was the representative organization that was
98. 19:39going around to these groups and so I was talking about the salon and all of you and saying you know this group was
99. 19:45started the week after chat GPT came out and I get to witness daily people Reinventing themselves and Reinventing
100. 19:53their businesses and I mentioned you know Jim Ross he said on a on a call the other day he feels like he's functioning like a
101. 20:0110-person company and he's a he's a solopreneur um that statement got people to sit up they were like oh that's really interesting
102. 20:11um the the staffer from Bud's office was he was smart he was taking a lot of notes he was asking questions he was
103. 20:19basically saying things like we hear you this is really important issue um he kind of said there's not going to be
104. 20:27anything coming from the Fed um for probably 2 or 3 years and um and I to to which I said so are we going to
105. 20:37have a patchwork quilt of nightmarish state-by-state laws and he goes yeah probably I said well that sucks so could you pass that
106. 20:50along um oh this was really fascinating so I um on the flight out there I was sitting next to a woman um and she was the executive director
107. 21:06for a um a what's it called Foundation that supports civil engineers
108. 21:21and I said are you using chat gbt you know I did I did my obnoxious AI evangelism [ __ ] [ __ ] that I do she said she had tried it but not
109. 21:33really and she she said well I do a lot of like writing and press releases and things like that and I don't want um
110. 21:39it's got to sound like me and so I said well here's some ways you can make it sound like you and I told her about take
111. 21:45some of your writing and put it in a document and upload it to chat GPT and it'll start writing like you and I was
112. 21:51given her ideas and she was taking notes and then we started I talking oh I said you know what are you going to
113. 21:58Washington for and she was doing some lobbying and I said oh I'm doing some lobbying and and uh and I said I actually testified
114. 22:06against the Colorado Bill and she goes oh that bill yeah she goes I didn't know where you stood on that I didn't want to
115. 22:12say anything but she says because so to give you some context of the Colorado Bill the Colorado bill passed and was
116. 22:22signed into law by governor polus with a signing statement that effectively said this law is a piece of
117. 22:31crap um I have grave concerns that it you know sweeps up small businesses so please fix it before it goes into effect in 20 late 2025 early
118. 22:422026 and um she knew the law and she said to me I've made a strategic decision not to use chat GPT
119. 22:58in any way because or any generative AI in any way because the way that law is written I don't want to put in a year or two of
120. 23:09learning how to do something for my business changing my processes and then ending up with a regulatory burden
121. 23:16sitting on top of that two years from now so I had a a an an actual specific example of the chilling effect of that
122. 23:26kind of legislation that this is you know this is someone who's got a big Statewide foundation for all the civil
123. 23:34engineers in the state and is actively not using chat GPT because of that law or or or you know not not using AI
124. 23:43because of that law um welcome to the AI learning lab we'll be watching the Oprah interview live with Kyle at 7 because I
125. 23:51live in [ __ ] Denver and all of the channels have been deregulated and decoupled and Disney is an ass ho there you go there you have it
126. 24:01[Laughter] um what if I want to watch the the show oh sorry you can't but you can you can stream it live tomorrow on Hulu no I
127. 24:13want to watch it live now oh yeah oh hey yeah we we uh we have a response to that maybe you could go on ahead and go [ __ ]
128. 24:21yourself does that work for you so yeah so we'll be watching we'll be watching Opie you can have ai and you can have ai and
129. 24:31you can have ai we'll be doing that in 45 minutes 42 minutes okay um so that was the bud the bud meeting um the next meeting I had with
130. 24:46was with Josh Holly of Missouri um George who I was with is from Missouri so we had a different Handler for the Josh Holly thing um I was
131. 24:59I'm not a fan of his but we didn't meet with any of the Senators so it doesn't matter we're just meeting with staffers
132. 25:06um but we who we met with was someone who was counsel so an an actual lawyer um from the Holly office um and he was
133. 25:14wicked smart like he knew the issues he knew the laws like when when I would say things like you know small companies get
134. 25:22swept up in this overly broad legislation he was like well how is it too broad tell me specifically so he was really digging in on specific
135. 25:31things um and so there was probably 10 minutes of just harsh cross-examination let's say um and I you know I think we did good I
136. 25:45think I think we navigated the the cross-examination well and he was a bit more pragmatic he said um I think he was
137. 25:54the first one that said to us um he said yeah the the AI stuff was really hot for a while and it's not now right
138. 26:05because you know we've experienced that on this channel that there's this kind of lull um in in AI news and excitement um
139. 26:14well it it translated to Capitol Hill and so basically if it's not a Hot Topic um things tend to take longer if it's a
140. 26:21Hot Topic they rush it through so so that was kind of another indication that we're probably not going to see
141. 26:28anything we've got two weeks left in the legislative session now then we go into um election season then you go into lame
142. 26:36duck so like nothing's going to really happen until 2025 um and if AI is not a big Hot Topic then probably nothing's going to happen
143. 26:46for 2025 so that that was some more of that um we got in we got in some of our message to him he he took notes he was
144. 26:54he was super engaged he was super um like he he was appreciative that we were there he didn't really want to be
145. 27:07there like it was it was he's like I got lawyer [ __ ] to do but you know go ahead this is part of my job you know he had a
146. 27:14little bit of that attitude uh so that was that let's see I'm trying to think who else we who else we went
147. 27:23to we can all watch it with you okay let's see um so those two were Senators then I think we took a lunch
148. 27:40break yeah and then my next meeting was with um Diana deette who's a congressman so we were in the Senate building which is on one side
149. 27:53of the capital and then the house building is on the other side of the capit capital and it's like you look
150. 28:01across you can look across and see the other building but it's like a long walk and it was hot and so we were taking
151. 28:07cars so you would basically take a car basically around the block but it was a big long block um and so what's fascinating is so if
152. 28:17you if you haven't ever been there I I haven't ever been in the bowels of of these buildings the Senate cuz the Senators
153. 28:26have more power right cuz there's only two per state their offices are a little swankier and you know Swanky is not
154. 28:34quite the right word but you know very lawyerly and big and um the house offices are a little janky so so we're in we're in D's office
155. 28:48and some other Congressman came in and I was shooting the [ __ ] with him and he was apparently he was a little anti- aai
156. 28:54he was a little he was a little nasty but he was fine uh um but the person we were supposed to meet with was in some other meeting and
157. 29:04then there was a whole group of people in this teeny little office and they're like okay you know we can meet with you
158. 29:10now but there's no there's nowhere for us to meet so can we just meet in the hallway so like this gaggle of like
159. 29:18eight people sort of scurries out of the office and they're holding the meeting just outside the door um and then we're still sitting
160. 29:27there we're supposed to be me like a half hour's gone by like our our meeting has sort of come and gone and this very
161. 29:34pleasant woman comes out and she says sorry the staffer you're supposed to meet with is still busy but I'll meet with
162. 29:41you and she goes we don't have anywhere to meet so can we just meet in the hallway so we're on the hallway there's
163. 29:47this gaggle of out PE eight people screaming there's people walking by there's other meetings happening in the
164. 29:54hallway and it's you know it's government buildings so there's no carpet it's just like you know hard concrete walls and Halls very noisy
165. 30:04um so she gave us a solid 10 minutes that was fine um so that was Diana deette then we met with I forget who it was a staffer
166. 30:20from was that in the house yeah it was it was a congress person from Minnesota and we got there late because
167. 30:35we were late from that other thing and we couldn't find it and when we finally got there he had already kind of started
168. 30:40the meeting he was really [ __ ] nasty he was older he was smart anytime he he sort of took me he latched on to me to like he came at
169. 30:54me um and you know when I would say things like you know I would sort of say platitudes about you know how bad the
170. 31:03legislation was and how it affected people well exactly how does it affect people what specifically about it
171. 31:08affects people and he was just like coming at me he's like what we what we what we sort of put together after because what what was
172. 31:19fascinating is he was he was coming at me and I was I was just this side of breaking into a flop sweat cuz you know
173. 31:24how when you're trying to say something and someone's like you get a little like and then I thought I don't give a
174. 31:34[ __ ] if this guy you know so so I I avoided flop sweat Panic so I just kind of held my own for a while and then he
175. 31:42and then he went on and talked to other people and then at some point he actually made the connection he was really smart he really knew his
176. 31:50[ __ ] Kathy from rxn was talking about the cloud storage thing and he goes oh oh he goes you're just talking about not
177. 32:03legislating the technology but legislating the use case and we said yeah and and then at that point his guard came down what we
178. 32:12put together after the meeting is that he was I think he thought that we were anti-regulation that we were like we
179. 32:19don't want to regulate AI at all and we're not we're like just regulate it smart regulate it in a sophisticated way
180. 32:26that doesn't paint like Frontier Model companies like open AI with small businesses with four employees in them
181. 32:33right cuz they don't have the same resources to deal with your stupid [ __ ] regulation so like be smart
182. 32:38about it right like make a grid and a matrix and understand the nuance and you know regulate the the use case rather than the
183. 32:46technology um and so once he got that his tone completely changed and he was really engaged and he was taking notes
184. 32:54we gave him some ideas so that was that one um then we went to Senator Bennett's Office from Colorado and the staffer we were
185. 33:06supposed to meet with wasn't there the staffer that we did meet with was a lovely young lady she was maybe I don't
186. 33:14know 20 wait did you show cranky old Mana no I didn't I didn't I didn't do uh Kyle from the AI learning lab didn't
187. 33:22really show up in Washington it's you got to be political you know it's it's if ever there's a place where you got to
188. 33:30be political it's DC right you you can't you can't burn Bridges and do weird [ __ ] so although you know apparently you can do really if
189. 33:43you're shit's weird enough you can do any of it um so she met with us and she was okay she didn't really know the issues all
190. 33:52that much she was like a year ago there was an AI series of summits um and uh and she kind of sat she was a little
191. 34:05cross-armed and and she just sort of gave us Pleasant nods and things like that um so she was okay it it was that meeting was just I'm
192. 34:18taking notes I'll pass them along I don't really give a [ __ ] was kind of the tenor of that meeting um did you say you
193. 34:25can make money with chbt I didn't although the last me so the last meeting was the best meeting so we then went to
194. 34:32hickenlooper's office and the staffer there was named Sam I forget Sam's last name but Sam had like his master's
195. 34:40degree in computer science um he was a software developer he was super passionate about this he goes this is
196. 34:48what I do like I'm you know I do Ai and I do software and I do this and he had he had like four or five different
197. 34:54things that he did but AI was in it he cared a lot about it um he asked really good questions he completely got um all the stuff we were
198. 35:06talking about and then he even came back and said hey if we were to approach regulation this way would that work um
199. 35:15and and I forget how he was saying it oh one of the things we were talking about was um there's different actors is is how he
200. 35:28put it which I thought was a really good term right that you want to regulate the right actors right so you you've got the
201. 35:35frontier models that are building the these core models and then you've got people that are just using the models
202. 35:41wholesale then you've got people that are using the models and fine-tuning them right and then you've got some
203. 35:47people that are making their own models and right and so there's all these different actors in the ecosystem and so he was talking about
204. 35:55like how do we determine the actor right and and he goes you know should it just be based on like number of employees or
205. 36:02number of Revenue and I said I don't think that can actually work because you might have a company that's a single
206. 36:08person company that's acting like a thousanders company because they're leveraging the [ __ ] out of AI right like
207. 36:15and then you might have a 1300 person company that doesn't fall under regulation because they're using it in a
208. 36:21really limited way that falls outside of that so he actually got that Nuance um one of the other messages that I talked
209. 36:29about how we doing on time one of the other messages that I talked about was um I said when you're thinking about regulation
210. 36:43don't be very careful to write the regulation in a way that doesn't assume that the technology we have today
211. 36:55is how it's going to look a year or two from now right right because if you say hey based on this model or this
212. 37:01capability of this model here's the regulation a year from now it's not going to look like that so I said you
213. 37:08run the very real risk of writing regulation that will um be obsolete by the time it passes and Sam really got
214. 37:17that he really got that um I was really impressed with him he he talked about um understanding the Nuance between big
215. 37:28companies and small companies right he understood the the difference between traditional Ai and generative AI um he
216. 37:37had this whole concept of different you know sort of layers and sizes of of companies and it at least you know
217. 37:44figuring out a way to do that um so so that one was really good so and then I just went from there to the airport
218. 37:52um the the last meeting was was actually really promising you know he similarly said probably not going not going to be
219. 38:00anything from the feds um anytime soon but um I kind of feel like so I I think I met with six
220. 38:17people I feel like two of them were really good the Josh Holly lawyer was really smart the the hick and Looper staffer
221. 38:26was really smart I have a feeling the Bennett staffer had we met with the actual staffer would have been fine so I
222. 38:33feel like Colorado's in a good place oh all of the Colorado people I said I said I I met with this executive director that is
223. 38:42not you know pursuing AI because of that law they all took Furious notes on that one like the fact that I had a specific
224. 38:52anecdotal piece of data from from the day before like flying in of someone where there was a chilling effect
225. 39:00because of this law and we you know one of the things George and I were talking about was like we want the US to remain
226. 39:08competitive and if you put us in a in a cage where we can't do our [ __ ] we're not going to be competitive so when I
227. 39:16had a specific story of here's a company you know where the executive director made a strategic decision not to not to
228. 39:23learn AI not not to do anything with it because of the legislation that you wrote that doesn't go into effect until
229. 39:312026 already having a chilling effect crazy all right so that was Washington um let me I'll see if you have any
230. 39:44questions because I I've just been talking and not looking at the screen AI wants to know if you notice
231. 39:50bias towards or against AI along party lines um I don't I haven't been looking for it but I I did see I did see a Tik Tok
232. 40:02today that addressed that specifically she took a transcript of the debate and had chat GPT analyze it and
233. 40:12then she took the transcript of the debate and um de deidentified it took out the specific names of the players
234. 40:20just said candidate a candidate B and she said it got she got a very different result from that so that makes sense to
235. 40:26me good Kyle sounds like you made some Headway yeah I I'll tell you like the the Minnesota Congressman the the the one that he
236. 40:37actually turned in the meeting where he went from being very combative to being you know really listening to us that
237. 40:44felt like we made progress the senator Bud dude I I think he was fine I think he'll pass along the notes um Josh Holly's Council
238. 40:55um he took it seriously the deette Diana deette one where we had the hallway meeting I think that was just a you know a nice to have the the
239. 41:06Bennett one was like e that'll that'll go along but but Sam from hickenlooper's office I think we made a a real impact
240. 41:14there he was the first one Sam was the first one so so one of the things that we said to all of them
241. 41:21is have small business at the table when you when you're you know thinking about this legislation and drafting this
242. 41:29legislation have small business in the room right so that you're representing the users of this technology not just
243. 41:37the manufacturers of this technology and um Sam was the first one I really got the sense that you know I'm
244. 41:46a constituent and um I really got the sense that you know they could potentially reach out at some point and
245. 41:52say hey we want to talk and here's what we're thinking about doing and what's your feedback on that so
246. 41:57um what Kathy from rxn told me was that I said you know what's the what's the end goal here is is the end goal that we
247. 42:06you know affect any specific change she said no she said she said there's no specific legislation that we're lobbying for or
248. 42:16against um she said this is relationship building right this is get the concerns into the conversation
249. 42:27so even if there's no specific action at some point in the future there might be just something that sticks in their head
250. 42:33you know we should probably make sure that we're separating big companies from small companies or you know so so
251. 42:39Concepts get through and then she said what what this really is is that over time you keep having you know these
252. 42:44meetings and you know the staffers get to know you and the staffers are the ones that that do all the work and so at
253. 42:52some point if they want to bring in a voice um for that particular issue they'll have record of all this and
254. 42:57they'll they'll call you call you out and you know bring you back out there and things like that so thought that was
255. 43:02fascinating um yeah so that's that uh Kyle for president his slogan is embrac the Jank four more years no My My
256. 43:12My slogan for 2028 is is it can't possibly be worse where where the AI is actually the president and I'm just the fancy little spokes model
257. 43:34Opie overwhelming all at once so I've gathered top experts in the field like Bill Gates Sam Alman and more to help us
258. 43:43understand what it is this is crazy to me this is the biggest technical advance in my lifetime Ai and the future of us an
259. 43:53Oprah infre special tonight an Oprah infree special that if if you're in the Mountain Time Zone go [ __ ] [Laughter]
260. 44:03yourselves there's a pre-show in mods FYI what's a
261. 44:18pre-show oh I see GPT 401 demo Hume update new voice model oh yeah I saw that I haven't played with it yet Gemini
262. 44:29live now available for everyone okay yeah we can so we could demo in the next 20 minutes we could go look at the new
263. 44:3701 model I can't believe they called it [ __ ] 01 gp40 and then 01 not 01 no
264. 44:52no Kyle want me to come up and share some legal stuff I learned learned I would love that yeah hop up here Ann Murphy
265. 45:06um oh I want to show you also the new notebook LM so notebook LM is something I haven't really talked about on this channel and
266. 45:16I it it need it needs to be talked about now um I probably should have talked about it a while ago but holy [ __ ] did
267. 45:26they just add a it a crazy new feature um so yeah hop up here Ann whenever you're ready oh remind me how oh I don't know you go to
268. 45:41somebody can somebody tell an how to click on the button which button where it is cuz I can't can you please volunteer to handle
269. 45:50their marketing oh can we please volunteer to handle their marketing honest to God seriously you've got a new thinking
270. 45:58model you can't call it like GPT cognition or GPT thinky mington GPT or it's not even called GPT it's just called 01
271. 46:14really really right by the comment box ah there she is an muy okay hey can you guys hear me uh wait my
272. 46:30volume's too low but go ahead yeah I can okay I'm G to start driving in a second so I'm just GNA be off camera is that
273. 46:36okay yeah that's fine okay so I just finished Paul reter's um conference oh yeah yeah it was amazing we had so many
274. 46:49like AI luminaries there and then a thousand people who kind of spanned the Spectrum it's all it's all marketers
275. 46:58some like agency people some like contractors some like you know entrepreneurs blah blah blah nice mix of
276. 47:07people y um but one of the sessions I kind of hated because what we learned is that the copyright cases are definitely going to
277. 47:22go to the Supreme Court yep and because of that and the additional scrutiny that will go that will you know go on listen
278. 47:32to what we are supposed to do our very own selves the recommendation because if we ever get sued because believe it or
279. 47:42not even if we taught someone and they go and do something stupid um or if we if it's like one of our clients or
280. 47:55ourselves listen to what we're supposed to do we are supposed to be keeping track of all of our search all of our conversations with Chad
281. 48:06GPT so that we can prove that AI wasn't part of it I mean sorry we can prove what was Ai and what wasn't and what
282. 48:17human contribution yeah yeah yeah you you know about this and that we're supposed to be able to yes produce
283. 48:26why we did it how we did it yeah the prompts and all of that that if we were to like do it on the up and up so
284. 48:36everything everything you do you would have to be keeping track of which just absolutely blows my mind yeah no it's
285. 48:43it's absolutely ridiculous and it's and and it's also just like I I've got this new book coming out called collective
286. 48:51intelligence and like a lot of it was written by AI but a lot of it wasn't right and it's also like there like most
287. 48:59of the concepts in there are [ __ ] that Kevin and I have been talking about for a year right yes exact prompting right
288. 49:06and exactly exactly and so and there's even there's even things like the um I don't know if you saw the thing where
289. 49:13timeland like his his the the rapper his his producer took a little beat that he made and put it Ino so it started with a
290. 49:21beat that he made theno basically finished the song like where's the boundary like what that is and what of
291. 49:28that isn't like it's it's absolutely ridiculous so I'm with you so I don't know so that that was a little bit
292. 49:35stunning and made me feel like oh [ __ ] great now I got to get like some kind of attorney and blah blah
293. 49:42blah yeah I don't know I've got I've got bad attitude on that because I'm just like well I do just because so many
294. 49:49people are going to be using so much stuff like and and so much of the work is going to be hybrid human and Ai and
295. 49:58not just like single AI like multiple systems are going to be used to go into stuff right so it's like the ability to
296. 50:06track all that's like literally impossible it's literally impossible to do that at that point it would be so
297. 50:13cumbersome you'd have to be like okay that's okay I guess I'm just going to go back to doing things the old way yeah or
298. 50:21something um so that was a big learning the other one wanted to share is that if there's a lull in AI right now not one
299. 50:32single PE person of that thousand people have any idea that there's a low like they are I mean they're marketers what was
300. 50:41what was the M what was the mix of you know I I'm here and I don't really use it but I think I might start to people
301. 50:48that are actively using it I assume that conference it was a lot of people using it but what yeah I would say there was
302. 50:55probably only like 5% of people who aren't using it and then probably 30% of people who are just like dinking around
303. 51:04and want to figure it out and then everybody else is the the one of the takeaways is that like the big big gap
304. 51:13is nobody's having it's not been easy for anybody to go from like some of the things you would do on an AI road map
305. 51:21and actually scaling it like all companies are having a hard time scaling it so that was interesting that makes
306. 51:30sense to me I don't think the tools are ready to scale quite frankly well that yeah that's part of it is people get burned and then the you
307. 51:39know the people who make the stuff or who are like the Brokers for the people who make the stuff like it's their brand
308. 51:48reputation and stuff so they don't want to give gadgets to people and be like well let's just cross our fingers and
309. 51:55hope that everything works out okay but the problem is is that like with co-pilot in Microsoft it's not even there's no
310. 52:07effort involved with scaling like it's it's going to but there people won't have any training they won't have any
311. 52:15governance like no one talked about governance at all nobody not even a single like hallway conversation not a a Breakout out I mean
312. 52:27they had we had like I don't know like there was like 60 speakers there and if it was if it came up I would I mean it's
313. 52:36not it wasn't in the write up on any of the sessions so that was surprising what was did you have what were there any sort of breakthrough
314. 52:44moments that you had or anything that got you really excited because yeah yeah thing I get but what what was on the other side yeah okay
315. 52:54um one of the things that I was really excited about was how excited everybody everybody there was like that seemed so
316. 53:02promising everybody has all this pent up energy and they want to do things and that's really cool um Liz grenan so
317. 53:12Connor grenan interviewed Liz grenan and she talked a lot about stakeholder trust and that's a topic that I really care a
318. 53:20lot about and there were some like good ideas about how you start working with people like you would should really only
319. 53:31work with the people who are ready to be on an adventure with you you shouldn't work with anyone like the the some of
320. 53:37the advice was like if you're in a space where people you have to teach people or convince anybody just leave the room
321. 53:46like we're not doing that yeah this is really good this is this is that that article that I wrote rise of the agile
322. 53:52human that that the people that you want to that you want to invest time into are the on on that are curious and hungry
323. 53:58and excited and flex right you know brave brave enough to break some [ __ ] right like like yeah I I I that's that's
324. 54:07good reinforcement because yeah yeah this stuff is to the the the all the aite tech right now is too malleable to
325. 54:17to have people in there that want predictable [ __ ] right yeah exactly exactly the other thing I really loved
326. 54:25was Rachel Woods um was there and you know I'm in her community and I've done courses with her and all that kind of
327. 54:34stuff but she shared so much about how you build your playbooks for AI and using AI to do it so that basically what
328. 54:46you're doing is you're creating these assets for your business where you're taking these like groupings of tasks and
329. 54:52you're just like you know totally making it routine and then you just um upgrade it like incrementally as you go and then
330. 55:00the goal is of course like those become the assets for your business and then you can hand off one thing after another
331. 55:07after another to other people and get yourself so you're only working on the highest level stuff so she's doing like
332. 55:16a four week course on it soon I think she's waiting to see how many people are into it but um yeah so to teach us all
333. 55:25how to do that which I like for me that's money in the bank yeah that's really cool that reminds me so Paul rer
334. 55:33last week released um problem GPT he's got like gpts now yes it's so good yeah it seems that seems like it's in the
335. 55:42same neighborhood it so so so so good and and uh the other thing he shared I don't know if Source Camp is is in is on
336. 55:52the live she was earlier yeah I'm pretty sure she's so this would be particularly relevant to Source Camp she he gave us
337. 56:00well he gave the people who bought his scaling AI course um a 29o assessment tool where what you do is you go through
338. 56:10the company that is your client and you learn about their business operations and everything but each task you
339. 56:16determine what is the exposure to the positive impact of AI so then you can H you can prioritize things so you're like
340. 56:26okay this one this one you know has a 70% likelihood of being better with AI maybe you know you go ahead and you do
341. 56:36that and so then you begin to be like cherry-picking you know the the projects that you work on so that the client gets
342. 56:45like value really quickly yeah um so that was really cool and then I met a bunch of people who solve a problem that
343. 56:53I don't have anyone to solve for which is when people people's data is a hot mess and you start going in there and
344. 56:59trying to do stuff and it's just like you know I can't even help you because your data is a hot mess so I I met so
345. 57:07many wonderful people there and like especially collaborators so great I hope you poed up the salon get him in the
346. 57:15salon oh I would love to do that he is like he is such a cool cat who Paul rer is yes yes yeah he's cool were you there
347. 57:27when he he spoke at the salon were you there that night no he was in he was already in the salon yeah well he spoke
348. 57:33he came and spoke at one of our one of our meetings oh man well the other thing I wanted to mention because people might
349. 57:42be interested in this kind of thing for future people there were walking around with rfps printed out to try to find
350. 57:50someone to train their teams to yeah to build [ __ ] for them I mean literally I walked away with people with rfps where
351. 58:01people had broken down like a Playbook and said this is exactly what we want to do all we all we need to know is how
352. 58:08much it costs and how how long it's going to take wow that's amazing W so I gave it to Josh already so Josh had you
353. 58:16know my friend Josh Houston he has it in his hands already and he so yeah I mean there's money by going to these conferences like
354. 58:26oh my God I made so I met so many potential clients and stuff it was so cool well that's great that's really
355. 58:33great awesome wanted share Co I forget how to I forget how to kick can you kick me off so I go back down I don't know
356. 58:41how yeah I'll get you out of here this this just leave an just go hie thank you bye all right
357. 58:56that was very cool I forgot she was at that at that conference yeah that's super cool um awesome fantastic all right so we are
358. 59:06did you get my memo about the RFP cover sheets we're gonna we're going to need you to all right what was that
359. 59:13conference called please that was called it was the uh uh makon maai marketing AI Institute ma AI n and it's it's the the marketing AI
360. 59:24Institute um they have the annual their annual shindig in Cleveland Paul rater is the is the I guess the CEO of that organization and
361. 59:36the main you know sponsor of that event um I got to go get some more water so entertain yourselves and then we'll be back into
362. 59:43Opie we will do Opie I'll be right back chiy everybody
363. 1:00:24after e
364. 1:01:03for
365. 1:01:54spe
366. 1:02:24e e
367. 1:03:20all right we're back we're back tonight on the AI learning lab Oprah Winfrey AI for everyone you want
368. 1:03:34some all right what's happening everybody what's shaking what's going down um we got three minutes left so I'll flip over
369. 1:03:44here time for the AI party Jim Ross is in the house hey Jim you're um I I invoked I invoked your name and your story a fair amount
370. 1:03:56um in DC the the thing that you said about you feel like you you're you're operating like a 10p person company you
371. 1:04:04know as a solo preneur um that went over really well like I tried a number of different things of specific stories um
372. 1:04:13the two that went over the best were that one and then the woman that I that that I met on the plane who said she's
373. 1:04:19not going to she's not going to incorporate AI into her business at all because of the Colorado law even though
374. 1:04:26it doesn't go into effect until 2026 she was just really pragmatic she's like I don't want to put time into that if uh
375. 1:04:34you know if if I'm just going to get regulated you know it's going to create this regulatory burden I was like damn laws actually have an
376. 1:04:42[Laughter] impact so yeah I know it was really cool it was very cool I told him I told him the one guy that was really interested
377. 1:04:54when I said that he was like oh that's amazing I said I said yeah he's he's got this this practice where every morning
378. 1:05:01for an hour he uses AI for his business he was like taking notes it was pretty cool all right let's see scripts we're doing okay let me let
379. 1:05:13me resize my window so that we got most of the screen on for y'all oh that's good we can just pop it right there and and
380. 1:05:27right do that oh that's good look at that we're going to get full Opie full Opie all right how are we going to know
381. 1:05:39how they going to know if they incorporate AI into their business well what what she was saying is like
382. 1:05:49she doesn't even she doesn't know right she's not using chat GPT enough to know how she would incorporate it but what
383. 1:05:56she said to me is when they passed that law she made a conscious decision to not pursue it to not learn it to not create
384. 1:06:06an AI committee to learn how they might do [ __ ] better because she's like I don't want to put time into something if it's just
385. 1:06:14going to create a headache down the line was that the DC trip yeah that was I I've been talking for the most of the
386. 1:06:21last hour about the the DC trip yeah no I know it it was really what was amazing was she and I were talking about chat GPT for for a long
387. 1:06:30time um and then um and then she uh once I talked to once she found out I I testified against that law she was like I didn't know where you
388. 1:06:44stood on that law so I didn't really say anything but once she knew where I stood on that law then she kind of admitted to
389. 1:06:49me that that was a choice she made so bonkers bonkas bonkas bonkas all right
390. 1:07:09Dion Oh let me cover up my keyboard where's my paper where's my black papier all right here she comes is this her no that's an ad councel
391. 1:07:27thing on social media the first thing that kind of comes to mind is just like the Terminator movies we've always been
392. 1:07:36interacting with some sort of artificial intelligence it's just much more in our face right now what scares me about AI
393. 1:07:43is that the technology seems to be advancing so quickly one day we weren't talking about it and then the next day
394. 1:07:52it wasn't just that we were talking about it it was there for to use without really even understanding it AI is
395. 1:08:00happening it's going to happen the question I would have for anyone who has power over the development is how are we
396. 1:08:08going to make this work for us I don't know what the agenda is with the people who's even creating AI to me it's like
397. 1:08:15what is really the point why do we need AI there you go yeah this is very much designed for introducing this to the world
398. 1:08:27fascinating I've been living on this planet for seven decades I've seen a lot of things change but about a year ago I
399. 1:08:36saw the first change I didn't fully comprehend it was a demonstration of the capabilities of artificial intelligence and I was
400. 1:08:47bedazzled fascinated curious and also concerned but there's no doubt it is here artificial intelligence is here and
401. 1:08:59life for all of us is about to be very different over the next hour we're going to talk about what is coming with a
402. 1:09:07brilliant group of people they have different backgrounds and opinions about the kind of future racing toward us but
403. 1:09:15they all agree the stakes are too high for anyone to sit this out now if you're worried this is going to be too
404. 1:09:24technical I promise it is easier than you think yep that's great this is the notion a computer that works like the
405. 1:09:30human brain has been around since the 1950s this is good the first chatbot arrived in 1966 but computers at the time weren't
406. 1:09:38powerful enough to really engage well it's been around for decades forever hooked but but it hasn't been
407. 1:09:46accessible to to Everyday People until data to learn from most recently a group of computer scientists taught machine
408. 1:09:56there's the Transformer find and process words more easily so they can understand how we naturally communicate that new
409. 1:10:04discovery is powered by these computer chips called gpus combine smarter programming with more powerful chips and you get AI
410. 1:10:14products like chat GPT there's your 2022 date that's the beginning of the revolution in my opinion so far beyond
411. 1:10:21our capabilities that it can seem like reality itself is warping Sam Alman is the co-founder of open AI which makes chat GPT I promise
412. 1:10:36everybody watching that this was not going to be so technical that they wouldn't understand you will do it all right okay what is artificial
413. 1:10:47intelligence it's really the next chapter of computing we have figured out how to make computers smarter to
414. 1:10:53understand more to be able to be more uh intuitive and more useful can you tell us in basic terms how chat GPT
415. 1:11:01works you know at the most basic level we are showing the system a thousand words in a sequence and asking it to
416. 1:11:08predict what word comes next and doing that again and again and again so we've seen this on our phones where it
417. 1:11:14predicts you know what the next word's going to be exactly but to do that as well as possible think how much you have
418. 1:11:21to understand about each word that comes before and the relationships every word and every word yes and by teaching the
419. 1:11:28model to look at a bunch of text and predict what word comes next and doing this incredible scale again and again
420. 1:11:36the system learns to predict and then in there it learns the underlying concept so that's what's going on behind the
421. 1:11:41screen yes how fast is this moving well four years ago most of the world if they thought about AI thought
422. 1:11:49about self-driving cars or some other thing or whatever it was only 2022 was the first time people said okay this
423. 1:11:58GPT thing like this is computer talks to me now that's new and then since then if you look at how much better it's gotten
424. 1:12:05it's been a pretty steep rate of improvement just take me to 2035 um I think it's hard to say where all this can go without sounding like a
425. 1:12:15crazy person I really believe the technology is going to be that powerful I hope that in 10 years the dramatic music of us we are limited by
426. 1:12:26what we can imagine we have this AI that can help us figure out anything we have abundant energy that can help us like reconfigure
427. 1:12:34stuff in the world and if you want to go build like a beautiful house you can dream with the AI you can figure out oh
428. 1:12:42I want it to look like this or do you have any other ideas and architect this amazing thing and then can you invent
429. 1:12:47some robots that can that can actually build this this house that we just imagined together so the machine that just
430. 1:12:55becomes a part of your family the way the phone is a part of our Lives now it becomes a tool for you to do whatever
431. 1:13:01you want for your family do you think it's moving too fast my current belief is no but I could imagine things that accelerating would
432. 1:13:11change that quickly and I would say okay this is moving too fast like what what would change that if we made an AI
433. 1:13:18system that was better at doing AI research then all of open Ai and you could push a button and just say improve really fast
434. 1:13:26that's an example this I think we be moving too fast if we're talking about AI benefiting everyone the industry is
435. 1:13:33dominated by a lot of people who look like you male and white and very few who look like me bias this thoughts about to
436. 1:13:43address that yeah obviously we want everybody to see themselves in our products uh we also want the industry
437. 1:13:51Workforce to be much more diverse than it is um and there's slower than we'd like progress but there is progress
438. 1:13:57there the most important thing that will happen here is everyone in the world will be able to create at a level that
439. 1:14:04is still hard for us to imagine this is going to be an enabler of human ability to create to flourish to make new things
440. 1:14:14to create new companies and services like we've never seen and we want everybody to get to do that not just the
441. 1:14:19white dudes that's good yeah what keeps you up at night the rate of change in the world from this we haven't seen it yet
442. 1:14:32we'll do because we haven't seen that yet doesn't mean we can get complacent let's say we can make the
443. 1:14:38technology super safe we can't kind of do anything about the rate of diffusion through the economy like people want
444. 1:14:44more and better and there's sort of some strong forces at play here if you were in charge of regulations what kinds of regulations
445. 1:14:55what kinds of boundaries what kind of restrictions do you believe should be put in place a partnership between the
446. 1:15:04companies developing this technology and governments is really important one of the first things to do and this is now
447. 1:15:10happening is to get the governments to start figuring out how to do safety testing at these systems like we do for
448. 1:15:17aircraft or you know new medicines or things like that and then I thinkr now we'll have an easier time
449. 1:15:24figuring out exact what the regulatory framework is later who do you speak with in the federal government we speak with a lot
450. 1:15:31of people in the executive branch um and then uh dozens of people fairly regularly in Congress um there's a new
451. 1:15:39AI safety Institute in the US that we work with I personally probably have some conversation with someone in the
452. 1:15:45government every few days every few days wow and what are y'all talking about what the US needs to do to lead here
453. 1:15:55how we're going to build enough data centers how we're going to deal with chips and where those chips are produced
454. 1:16:00and make sure politically we have a good strategy there how to think about safety for AI and how the government should
455. 1:16:06test it how to think about collaboration with other governments um what the economic impact that this is going to
456. 1:16:12look like how the government should be using this all that the bottom line for I think most people is whether or not
457. 1:16:21you can trust the people who are in charge the the um I know you went through a phase where your own board
458. 1:16:29said we don't trust you and then you got yourself another board how do we know we can trust you so the bar on this is clearly extremely
459. 1:16:41high um the best thing that we can do is to put this technology in the hands of people talk about what it is capable of
460. 1:16:49what it's not what we think is going to come what we think might come give our best advice about how Society should
461. 1:16:59decide to use this um say when we think it's important to not release something which we also might get wrong and build
462. 1:17:06up that trust over time but it is clear that this is going to be a very impactful technology and I think a lot
463. 1:17:15of scrutiny is thus super warranted I actually saw a headline that said you were the most powerful and perhaps most dangerous
464. 1:17:25man on the planet and I'm wondering how that sits with you I it's it's definitely strange to hear you say that it doesn't I don't feel
465. 1:17:39like the most powerful person or anything even close to that like I feel the opportunity responsibility in a
466. 1:17:44positive way to get to nudge this in a direction that I think can be really good for people and that is like that's like a serious
467. 1:17:58exciting somewhat nerve-wracking thing um but it's something that I feel very deeply and I and I and I and I realize I
468. 1:18:06will never get to touch anything it's important started 10 minutes ago just started just what we seen thus far of
469. 1:18:13people with nefarious ideas misusing it um causes me a lot of concern and yet when I sit with you you're so calm you're so
470. 1:18:26relatable it seems like everything's okay is that how you really feel I don't want to give a false sense of security
471. 1:18:34here uh I think I'm I'm tremendously excited for the good I think the good will be incredible but there will will
472. 1:18:40be bad too and we'll mitigate as much as we can we the whole field um but this technology will be misused like every
473. 1:18:47other technology before it and we need to be cleare eyed about that even though even though I'm so convinced that the
474. 1:18:56upside will be so tremendous and I think we're all so excited about it um we do have to be responsible for the whole [Music]
475. 1:19:08package yeah really interesting really interesting I mean you know listen he he's this is a you know this is a this is an edited piece
476. 1:19:18right um I would assume that this this was likely architected by the pr team at open AI is my guess um you know not quite a puff piece but
477. 1:19:33pretty close to it so far um this but this is definitely this is definitely feeling designed for the people that haven't
478. 1:19:43heard of chat GPT did you notice at the beginning when she said you know Sam Alman the the you know from open AI the creator of chat
479. 1:19:51GPT she said it like she was talking to second graders right um this is definitely designed for that
480. 1:20:02the the downside of it so far like while while he's explained it okay it still seems like nebulous like it's a machine
481. 1:20:13like when he talked about it's a thing that predicts the next word people don't know what the [ __ ] that means if he said
482. 1:20:20you ask it a question and it answers you uh like it feels like he could have been a bit more clear about what it actually
483. 1:20:29does right like like I like if if people didn't know what chat GPT was the fact that it predicts the next word over and
484. 1:20:38over and over again what does that [ __ ] mean well you know what that means is it can write you a poem it can write you
485. 1:20:46you know a chapter of a book it can write you a computer program oh and it can just write it you know that like
486. 1:20:53that that's a bit missing right now but we'll see I assume who who she probably goes to next is Bill Gates so what Bill Gates does for
487. 1:21:03this is you know because he was sort of the the poster child with Steve Jobs for the for the opening of the
488. 1:21:13um for the beginning of the PC and then when the internet came along you know Microsoft was really slow but when they
489. 1:21:21decided to go they went right and so and then and then because of the Gates Foundation he's got a Global Perspective
490. 1:21:28so he's got Gates has got the ability to I I'm sure he'll talk about the first time he used chat first time he saw chat
491. 1:21:36GPT I think Sam Al he Sam Alman came to Bill Gates's house and showed it to him he'll probably talk about that and talk
492. 1:21:44about how significant it is so you know Sam's Sam's there to be the calm voice of it's going to be okay people the
493. 1:21:52robots aren't going to kill you and I think he he does that pretty well and then um and then Bill Gates is going to
494. 1:21:59say this is important and it'll it'll be interesting to see what the other what the other folks do um by the way if you're wondering you
495. 1:22:09know where this is so I'm in Denver so we're an hour delayed so this already happened on the East Coast so we just started at the top of
496. 1:22:19the hour so we're 15 minutes into this into this show I use perplexity a lot okay she's definitely asking alarmist questions yeah I mean
497. 1:22:33it's I I mean it's the media right like I I mean even when when I was in DC like a lot of what I talked about was you
498. 1:22:43know I think part of what we're swimming against here is Hollywood has told us for 50 years that the robots are going
499. 1:22:48to kill us and every time I said that they were like yeah yeah um I asked three people the direct question do you use chat
500. 1:23:01GPT no one is no one is it's back on well it's not back on here oh yeah it is sorry about that holy [ __ ] two holy [ __ ] oh Maris
501. 1:23:15browny marz brownley have 20 million subscribers who all rely on you for all things Technical and this is a moment
502. 1:23:25we're having tell us what's going on yeah look I I've been in the tech World a while and a big part of being in this
503. 1:23:31world is trying to figure out when this new thing is a big deal YouTuber Tech expert I can say confidently it does
504. 1:23:38feel like a big deal I think it's a very big deal yeah I want to show you what an AI generated video one year ago looked
505. 1:23:46like oh this is great okay and it's this this is great this is the rock eating rocks okay it's kind of a meme it's it's
506. 1:23:53a pretty funny video it's obviously not a real video obviously the a generated now you can tell that and you
507. 1:24:00look at that and you say okay that's a pretty ridiculous looking video yeah this was one year ago yeah right now now
508. 1:24:09this is what AI generated videos look like what do you mean this is not a real person not a real scene no no real
509. 1:24:17cameras were used this is completely generated by an AI generated video tool now you can still kind of look at pieces
510. 1:24:25of this and something's not quite right no I can't but that looks like a video uh here's another one this one's a
511. 1:24:34little more interesting it's a a handheld cinematic shot of a child on a beach holding an ice cream con that's
512. 1:24:40what you ask for that's what you typed in that's what I typed in I'm assuming that AI had dist stored somewhere the
513. 1:24:46picture of the Child AI has millions of photos and videos of millions of children stored that it's trained on
514. 1:24:57okay are they real children is this a real child it's not a real child but it looks like one because it's trained on a ton of real
515. 1:25:05children Is that real ice cream sure it looks like it so and this is again like if you pay super care that's really it's really
516. 1:25:15interesting she what she's asking right there is basically is this thing copying and pasting real people like is this
517. 1:25:24compositing real people together she's not quite grocking yet that this is generating a original content out of all
518. 1:25:34of the training data which is that's where everyone's going to be with this right everyone everyone that hasn't used
519. 1:25:40this and hasn't really looked into it is thinking oh it's copying and pasting other people's work it's not how it
520. 1:25:46works so that's that's really fascinating AR really any rules to it so everyone who makes tools like this of is
521. 1:25:54coming up with their own rules on the Fly oh I'll add a watermark to the corner or I'll add some metadata to the
522. 1:26:02file so if someone downloads it then they we supposed to know where to look for a watermark it's a couple layers
523. 1:26:07deep so the point is there isn't really any good set of rules yet I know this has happened to you it's happened to me
524. 1:26:15is to have people take an image of you put words into your mouth that you didn't say you're selling things you
525. 1:26:20don't even know you're selling I mean I can't even keep up with all the things I'm about to show you actually is a
526. 1:26:25voice generator yeah this is actually a real one right now that's trained on my last video yes and you can ask it to say
527. 1:26:32whatever you want and it will sound just like me is there anything you would you'd like it to say I'd like to say
528. 1:26:38Peter Piper picked a pack of pickled peppers Peter Piper picked a pack of pickled peppers Peppers let's see if we
529. 1:26:47can do it okay generate speech 11 labs to the untrained ear this sounds basically exactly like a recording of my
530. 1:26:55voice and even to the trained ear it's pretty convincing Peter Piper picked a pack of pickled peppers oh my God so in a matter of
531. 1:27:06seconds this tool is generating really high quality recording level this is good this isn't good so that's what I
532. 1:27:16mean so yeah she's very much in the alarmist place fasinating but it leaves me a little bit more concerned you don't seem to be as
533. 1:27:27concerned I'll say I've seen this story before cars in general went through this smartphones of course went through this
534. 1:27:35computers went through this the internet went through this where at their very formative years they're really really
535. 1:27:41confusing and potentially amazing but potentially horrible 80% of users just swore off AI forever humans sort
536. 1:27:49themselves out to lay the rules down in a way that we can actually trust that will be better than it is bad so I've
537. 1:27:58seen the story before and I'm hoping it plays out the way it's played out before thank you no problem next we have
538. 1:28:05Tristan and ASA who here to talk about the incredible speed in which this is happening it's moving really fast um we
539. 1:28:14had experience with another technology that was introduced a little bit recklessly and that was social media and
540. 1:28:19we're here because we don't want that to happen again with AI what was your experience I made infinite scroll you
541. 1:28:25know that thing where you scroll and the phone just keeps loading more I invented it before social media and I invented it
542. 1:28:33to make people more efficient and it's now used to waste over half a million human lifetimes every single month
543. 1:28:42addiction Doom scrolling polarization these are all consequences and what I learned the hard way is that my good
544. 1:28:50intentions as an inventor just wasn't enough and is moving way faster and is way more powerful in its impact on
545. 1:28:58society than social media I thought marquez's example uh actually caused me to have an aha when he was talking about
546. 1:29:06all technology has gone through a similar phase where people question he invented infinite scrolling so he
547. 1:29:14invented you know how you scroll and it loads the the extra things at the bottom so so you can do that
548. 1:29:24well you say what to that it took time it took decades to go from the invention of the car to adding seat bels and
549. 1:29:34driver's licenses and speed limits that's right we had to figure it out so the problem with the AI is that we don't
550. 1:29:41get that time so explain to me why AI is now moving so fast in 2017 there was a huge breakthrough at Google and research
551. 1:29:54Searchers there discovered a way that AI could start to learn a kind of universal language where it could speak the
552. 1:30:01language of anything and then suddenly the same lines of code could learn the language of DNA and learn how to make
553. 1:30:10drugs could learn the language of images and could learn to do all those things that Marquez was showing and so the rate
554. 1:30:16of innovation went through the roof and of course once AI can do anything people start to use it to do everything that's
555. 1:30:24what's so confusing about AI is it's going to give us all these benefits at the same time that it's undermining the
556. 1:30:30foundations that we depend on and you can't separate the promise from the Peril so I think instead of getting
557. 1:30:36caught up in the question is AI going to be good or is it going to be bad it's about is it happening at a pace that our
558. 1:30:43society can respond appropriately but guys it's that that that's really important I'm interested to see what she
559. 1:30:49says here but it's like that the whole point of this channel is to judge whether it's good or bad I don't give
560. 1:30:56too much of a [ __ ] about whether it's good or bad what I care about is it's [ __ ] happening whether we want it to
561. 1:31:03happen or not and what this channel is about is like if it's happening whether you want it or not don't have it happened to
562. 1:31:12you like have you happened to it right can respond appropriately but guys it's here that's why I started this
563. 1:31:22program by saying it is here so I I the argument isn't whether or not it's going to continue or whether the speed's going
564. 1:31:29to continue it it because it is already here it's already here some of it's here but even more powerful forms of AI are
565. 1:31:35coming open ai's stated goal is to be able to build systems of AI that are even smarter than humans because there's
566. 1:31:43trillions of dollars to be made on the other side and so now is the time for us to say what are the choices that we want
567. 1:31:49to make about how this technology comes out what do you want to see happen now open AI Google Microsoft they need to be
568. 1:31:59for the har they would create so that they're all incentivized to pull back and go at the pace that they can help
569. 1:32:06Society be prepared and not be overwhelmed and who is holding them accountable we need loss yeah interesting interesting interesting
570. 1:32:19um yeah this is definitely what what they haven't quite done yet and they may not right this is this is mass media what they've successfully
571. 1:32:35done is said holy [ __ ] it's coming it could do really bad [ __ ] nothing you can do about it is essentially what the message has been so
572. 1:32:45far um what they haven't done is kind of the the the optimistic view of well here's the good things it's done here's
573. 1:32:54the good things it's doing here's the accessibility it's providing to people um you know here are people that are
574. 1:33:03using that video tool in in positive ways um things like that so the AI version of when sharks attack yeah this
575. 1:33:11is this is Shark Week AI shark [Laughter] week uh you know what's actually fascinating is I wonder if viewers ship dropped off
576. 1:33:25after the after he showed the voice thing and she kind of freaked out a little bit shark NATO yeah exactly AI shark NATO
577. 1:33:40fascinating I just saw an email let's see where's it posted
578. 1:34:01where am there I am um it looks like the um the AI Salon got um we sent out a press release today from my trip to Washington and it looks
579. 1:34:13like we got picked up in a political article so I I'll look for that in a second actually maybe I wonder if it's over here
580. 1:34:41[Music] but
581. 1:34:57D people love a good train wreck I bet you can fast forward since you paused yeah I probably can [Music] boom yeah it's back I've been in a
582. 1:35:12terrible accident and I am so so sorry and a man gets on the phone and he says is this chast's mother she interrupted
583. 1:35:23deal that I was doing and I lost $900 because of her so I want my $900 back or I'm going to kill her so I'm driving
584. 1:35:33like 100 mph down the interstate and just as I got ready to get off the interstate a text come in and it was my
585. 1:35:41daughter and she says mom what is going on I'm at the doctor's office my immediate reaction was how the heck did
586. 1:35:48they get my voice um how did they know I call my mom mama I'll tell you what I I learn from it I will not ever take
587. 1:35:57another phone call unless they're in my contact list wow more scary AI is going to kill you that's robots are going to kill us
588. 1:36:05bad guys okay this this is this is the robots are going to christop want talk about threats from Ai and other aspects of law in order do
589. 1:36:18you remember the first time that someone showed you evidence of AI being used to commit a crime what was your reaction
590. 1:36:28yeah if I if I said a bad word we might get rate limited or they might kick long time even before I became FBI director
591. 1:36:34but one of the first memories I have of dealing with it in this job uh was I was in a conference room and a bunch of our
592. 1:36:43folks got together to show me how AI enhanced deep fakes can be created and they had created a video of me saying I
593. 1:36:53had never said before and would never say uh and I was staring at this video of myself and I found it incredibly
594. 1:37:01convincing and believe me it caught my attention I kep say wait that's not me I never said that what is this you know oh
595. 1:37:08the lady on the video right now ai is most useful in taking what I would call JV criminals and making them
596. 1:37:18varsity and that's a problem sure we had a a case not that long long ago with a child psychiatrist in North Carolina who
597. 1:37:27took images that he was able to get online and then manipulate them using AI to make them into child pornography one
598. 1:37:36of the things that made this particular example uh so egregious and horrifying is that the child psychiatrist took
599. 1:37:46photos from decades earlier from first day of school a school dance that kind of thing and essentially was able to
600. 1:37:53manipulate those images to show them uh naked wow there's so much out there now on social media and on the internet
601. 1:38:05well-intentioned things that people post about their lives but in the wrong hands enhanced by AI people are
602. 1:38:15vulnerable so you said more and more types of crimes so that would be one another one is something called
603. 1:38:22sextortion which more and more often somebody posing as a peer targets uh a teenager typically more and more it's
604. 1:38:32young boys so you're a teenage boy you get a reach out from someone who claims to be a friend of a friend she sends
605. 1:38:42compromising pictures of herself and invites you to send pictures back except it's not a girl those images are
606. 1:38:50enhanced and in fact it's some guy behind the keyboard board in Nigeria and then once they have the images they
607. 1:38:57threaten to Blackmail the kid and say if you don't pay up we're going to share these images will ruin your life
608. 1:39:04unfortunately this has real word consequences there have been at least 20 teenagers who committed suicide in these situations and it's just
609. 1:39:15heartbreaking do you think we have the laws or regulations in place now to keep Americans safe from all that's coming
610. 1:39:26all the AI threats I would probably leave legislating to the legislators but what I will tell you from an FBI
611. 1:39:33director's perspective um is that this is a type of technology that we see manifesting itself in more and more
612. 1:39:42situations more and more types of crimes more and more types of threats uh and there's a degree to which over not up to
613. 1:39:49speed that can be done without a with technology you sure yeah sure what that's a good point I mean listen what what AI does both good and
614. 1:40:01bad is while yes you can do all of what he just described without AI you have to have pretty significant technical skills
615. 1:40:12to do it with AI it democra so AI democratizes all skills right and it's going to increasingly democ democratize in you
616. 1:40:24know increasingly sophisticated skills are going to get democratized so everyday people can now do things he he
617. 1:40:31talked about like going from junior varsity to Varsity that's the thing that's happening with AI is more and
618. 1:40:37more people are going to be able to do this stuff and the elite bad guys are doing what so the elite bad guys are the ones
619. 1:40:51you know for example conducting the most phisticated cyber intrusions to me no country no country represents a broader more severe more
620. 1:41:03comprehensive threat to American innovation American ideas so this stuff they're showing here like this is this
621. 1:41:11is relatively old convolutional Network Technology so they're showing China here they're basically saying China's using
622. 1:41:18AI to watch all of their people and we're going to be able to do that here we are already doing that here right
623. 1:41:26like this is old technology what things like segment anything from meta does is it makes that technology available to
624. 1:41:34everyone else you can Rel with relatively you know trivial amount of programming skills and now with something like repet um with their
625. 1:41:43autonomous agent where you can create an application like I did last week without any programming skills individuals are
626. 1:41:50going to be able to do this this kind of sophisticated you know person tracking and identification and things like
627. 1:41:58that American ideas to our Economic Security and ultimately our national security than the Chinese government China has a bigger hacking
628. 1:42:08program already than that of every major Nation combined and has stolen more of American's personal and corporate data
629. 1:42:19than every nation our data is already gone it's long gone the scale of the threat is significant if you took the
630. 1:42:26FBI's program and just said forget Russia forget Iran just do nothing but China the Chinese government's hacking
631. 1:42:34program they would outnumber us 50 to1 does that mean the race that we all feel that we're in now the race to be
632. 1:42:43first to be better is to our advantage I do think we need to win the race I think we are ahead in the race
633. 1:42:53but I would also say that we shouldn't in the course of trying to win the race disregard the importance of building in
634. 1:43:01safeguards and controls and that sort of thing to help protect abuse so speed yes recklessness no we've got an election coming do you
635. 1:43:13think this election cycle will be compromised by disinformation I fully expect to see um disinformation at by Foreign
636. 1:43:29adversaries the the 50x behind thing was just on the on the hacking um so I think he's saying we're ahead in the in the
637. 1:43:37development of the core technology but I don't know and I you know if you look at what China's releasing with their video
638. 1:43:44generation tools there's there's three major video generation tools from China right now that are at least in the
639. 1:43:50neighborhood of Sora so so I don't think we actually know where we stand relative to China right now finding all too often that something
640. 1:44:01on social media that looks like Bill from Topeka or Mary from Dayton is actually you know some Russian or
641. 1:44:08Chinese intelligence officer on the outskirts of Beijing or Moscow what should we be on the lookout for all
642. 1:44:16Americans should try to be you know more Discerning consumers of information it is incumbent on everyone in America to
643. 1:44:24bring an intensified name the video AI China has they've got cling there's one that came out last week and then there's
644. 1:44:32um cing is a really good one cing uh it's cing clling ai.com I think or king. just look up cing that's the one that's
645. 1:44:41farthest along um but then one came out last week and then another one came out like two weeks ago or three weeks ago so
646. 1:44:47there there's a number of them out there now um some of them you can't sign up with if you don't have a Chinese um
647. 1:44:53phone um cing made it possible I think it's clling ai.com you can sign up with an American email for that one then you can use it
648. 1:45:06caution to uh the use of AI and how AI can be used by bad guys against all of us but not to panic I don't think this is a time for
649. 1:45:17panic let's talk for 15 minutes about all the reasons you should panic and then we'll have Christopher say this isn't a time for
650. 1:45:26panic that's radiation or have system ah TV isn't media the best they're watching you they're coming
651. 1:45:41to get you China's better than us except we're better than China because America's number one uh but but you know you know what you
652. 1:45:50shouldn't do you shouldn't worry about the robots coming to kill you just cuz we told you the robots are coming to
653. 1:45:55kill you why would you do that don't you worry your pretty little head go get some more Fritos no point in panicking just
654. 1:46:08panic I don't think that FBI guy knows completely what he's talking about he sounded special um you got to imagine that we're
655. 1:46:16hacking them too like again all that stuff has been there forever AI is going to take a lot of people's jobs I like
656. 1:46:22making analogies of the Industrial Revolution we found other places to specialize I have a robot in my Pharmacy
657. 1:46:29that fills drugs all day long for me I did not like it when it first got there but I couldn't imagine being without it
658. 1:46:35now I hope that AI will be able to change Health Care in a way that you I bet they go to Bill Gates right now so
659. 1:46:43that that information would go from the doctor to the patient uh in a more efficient manner I think that made me Shi to nursing school
660. 1:46:53because basically I'm always going to be needed I'm trying to Future proof myself if am I going to be successful is a real
661. 1:47:04question AI in the future of us for health education and jobs Bill Gates so Bill Gates you've been on this Tech rocket ship for quite
662. 1:47:15some time in the 70s this is the it's bigger than anything it's going to be okay the positive outweighs the good I
663. 1:47:21bet the internet and now I'm just wondering how you think this AI moment Stacks up to where we've been to where we are right
664. 1:47:30now ai is going to have more impact than even the personal computer the phone the internet right this what I said right
665. 1:47:39this is what he's known for in this phenomenal way but this is the biggest technical advance in my lifetime what does
666. 1:47:47the I mean that's important for people of AI in 10 years I'd Sayed ucation and healthare will have changed dramatically
667. 1:47:56y so in medicine there's the research part the speed of scientific inovation across all diseases even the neglected
668. 1:48:04ones like malaran and HIV that will be accelerated but perhaps even more profoundly whenever this guy's stolen
669. 1:48:13everything that made him what he is he's no Visionary yeah but but he was a guy that figured out how to take all that
670. 1:48:18stuff that other people created and scale it open I a lot of what they're doing is taking things like the transformer technology
671. 1:48:30and they figured out how to scale it better than anyone else so just because he didn't invent anything doesn't mean
672. 1:48:38he didn't innovate what he innovated in was leveraging that technology that had been there and scaling it right Steve
673. 1:48:46Jobs You could argue was the inventor and and innovated Bill Gates was taking things and scaling it I would argue that
674. 1:48:54open AI right now is taking technology and scaling it I think I actually got that idea from Pate who I just saw posted things will change
675. 1:49:03Wow E easy there with the wild speculations you almost can think of the AI as like a third person sitting in
676. 1:49:13doing the transcript suggesting the prescription and then you have this whole session has been transcribed and
677. 1:49:19so instead of the doctor facing that computer screen they're engaging with you in this software is making sure there's a really good
678. 1:49:28transcript one of the things it's exciting to me is how this is going to impact kids and their personal learning
679. 1:49:37yeah yeah the dream that you could have a tutor who's always available to you and understands how to motivate you what
680. 1:49:43your level of knowledge is this software should give us that you know when you're outside the classroom that personal
681. 1:49:51tutor is is helping you out encouraging you and then your teacher you know talks to the personal tutor and says okay
682. 1:49:58what's your advice about how I do things in class the next day they won't have to spend time grading homework they'll be
683. 1:50:04immediately know how that went um oh my God they're going to have so many mistakes can't replace a nurse or ma Medical Associate I assume
684. 1:50:14um do not this is just my opinion do not assume that the inaccuracy of the models of today are going to persist for that
685. 1:50:24much longer what open AI released today is a version of its of its language model that now thinks before it answers
686. 1:50:36these things are going to get smarter and smarter and smarter so the current technology I would agree with you but I
687. 1:50:41don't think that lasts for that long we know how that went does this mean you eventually don't have teachers
688. 1:50:49and everybody just has their own AI teacher we're trying to help teachers uh you know that personal engagement motivating
689. 1:50:58a kid uh meeting them where they are you know the AI in no way uh is a replacement for that but the fact that
690. 1:51:06as the teacher saying okay what should you work on it's working at your level that personalized notion you know we can
691. 1:51:14see it's already working pretty well I talked to Sam Alman and he was pretty direct about ai's impact on the work
692. 1:51:22force a lot of jobs may go away do you agree and how should we start preparing ourselves to navigate that yeah
693. 1:51:31particularly as you get in this not the 5 to 10e time frame but beyond that or the AIS not only are they very good at
694. 1:51:40kind of White Collar things but also the robotics embodiment is taking a lot of blue collar type things whether it's
695. 1:51:49warehouse work or factory work 3D printed house rooms cooking things like that and so that's not only just a
696. 1:51:57threat to people's livelihoods but you know people gain dignity and a sense of purpose and meaning from work that's
697. 1:52:04right we've organized our society around the fact that if you you know get a good education if you get a unique set of
698. 1:52:11skills that you'll be rewarded for that if we get 20 or 30% more productive that's all to the good because there's
699. 1:52:18so many unmet needs but if you get you know 300% more productive yeah in a way and say well that's only good because
700. 1:52:27then we can make more stuff and we can have more leisure time but the the way that we organize it and maintain
701. 1:52:34people's dignity will require a lot of Engagement and you know picking leaders who are attuned to taking that access
702. 1:52:44and and making sure we're all sharing fairly people this is a big deal the the the thing that I think is going to be
703. 1:52:49the biggest transition for people is is not necessarily the loss of jobs I feel like we're we're going to find a way to
704. 1:52:56navigate that I think that that concept of of people having their self-worth tied to the things they do the skills
705. 1:53:05they have the [ __ ] they learned in college that they now have been doing for 20 years in their job and then all
706. 1:53:11of a sudden that goes away and they have to they they have to learn to do something else that's not where they
707. 1:53:17have their their core you know personal dignity and value that's the thing I think is going to be from a mental
708. 1:53:24health standpoint the the biggest transition I don't have a good answer for that one I think that's just going
709. 1:53:30to be time heard you speak be a lot of morning yeah in the long run you know as we get past that 10 years work won't be
710. 1:53:38as important and you know that uh you know I think is work won't be as important thing yeah what are the jobs
711. 1:53:46yeah it's that feels right to me that it's we're we're in the middle of a 10 we're at the beginning of a 10e transition
712. 1:53:55where the value of work is completely redefined completely redefined yeah people people will find their creativity again that that the
713. 1:54:08value of creativity and empathy and personal connection I think that goes up in value and that maybe the job of of
714. 1:54:16humans becomes to engage with one another and let the machines do all the shitty work that's that's my hope that you believe are going to be
715. 1:54:26undisturbed what skills are going to persist in these in this new world yeah that's very hard to make a prediction I
716. 1:54:34mean be aspects of creativity there'll be aspects of appreciating people you know social work the teacher really engaging that student
717. 1:54:44in the rich way even that I can't guarantee that software won't eventually get good at that but you know for the Next Generation
718. 1:54:53the more human the more engaged you are with other people that is a a skill set that is in such short supply whether
719. 1:55:02it's me the more engaged you are with other people right I think that this is why joining the salon joining
720. 1:55:10communities of people who are engaged and generous and empathetic and curious and adaptable all that sort of stuff
721. 1:55:22feels to me like it goes up dramatically in value but that's the the change is is going to be profound for a lot of people
722. 1:55:32that where they they put their value in their skills and their work um there's going to be a lot of loss and
723. 1:55:40mourning in such short supply whether it's mental health or education uh that we can never have must soul I want to join but I'm
724. 1:55:50weird we are weird friend do you have none do you have no I have significant fears about the risks the this is the
725. 1:55:58first technology that is happening faster than even the Insiders expected it's happening faster than you guys
726. 1:56:06thought absolutely do you think that your company uh that that Microsoft and other companies that are going to make a
727. 1:56:14lot of money off of AI have an obligation to help us manage and navigate through this change absolutely the fact that there will have
728. 1:56:24to be regulation that the way we think about taxes will have to change quite s Papa we're not weird we're irregular
729. 1:56:32with the government then at least we can craft something that's that's not just profit driven but what did we miss with
730. 1:56:39the internet and social media that we can use to apply now to this moment of AI well I I would say speaking for
731. 1:56:50myself personally and I think many others that we were a bit naive yeah so Sandy Hook is a synthetic completely
732. 1:57:00fake we thought the internet the availability of information would make us all a lot more factual the fact that
733. 1:57:06people would seek out kind of a niche of misinformation we were a bit naive and and bringing those lessons of you know
734. 1:57:13let's get ahead of this let's get the politicians to be users and and have a dialogue about uh maximizing the
735. 1:57:22benefits while uh minimizing the negatives that's that's actually really interesting the the um if the value of the past 10 years was
736. 1:57:33to teach us how susceptible we are to misinformation and media manipulation um maybe it does set us up to be a bit
737. 1:57:44more Savvy I I mean I think as scary as it was to hear from the woman that was like you know my daughter called me and
738. 1:57:50told me you know she I need $900 as scary as that is I think creating an awareness that okay we've seen what we can do to each other with
739. 1:58:00just you know basic media manipulation over the past decade um and now we've got tools that make that way more
740. 1:58:07sophisticated I mean may maybe that's at least prepared us to be ready for what's coming and the second root word of it is
741. 1:58:19which is to be loved are we going to lose that aspect of human let me go back good question humans we have a soul
742. 1:58:29we SPS book when it comes to your morning news we got you covered from Lis with your family BR you the news in
743. 1:58:36Colorado that matters most we're out here in the community telling your stories and making a difference start
744. 1:58:41your day the right way get Denver 7 Morning News weekdays starting at 4:30 yeah awareness is really important
745. 1:58:48right now y I'm broke my family thinks I'm a l love life is is this being streamed on YouTube what's the name of this program with
746. 1:59:01Bill and Company Bill and Oprah it's Oprah AI in the future of us on ABC human is to be boy that's a good question humans we
747. 1:59:12have a soul we went through different levels of life from newborns toddlers children teenagers human in Arabic is
748. 1:59:20insan and it translates to two things one is forgetfulness and that's a quality that all of us share and the
749. 1:59:26second root word of it is UN which is to be loved are we going to lose that aspect of humanity I think that gets
750. 1:59:32Amplified I think that as humans there is nothing that brings us greater discomfort than lack of knowledge and I
751. 1:59:40think that AI in many ways comes out of that drive know all the answers to be in control one of the things that I reflect
752. 1:59:50on is be careful about Comfort because too much comfort is actually going to make your life worse not better because
753. 1:59:57it'll decrease your overall capacity this this is a Wellcraft that I locate the Divine in relationships relationships between
754. 2:00:05humans and the environment to me that's God and that's the Divine and that's something that technology can never
755. 2:00:13stand in for and can never replace totally agree with that to that's that's really good messaging and I think
756. 2:00:20counterintuitively that's where AI pushes us I am so delighted to be able to talk to Marilyn Robinson one of
757. 2:00:28America's greatest novelist who writes of all the messiness and all the grace of Being Human so what do you make of
758. 2:00:38this moment that we're in and these two words artificial intelligence well you know uh it's a certainly a departure
759. 2:00:47from any sense of intelligence we have had up to this point you know mulling cumulative knowing and interpreting the world that has been
760. 2:00:57called intelligence because we've always thought of them as uniquely human traits exactly the impulse behind it seems to
761. 2:01:04be to eliminate the human hand the human eye in the making of the reality that we inhabit to to the farthest extent
762. 2:01:14possible with the effect of of U concentrating power in the sense of the control of this strange fantasy uh within very few people's
763. 2:01:25hands and consolidating masses of capital and masses of Natural Resources devoted to this very dubious project
764. 2:01:34it's just a a human impulse to escape from its Humanity which is something that happens over and over again
765. 2:01:42historically the impulse to say uh that there is nothing intrinsic about a human person that cannot in theory be
766. 2:01:52replicated by something that frankly Works cheaper Works faster I don't know why it is but it's part of our human lot
767. 2:02:01to be uncomfortable with our humanity and to mechanize it you know out of existence in some [Music] way we're not enough in love with our
768. 2:02:15own existence and we're not enough in love with everyone else's existence and I think that's very d [Music] dangerous I think that's just so
769. 2:02:26powerful we're not I vehemently disagree with it by the way no surprise there but hasn't that been the nature of who we
770. 2:02:35are as human beings always seeking the next the more the better the other we generalize from the behavior of
771. 2:02:44people who tend to dominate historical narrative and are very frightening people in many many cases not the kind
772. 2:02:52of people we would have wished to be making great decisions what is the thing that concerns you the most about
773. 2:03:01AI I'm afraid that there will be so much investment of every kind into this project that it will have it this project that
774. 2:03:13like her idea that there's this there's this single entity that's doing this is just bizarre to me it it's like this
775. 2:03:21project is not a project right there's a technology that was invented in 2017 that millions of developers are pursuing
776. 2:03:31and a handful of companies are exploiting better than others but fascinating that it will have its consequences before it really deserves
777. 2:03:43them and so there will be all kinds of cutting back on all kinds of employment and so on anticipating this insurgence
778. 2:03:53of of AI and then it will turn out that it's a it's a bad project you know that it's more dangerous than it is valuable
779. 2:04:02and so on and so we will have a a disrupted economy on one hand and an unusable technology on the other yeah
780. 2:04:11that's horseshit it it like like what what she's basically saying is we'll realize oh it was a big mistake and and
781. 2:04:20the way we used to do things things doesn't exist anymore and now we just have this failed experiment she
782. 2:04:25completely is missing the fact that we as humans will evolve right as as these tools change things we will step into a different
783. 2:04:35role right yeah Source Camp she's clearly never used generative AI you guys she represents hundreds of
784. 2:04:41thousands of people be nice no listen I I don't this is not about nice or not nice this is this is about so well whatever let's let's just
785. 2:04:52keep listening to her cuz she she's clearly very smart and articulate and thoughtful about this but she's also got a very limited view
786. 2:05:04of what AI is and its role but it's also going to no question about it it's going to make life a lot easier for a lot of people so what's
787. 2:05:18wrong with that American schools as for a lot of written work from students and that's a very good thing that's an autonomy
788. 2:05:26creating exercise you know but what we don't do we we don't tell them that this is a discipline like an athletic
789. 2:05:35discipline that they have undertaken to strengthen themselves not just to bring themselves over some arbitrary line of
790. 2:05:42sufficiency that it's the discipline of doing the work and the thinking process exactly and students say writing is
791. 2:05:49difficult which of course it is it is but difficulty is the point and so if you were in charge of this whole AI
792. 2:05:58movement what would you be doing she turn it off I would advise that it be used for you know the kind of research that
793. 2:06:08produces uh drugs and immunizations and so on I would uh make a fantastic Library well that's what it is it's the
794. 2:06:20world's greatest Library instantly accessible very very judiciously to replace work that is drudgery without
795. 2:06:29the arrogance of replacing the work of people who you know who are of very great value you know just because they
796. 2:06:38are the substance of our ordinary interactions with the world but you know I also age as I'm growing older and I
797. 2:06:45think as you grow older there's always the risk of trying to hold on to what was and when there are currently now children
798. 2:06:58who are born and who will be born who will know nothing but this generation of AI so there will be no looking back this
799. 2:07:07hour what used to be or the or the way it was and that's always been the case but I I get the sense that you somehow
800. 2:07:15feel that this moment is different than others in past Generations why well you know the the thing that occurs
801. 2:07:25to me to compare with it is the Manhattan Project that was carried on among very few people in deep secrecy
802. 2:07:34these things are talked about afterward as if they're Collective decisions but they're not they're scientific
803. 2:07:40thresholds that are crossed only by people yeah this is this makes me nuts right the comparison to the Manhattan
804. 2:07:47Project is right that was a handful of people creating you know a machine right now you got millions of people
805. 2:07:54developing these Technologies The open- Source Community I mean one of the things that meta did here by open
806. 2:08:00sourcing their powerful models is that they're giving access to these tools to to lots of different people
807. 2:08:08the the democratization of access to this technology as it's being developed is completely different than the
808. 2:08:15Manhattan Project was which was just designed as a weapon of Destruction right she's only talking here about the
809. 2:08:32downside of this not the the the humanity impact of it my experience with with these tools is that the benefit I get from them is as is as
810. 2:08:48if I'm tapping into the collective intelligence of humanity not the artificial intelligence of a machine
811. 2:08:55that's here to replace us that that this is giving us access she said it she said this you know she wants us to build the
812. 2:09:02ultimate Library that's what large language models [ __ ] are they're the world's greatest Library where we get to
813. 2:09:09tap into the collective output of humanity so um this particular point of view makes me a little cranky CU it's it's just not the same at
814. 2:09:20all sequence and there's no going back from nuclear fishing but uh that doesn't mean we should put ourselves in the same kind of position
815. 2:09:31again yeah that that was a very lopsided segment didn't like that at all cond over and only one was named the September
816. 2:09:487 did she ask to speak to ai's manager yet it's very very CL thank you to everyone who spoke with us tonight I
817. 2:09:55hope that they have left you with a clearer picture of what is coming artificial intelligence is still beyond
818. 2:10:03our control and to a great extent our understanding but it is here we're going to be living with technology that can be
819. 2:10:10our Ally as well as our rival it is something no other generation has ever experienced this moment requires a
820. 2:10:22different level of alertness awareness about who we are and where we're going what's real and what's not what's
821. 2:10:32artificial intelligence and what is human wisdom authentic intelligence earned through study and experience we
822. 2:10:40are this planet's most adaptable creatures we will adapt again but keep your eyes on what's real the stakes for all of us could not be
823. 2:10:52higher don't disagree with that the authenticity is a big one here's here's where I see this special like parts of this special are
824. 2:11:01fine parts of this special are not going to age well here's what this feels like to me
825. 2:11:19um um this is what this feels like 56 pass I wasn't prepared to translate that as H doing that little tease that little Mark
826. 2:11:33with the A and then the ring around it at see that's what I said MH um Katie said she thought it was about yeah oh
827. 2:11:41but I'd never heard it i' never heard it said I'd always seen the mark but never heard it said and then it sounded stupid
828. 2:11:46when I said it violence at NBC the lunch there it is violence at NBC geom GE Allison should know what internet
829. 2:11:59anyway what is internet anyway internet is uh that massive computer network the one that's becoming really big now what
830. 2:12:08do you mean that's how does one what do you write to it like mail no a lot of people use it and communicate I guess
831. 2:12:13they can communicate with NBC writers and producers Allison can you explain what internet is no she can't say
832. 2:12:19anything in 10 seconds or less oh Alison will be in the studio shortly Frank such a dick computer network made up made up
833. 2:12:27of uh started from Oh I thought you were going to tell us what this was the it's not an it's it's it's a computer
834. 2:12:34billboard but it's Nationwide and it's it's several uh universities and everything all joined together right and
835. 2:12:39others can access it right and it's getting bigger and bigger all the time it came in really handy during the Quake
836. 2:12:44a lot of people that's how they were communicating out to tell family and loved ones they were okay because all
837. 2:12:48the phone lines were down I mean that Oprah special that that that what this feels like I think we'll look back on
838. 2:12:52the over special at some point like it it didn't have any really egregious things in it but it but it was
839. 2:13:00definitely with the exception of Sam Alman a little bit of the Bill Gates um that was definitely um it it was definitely
840. 2:13:16reinforcing the Hollywood the robots are going to kill us Ze gist right so you know um I don't think that special did AI any favors and
841. 2:13:30I don't think it [ __ ] matters I you know again the whole purpose of this channel is regardless of whether you
842. 2:13:38know that woman feels like you know AI is a project that is designed to replace people or not doesn't change the fact
843. 2:13:50that this shit's coming and it's coming fast right my push back on the woman talking about the humanity thing is that I don't
844. 2:14:00think she's used generative AI like I don't think you can use it and not feel a profound sense of tapping into humanity
845. 2:14:15and this community the community that hangs out here a reminder of how important our empathy is that's that's it
846. 2:14:22an my my experience with generative AI is I feel more profoundly connected to People by a lot than I ever did with social media social media
847. 2:14:36cut you know it's an oxymoron right because it makes us antisocial I feel like when you tap into these tools and you feel that profound
848. 2:14:48sense of like holy [ __ ] the fact that I can say write me a computer program and it writes a computer program that is the knowledge of
849. 2:14:58millions of people that have contributed code to the internet for years and years and years that makes that possible like there's a
850. 2:15:11profound for me there's a profound realization of the value of humanity when I use these tools not the opposite
851. 2:15:22of that that was my my the thing I had the most trouble with I mean okay fine Oprah didn't get it and Oprah's got a
852. 2:15:30bit of a point of view on it that's like that's fine and that that I actually expected that woman in particular like she's
853. 2:15:41missing the magic of AI is that it is the world's best Library already it is a reflection of humanity already it's not there to replace us
854. 2:15:58right it is part of us it is collective intelligence not artificial intelligence but whatever um I bet Oprah's going to
855. 2:16:07launch something with AI classes more specials she has a profit somehow well I listen I think Oprah's smart Oprah is like this is going to
856. 2:16:18change everything I like I at least want to have a voice voice in the conversation I mean I don't like I don't even with the woman I
857. 2:16:28disagree with I don't hold it against her that she hasn't used this stuff I don't hold it against Oprah that what
858. 2:16:36she's that what she framed up was this shit's scary CU it is scary and the more you use it the less scary it becomes right so I see just
859. 2:16:51like the Katie KK piece here I see this as one of those things that's looking at AI through a lens of not having used it
860. 2:17:01that's where everyone starts and so the only the only information they have is what have people said about it well
861. 2:17:08they've said that the robots are going to kill us oh so I should be scared afraid of it oh [ __ ] they deep faked
862. 2:17:14someone I should be afraid of it right there's there's all the reasons to be afraid of it the Brandon tid in a couple of hours making a
863. 2:17:25picture book for his kid and being able to publish it with his cute little Pixar kind of version of himself they're not talking about
864. 2:17:38that that doesn't get people to watch TV but that changes lives right so like did did this did this show advance
865. 2:17:56advance the cause at all I don't know people over 60 are the best source of knowledge on AI we all know this yeah exactly pay I you know
866. 2:18:08exactly um AI does hallucinate information for now it does but so do humans so do humans humans get [ __ ] wrong humans hallucinate
867. 2:18:22humans don't apply logic to the situation right I mean the past decade has been that just people sort of
868. 2:18:32getting in camps and going you suck no you suck no you suck well here's facts yeah we have alternative facts
869. 2:18:41like humans are worse hallucinators than ai ai is at least getting better um nobody slowing down I agree with
870. 2:18:53that AI is our mind already I have developed another part of my brain than uh that talks AI thinks AI yeah AI is part of our
871. 2:19:04brain there's another Oprah special on AI on 916 is it a repeat or a new program what's today the 12th if it's on the 16th I doubt it's
872. 2:19:22unless it's a follow on where they're going to announce some some of the new [ __ ] that open AI announced I doubt
873. 2:19:31it I I I I don't know I don't I don't know the answer to that Amelio's wife we humans are AI we were created well we'll
874. 2:19:47see are we in a simulation I you know I personally don't think so but you know it's possible I found a free AI from Opa
875. 2:19:56under my seat anyone else you get AI and you get AI yeah you know the the the thing if if I were to sort of criticize
876. 2:20:04one thing about that that I feel like was missed in that special was let's hear from someone who's used it that that did something
877. 2:20:12really remarkable with it like we've the the salon and this channel is full of people and and some of the victories are
878. 2:20:22very very small like like the kids book and what Jim Ross is doing with his business and what source Camp is doing
879. 2:20:28you know educating people and Anne Murphy in fundraising like I I there are so many there are so many things that could
880. 2:20:38have rooted it in reality as scary as it is here's someone that did something you know amazing or cool or powerful um but
881. 2:20:48listen again when when this Katie kurick thing came out so this was 1994 so I had just started Urban desires
882. 2:20:59I just started agency.com and the media was just like what is this internet thing who are these weirdos right like it
883. 2:21:10was people were discombobulated and most people hadn't heard of it they weren't paying attention to it and so I I think
884. 2:21:18we're at a very natural state that that that's special to me is further Evidence so there's there's a there's kind of
885. 2:21:30a um a Trope going around right now that AI the AI hype has peaked and and we're coming down now and it's all over and
886. 2:21:40these tools have plateaued and they're never going to get better um and and they equated where we are with AI to the internet bubble of of
887. 2:21:522000 where right where the it the hype got so big and then everything burst I've argued and this this special with
888. 2:21:59Oprah reinforces it for me the hype cycle hasn't even begun yet not enough people know about AI not enough people
889. 2:22:07are using it the technology is still too janky I think we're probably two years out from where the real hype begins and
890. 2:22:16it really gets blown out and Amplified I think that happens two years from now um and I think I there probably will be
891. 2:22:27a hype cycle where it it gets so overblown and then bursts but I think we're not even close to it I think we're
892. 2:22:32closer to 1994 metaphorically than we are to 1999 Tyler Perry just invested millions in a movie producing use using AI that
893. 2:22:42that's it oh did he that's interesting so Tyler Perry also cancelled an $800 million Studio expansion in Atlanta
894. 2:22:51because of Sora so that's interesting that he's investing in it he's like well [ __ ] it I was going to spend $800
895. 2:22:56million to build a studio why don't I spend aund million and make an AI movie I think that's pretty smart
896. 2:23:07actually for people who have never used AI how will they respond to this show prediction be scared I agree with that Dan I mean yeah the the
897. 2:23:22what what will happen from that show is that more people will not just hear AI they'll hear chat GPT or they'll
898. 2:23:33like you know they they making videos or they got voice things like there may be an awareness of it I don't think that
899. 2:23:42special inspires anyone to go get curious about AI again the reason to be part of you know Rocky Mountain interest
900. 2:23:49group or the salon or communities that are curious about this stuff right now is that we're still in a pocketed
901. 2:23:59minority of in the corners of society so so we have time you're not as far behind as you think and we're not as far ahead as we
902. 2:24:10think we're very early still would you call it underhyped Kyle in fact I would if you look in the corner I've got a video series that I
903. 2:24:20did that if you want to buy it it supports the channel called underhyped where these are videos where I talk
904. 2:24:26about why this stuff is significant I seriously wonder if Oprah has any investments in AI she probably does um
905. 2:24:33she well although I would say if that special was to Hype up AI Investments that she has it would have been a very
906. 2:24:42different special so I would assume that she might have some Investments but but that special did not feel like it was hyping AI
907. 2:24:51that special felt like it was you know a a continued um a continued part of the narrative that AI is scary and you
908. 2:24:59should be scared of it you make Mone not like that um let me take a quick break I want to come back I want
909. 2:25:07to show some things so we're going to play with the new so open AI did launch I think what's project strawberry
910. 2:25:17today they launched a new model um that's it's in chat GPT right now at least it is for me they've been rolling
911. 2:25:23it out all day um and it's interesting it's not multimodal it's it's a very limited version of of what it is but
912. 2:25:32it's pretty fascinating so we'll look at that I want to show you this new feature in Notebook LM that I think is
913. 2:25:39remarkable I've been playing with that a lot today um and then Brandon had hang on
914. 2:25:57um oh there's a new Hume voice model out and then Gemini live is now available to everyone I don't even know what Gemini live is let
915. 2:26:08me go get another drink take a break I'll be right back entertain yourselves oh please
916. 2:26:49e
917. 2:27:19e for
918. 2:28:44yeah I got to three things
919. 2:29:05all right all right good people have we got anyone in here still yes while I was waiting I had GPT roast
920. 2:29:14a picture of Kyle's workplace so funny nice I like it um that's awesome beautiful and Murphy all right relevant to the
921. 2:29:25conversation let's see what's going yay Kyle is back he did it okay um my back is holding out my back has been shitty all day but to
922. 2:29:40right now it's okay so um if it goes south I'll let you know but let's go play with some [ __ ] going to post it and show and tell yeah
923. 2:29:52please do that um okay so couple of things if you're new here welcome um that's not what I wanted that's not what I wanted that is what I
924. 2:30:03wanted the salon the salon. go there Kyle's all hyped up and his back is feeling good yeah exactly lumbar support no I think
925. 2:30:17it's what it was was it was Friday night I did a live and I you know how when I get in my Rants and I just go for like an
926. 2:30:30hour and a half or two I did one of those and I think I was sitting in the exact same position for more than an
927. 2:30:38hour and then when I stood up like my ass muscles got really tight and my and it just pulled something in my lowered
928. 2:30:44back um and just pulled something out of whack and it's been [ __ ] up ever since so I think so long as I keep reminding myself to change
929. 2:30:54positions um I'll be okay Gemini live is Google's version of advanced voice oh interesting oh it's oh it's
930. 2:31:04available for everyone is it available for everyone is there a web version of it it was the cartwheels yeah exactly
931. 2:31:11you know what it was Silver Fox it was the twerking if you weren't here Friday night so I've been promising for for
932. 2:31:18months that I was going to uh Donna manini and do some twerking on this channel as it deserves it is Tik Tock
933. 2:31:26and uh so so I was twerking in the manini on Friday and just you know threw the back out boy do I feel old not as
934. 2:31:37old as you looked in that man Keeny all right ah please no Dan all right if you've just walked into the channel yes this is normal
935. 2:31:52conversation um anyway the salon so the salon is a community that I co-founded with Leah faston who's a
936. 2:32:00photographer out of Boston we founded it in the week after chat GPT came out um December 7th 2022 was our first meeting
937. 2:32:11and we've been doing a meeting every other week for since then and we've got an online community so if you go to the
938. 2:32:17salon AI it's a sitee you can read about what we're about um and then click on join our community and that'll take you
939. 2:32:25to a site that is the community and there's a nice little onboarding there's like a little seven step you know
940. 2:32:31welcome just follow those steps um and uh and it's it's a really remarkable community of AI optimists people that
941. 2:32:39are trying this stuff so if you're curious about Ai and you haven't really done much with it and you want to be
942. 2:32:46around people who have and and are not douchy about it that's this community it's a very generous empathetic
943. 2:32:56supportive enthusiastic optimistic Brave group of people and especially if you're one of those people that's really excited about
944. 2:33:08AI but everyone around you is like could you shut the [ __ ] up about stupid [ __ ] AI Janice we don't care you know
945. 2:33:20it's it's jalapeno poer night at TGI Fridays and we just want to enjoy our our icy margaritas and our jalapeno
946. 2:33:31poppers all you can eat jalapeno popper night is what I live for and your AI is just H we're tired of it we're tired of
947. 2:33:41hearing about how excited you are about that stupid Ai and those robots the robots are going to kill us Janice so enough with your AI
948. 2:33:53okay can I have another Margarita please thanks thanks Sandy that's great yeah no yeah it's delicious it's always
949. 2:34:01delicious hey and you know could you bring us some more poppers I think I'm I'm feeling I'm feeling radical tonight
950. 2:34:08let's get another let's get another round of Popp poppers on me everybody so if you're used to hearing that come to the salon
951. 2:34:19okay boycott Oprah Oprah's fine um Oprah was Oprah was banned from the community I listen I I don't here's the thing about that
952. 2:34:30special I don't have any ill will for that special um I think that what that special did is so here's something I heard when I was in
953. 2:34:44DC um one of the staffers said to be quite honest AI was was the topic of conversation about eight months ago and
954. 2:34:58everyone was talking about it and everyone was focused on it and that's gone away that sense of urgency is gone
955. 2:35:05away so if the Oprah special just introduced the topic to more people and more people like hey what are we doing
956. 2:35:13about this AI hey what is this AI some percentage of those people are going to get curious and they're going to try it
957. 2:35:20and they're going to have their first Kevin mallister moment oh my God it can do that and they'll get excited about it so
958. 2:35:31I don't think it's a bad thing I don't you know people are where they are and some of what AI represents is scary and
959. 2:35:39a lot of what AI represents is very very very very very hard to communicate without the other person actually
960. 2:35:54experiencing it I can't I can't tell you how viscerally satisfying it feels to take a screenplay that I wrote 30 years ago and
961. 2:36:06have chat GPT understand it tell me who it would cast in the roles and generate frames from the movie that I had in my
962. 2:36:14head 30 years ago I Can't Describe to someone how Lally satisfying that is they have to experience their version
963. 2:36:25of that on their own before they actually understand what the [ __ ] is going on and so what the salon is about is our our we we've got a a
964. 2:36:36three-step journey to AI literacy and the first step in fact Let Me Go show you the page
965. 2:36:53the first step of AI literacy where we say to start is play first play play first it's funny when when I got um invited to have the salon
966. 2:37:09participate in some of these Washington events um this word this phrase play first through through the people that we were
967. 2:37:21talking with they were like let's not that's not serious enough AI is really serious these are you're talking to politicians this is serious
968. 2:37:29business I actually think that playing with AI is like I am more convinced than ever this is very serious playing with
969. 2:37:41AI experimenting with it do like Source Camp does and and do parlor tricks go make yourself write yourself a silly song with sun
970. 2:37:50um go clone your voice like Mar Mars brownley did and and have yourself say silly things in that voice play with it
971. 2:38:04because if your only experience with it is oh I'm afraid of it and now I have to learn AI that feels [ __ ]
972. 2:38:14daunting it feels like the thing that's the the farthest from reality if if you don't know what's going on with AI and
973. 2:38:21you see all these crazy people going look at the picture I made look at the voice I made look at the video I
974. 2:38:27made your current experience with how you produce content like that is you've got to learn Final Cut Pro and you've
975. 2:38:34got to learn cinematography and you got to learn this and you got to learn that you can't possibly know sitting from the
976. 2:38:41outside what's actually going on and when you type in the words make a video of of a cute red puppy you know
977. 2:38:51running down the beach you know with a little girl in a yellow dress and then it does that and you go oh my God I had no idea
978. 2:39:02you can only get to that through playing just play and in playing you'll figure out where the boundaries are you'll figure
979. 2:39:11out what you can now do that you didn't imagine was possible and then you can move on to the next phase of the AI literacy Journey
980. 2:39:22which is mindfully create now that you know a few tools and that you can do some stuff now you go oh maybe I should
981. 2:39:29make a book maybe I should make an art project maybe I should solve that problem at work maybe I should start
982. 2:39:34that business of I've always thought of and then the generously lead part is tell people about it Learn out Loud hey
983. 2:39:44I was using Ai and I thought I could start this business and I went to start this business but then the AI sucked but
984. 2:39:50then someone in the AI Salon told me that I could use this other tool and then I tried that and it sucked less and
985. 2:39:55so now I've started this business now I don't know what to do and bang all of a sudden you are an AI thought
986. 2:40:04leader where maybe two weeks ago you were just like huh what's this AI thing that's crazy so join this community all
987. 2:40:15right could have had Jansen H and how Nvidia is leading the US and stock markets higher yeah there could have
988. 2:40:21been there's a lot of opportunities where they could have done some more positive stuff there but that's okay it
989. 2:40:26was only an hour I mean the problem with AI is it's so [ __ ] vast I mean it's it's one of the things I struggle with
990. 2:40:32on this channel is like it's so Vast at this point I like I don't know what to talk about anymore um all right we could have woken up to a
991. 2:40:44huge Uprising tomorrow that's all right I'm fine I'm fine without it totally agree on play equals learning that's
992. 2:40:49profound yep yeah it's like this is this is where it's at this is where it's at I th this little trium of of activities is
993. 2:41:00is super amazing and and you know quite frankly if you're if you're in the salon if you've been in the salon like this is who the people
994. 2:41:08are there's a playfulness there's a curiosity there's an adventure sense of adventure there's a sense of gratitude there's a sense of
995. 2:41:18empathy there's a sense of curiosity and bravery and just a willingness to go ah [ __ ] it maybe I should just reinvent myself all
996. 2:41:29right okay so let's go look at some stuff um so let me jump over to chat je generously lead is my fave yeah
997. 2:41:44generously leads important here's the thing about generously leading when you think about leadership you think about
998. 2:41:50okay I'm going to learn some stuff and I'm going to go teach people I'm going to lead the troops um that's
999. 2:41:58great another part of generously leading is asking for help when you need it and this is this is when I came to this
1000. 2:42:07Summer that it struck me that there's a lack of ego in the salon that allows people to ask for help hey stuck hey how did you do that an
1001. 2:42:20that that thing you did was really cool what did what tools did you use oh and then she'll tell you when you ask for
1002. 2:42:27help you actually provide someone else the opportunity to lead and that's leadership creating space for others to lead is
1003. 2:42:39leadership so not only being willing to generously tell people what you're learning and teach them how to do what
1004. 2:42:45you're doing but asking for help when you need it is as much leadership as the other side and that that was a big
1005. 2:42:51breakthrough for me okay so now when you go to chat [ __ ] you should see this GPT 40 right which is the the latest best
1006. 2:43:08for complex tasks then you should see 01 preview uses Advanced reasoning 01 mini faster at reasoning these two are the new model
1007. 2:43:20when someone says I'm behind they're not expecting to hear play more exactly exactly I'm behind go play [ __ ]
1008. 2:43:30it exactly th that's the thing Dan is I mean you know this as much as anyone like all of us are so far behind we will
1009. 2:43:45never catch up and and you can't catch up it's like like getting caught up on AI is kind of like uh you know I want to
1010. 2:43:58get caught up on every single detail of geopolitics globally right like you can't do it it there's too much it's
1011. 2:44:06moving too fast AI is the same thing where it has now surpassed our ability to keep up with it so given that well
1012. 2:44:17[ __ ] it have some fun with it play with it just see just pick a corner and just dick around with it and see what you discover
1013. 2:44:30because the other thing that happens when you play with AI is you will invariably have a breakthrough where you
1014. 2:44:38realize it can do something you've always wanted to do in your life but you you have written off as oh I'll never be able to do that
1015. 2:44:49yeah I dropped out of that drawing class in when I was 17 and I've always regretted not being a better drawer well now you
1016. 2:45:02can oh I didn't get into Berkeley and so now I'm I gave up on being a musician well now you can be a musician you're not behind but you'll
1017. 2:45:13never catch up Kyle Shannon turn turn that into a meme on the regulars Channel that's great all right and then you've got your
1018. 2:45:23old models are in here GPT 40 mini and the Legacy GPT 4 honest to God their naming their naming is such a [ __ ] disaster it's
1019. 2:45:36unbelievable it's so bad like how are you supposed to know the difference between this thing that cost them a lot of money but does really
1020. 2:45:46cool [ __ ] and this thing that's sort of like this thing but they've hobbled it because they haven't released all of it
1021. 2:45:52yet cuz the testing isn't ready and these things these things are completely different models I assume
1022. 2:46:00what these are is they're sitting on top of this engine just behaving differently but I don't know and then 01 mini is
1023. 2:46:08probably sitting on top of this model doing its reasoning thing this is sitting on top of this model doing this
1024. 2:46:13reasoning thing and this is a completely different thing no one knows that no one should know know that no one should care
1025. 2:46:18here but if they name them differently in and somewhat usefully then it would suck less okay so let me
1026. 2:46:31show you what I know so far I don't know much I think I've done two total prompts on this and in fact let me go look at
1027. 2:46:39the prompt that I did do yeah oh that that's interesting so when you do when you do a prompt with the 01 preview you got a blue dot beside
1028. 2:46:53it so they're now in they're now indicating that you've done something different so I said oh this was a this was a fun one
1029. 2:47:05let me show you what I did and then we we'll go play with it I said I want to take a classic movie and write a new movie that maintains the
1030. 2:47:16story structure number of characters Etc but completely comes up with A New Concept so give me 10 diverse Classic
1031. 2:47:24Movies first I'll pick one and then give me 10 concepts for the new derivatives so what it says here is it
1032. 2:47:33thought for 21 seconds but just like when you use code interpreter I think the O Oprah novelist L lady did the naming oh that's funny
1033. 2:47:45[Music] so just like when you use code interpreter it gives you that little drop down menu and you can see what it
1034. 2:47:54was doing and it's writing python code look what this is doing so if I open that up it's come up with let's see 1 2 3 4 5 6 seven steps
1035. 2:48:14and then it's it's written a prompt for each of the steps so step one weighing the concept I'm evaluating the users's
1036. 2:48:20request to generate a list so step one evaluate step two brainstorming movie options I'm listing 10 uh that provide a
1037. 2:48:28choice of options ensuring that they're factual and within copyright rules interesting generating classic film
1038. 2:48:36ideas now I'm creating a a list of 10 movies mindful of copyright rules and avoiding certain inappropriate themes
1039. 2:48:43right so it's actually telling you what it's doing um I'm thinking through a of classic films spanning genres periods
1040. 2:48:51and cultures like Casablanca and Seven Samurai this approach highlights the importance of diversity and broadens
1041. 2:48:58classic film appreciation so it's actually really thinking about you know film making curating a list compiling a
1042. 2:49:07list Gathering Classic Treasures and then identifying gaps and then it gives me the list and then I said I basically
1043. 2:49:16said oh give me short plot summaries of cuz I didn't know all of them and so I read them all and some of them were
1044. 2:49:23fine um and then and then I just said let's do Jaws like I don't want your stupid High highbrow Cinema stuff and then it says shirely here's
1045. 2:49:36jaws and then it gave me 10 Concepts that could layer on top of the Jaws story structure and they were all pretty good and then I picked one
1046. 2:49:52let's see and I said give me give me a table that's got the character in Jaws what they did and then the character in the
1047. 2:50:02new movie and what they do which was pretty cool and then I said give me an outline which it did then I said give me
1048. 2:50:09a detailed outline which it did and so so anyway so so that was the that was kind of the first thing I did
1049. 2:50:17where I kind of dug into this thing or played with it so let's do the we'll do the classic how many RS are there in Strawberry so
1050. 2:50:25if you don't know this one with traditional GPT in fact let's switch over let's go to GPT 40 oh they fixed it in that one let me go to
1051. 2:50:34classic GPT so now we're in chat GPT 4 and I say um how many RS are the
1052. 2:50:50there in the word straw Berry there are two RS in the word
1053. 2:51:06strawberry why did it get that wrong well because the way these things work is um they create what are called
1054. 2:51:15tokens and a token could be a space it could be a period it could be part of a word so the word St strawberry is
1055. 2:51:23probably broken into two or three tokens and so I would assume that Barry is the second token and the second token
1056. 2:51:33has two RS in it so when you say how many RS are in the word it goes back and looks at the last token of one of two
1057. 2:51:41and it says there's two RS in Strawberry not three because that was in a different token so the way historically these tools have worked is
1058. 2:51:52they vomit out tokens and then when you ask them to think about it they just vomit more tokens they don't ever really
1059. 2:51:59understand what they vomited out the new stuff is different so let's go to the 1.0 preview in fact let's go to the Mini
1060. 2:52:08version because it'll be faster so we're going to start a new chat and I'm going to say how many RS are in in the word
1061. 2:52:28strawberry oops internal server error regenerate thought for a second the word strawberry contains two letter RS oh it
1062. 2:52:43didn't work here's the breakdown hauh Minnie got it wrong fascinating all right let's copy this there wait this Minnie got it wrong that's
1063. 2:53:13fascinating there are three Rs in the word strawberry so this got it right let's see if it explains its reasoning
1064. 2:53:23seems to be hung a little bit right now what was your
1065. 2:53:42thinking they still have a bug where sometimes when you hit the enter key it it doesn't actually submit the prompt
1066. 2:54:11huh now it's totally confused and got it wrong all right well that's [ __ ] up let me let me I'm going to option reload
1067. 2:54:21this page it it seems like it's it's in a bad way all right paste how many RS are in the word strawberry I think my browser a little
1068. 2:54:39crashy huh there it just said it got it right but it didn't show its thinking all right let's let's do something more
1069. 2:54:46interesting than this let's do um [Music]
1070. 2:55:01help me create a logical framework to determine what's better a hot dog or a fidget spinner
1071. 2:55:21So This falls into the category of play first like oh God damn that thing um giving it ridiculous prompts just to
1072. 2:55:33see what it does is is kind of fun let's see response count one oh I'm oh that's why I got it wrong
1073. 2:55:49CU I was I was using chat GPT 4.0 I got your back mod Channel please oh thank you Brandon 15 minute reminder to move your butt oh that's
1074. 2:56:01really good ah thank you you're now on 4.0 that that that okay so wait so let's go back now okay so 4.0 got the
1075. 2:56:10strawberry thing sort of right sometimes but sort of wrong others I was on the wrong model all right 01 preview okay so
1076. 2:56:19this is the big one how many RS are there in Strawberry so that's one of their pre-done prompts thinking thought for a few seconds so
1077. 2:56:30what did it do counting the letters by assigning each letter a position and checking how many times it
1078. 2:56:38appears there are three letters in the word strawberry so all right so it gave us a little thinking I've seen other
1079. 2:56:44things where it it does that okay so let's create a new prompt now and we'll do the same one [Music]
1080. 2:56:54again help me come up with two different strategies to determine what is better between a hot dog and a fidget
1081. 2:57:13spinner so thinking formulating strategy evaluating options assessing value thought for 6 seconds certainly comparing
1082. 2:57:24a comparing a hot dog to a fidget spinner is an unconventional task since one is a food item the other one is a
1083. 2:57:31toy however here are two different strategies you can use strategy one needs-based evaluation identify the fundamental needs that each
1084. 2:57:40satisfies step two assess the important of these needs step three consider health and well-being conclusion
1085. 2:57:48if fulfilling basic survival needs is the priority then the hot dog is better if mental relaxation is more of a focus
1086. 2:57:55then the fidget spinner strategy two attribute attribute-based comparison Matrix you know I've been thinking for
1087. 2:58:04years I've got to come up with an attribute-based comparison Matrix for making decisions in life all right create a list of
1088. 2:58:13attributes that are important to you such as taste nutritional value convenience cost cultural sign significance entertainment value blah
1089. 2:58:20blah blah assign weights to each attribute from 1 to five taste and then it does it and then compare
1090. 2:58:28them calculate the weighted scores taste 5 * 8 = 40 nutritional value 16 total 56 fidget spinner 51 therefore hot dog
1091. 2:58:42better so that's a ridiculous thing but the fact that it's doing this multi-step reasoning that's that's the new thing here um so
1092. 2:58:58okay so let me go back to this idea of play first we have no way of
1093. 2:59:13knowing like zero ways of knowing when you should use chat GPT 40 the the latest model before these new ones or when you should use these
1094. 2:59:29reasoning models when do you when is a problem you're trying to solve rise to the level of wanting to use reasoning I
1095. 2:59:38don't [ __ ] know you don't know we don't know so the only thing that can make sense sense is if you've been using chat GPT for a
1096. 2:59:53while and if you've ever had oh whenever I try to get it to do this kind of problem it never works that would be a
1097. 2:59:59perfect thing to experiment with with the new model see if it does it better um if you've tried coding or if you
1098. 3:00:09tried math problems historically try those with the new model apparently it's going to be better at math because it's
1099. 3:00:15actually doing reasoning and breaking problems down into sections um there are probably things that the new model can do really really
1100. 3:00:25well that we never would have even considered in the previous model just cuz we're now thinking differently we're
1101. 3:00:32we're trying to think of things we can tell it to do that are more complicated what might those be I don't know start
1102. 3:00:39playing joining late sorry if this is repetitive but we think the new model Maybe trained on old data I don't think the new model is
1103. 3:00:53trained on old data I think that well I I don't I don't know what what this new model was trained on what it what it feels like to
1104. 3:01:07me is that this new model is just a different approach to using the old model so whatever datea the old old one
1105. 3:01:18was trained on I I assume this is in that neighborhood it feels like this is like a fancier version of code
1106. 3:01:33interpreter um hater ha okay yeah I don't know what you're what you're looking for um so wait old data and what capacity
1107. 3:01:43it's not like it becomes irrelevant yeah exactly is it still training to April 2023 isn't all data from the past yeah
1108. 3:01:51absolutely um that that I don't know about I I don't think that's as relevant um what what one thing I have heard that
1109. 3:01:59that is happening which is kind of interesting is that they have gp5 but they haven't released it yet but they're using
1110. 3:02:09gp5 to generate synthetic data to train GPT 6 and then I heard the term who did I hear this from Paul rater Paul rater in
1111. 3:02:23his in his uh his weekly podcast this past week um introduced me to this term called copyright washing which okay we should
1112. 3:02:39assume that every large language model that every image model that every Audio model that we've ever used have been trained
1113. 3:02:50on not perfectly procured data these things couldn't exist without being trained on copywritten data copyright laundering
1114. 3:03:05yeah and so what they figured out how to do and it's fascinating um David Shapiro did a really interesting YouTube video on
1115. 3:03:16this where you basically say to the large language model tell me about a topic and then it vomits out a bunch of
1116. 3:03:22stuff and you use one large language model to to prompt another large language model to just vomit everything
1117. 3:03:29it knows about a certain topic and then you use a third large large language model to look at the [ __ ] that it
1118. 3:03:36vomited out and say what in this vomit list is valuable and and right and good and then you take that refined data and
1119. 3:03:44that becomes training for a new model what you're also doing by training the new model on that cleaned up content is that you're copyright
1120. 3:03:55washing it you're saying here's some data that we produced from copy copywritten trained models but that is
1121. 3:04:03not copywritten and we're going to train it on that so they're going to be able to make a claim that their new model is trained on
1122. 3:04:11data that we own and and Peak TMZ unreleased model being test to train another unreleased it's it's Peak it's
1123. 3:04:19Peak TMZ P that's that is that is what I live for um but I kind of like the I kind of like the concept so um so we'll see um
1124. 3:04:33but but this this 01 model from what I can tell just looks like it's doing a multistep not just it's pretty significant
1125. 3:04:49that as it's producing content it's looking at what it's producing before it gives you an answer the current models don't do that
1126. 3:04:59they just vomit answers out so why it's getting better is that this is what we talked about a year ago when these
1127. 3:05:05things will get better is instead of just vomiting tokens at you they vomit tokens they throw up a little in their mouth
1128. 3:05:15first and they swish it around and say huh is that tasty and then they spit it out um yeah that they're they're they're
1129. 3:05:25creating their tokens and then they're looking at what they did and refining that before they give you an actual
1130. 3:05:30answer so it makes sense that it's better and then it also makes sense that if you take that approach and put it on
1131. 3:05:36a new model that's when you're going to get something that's really really powerful so I think that what they just
1132. 3:05:43gave us is a new way of using the existing model but I don't it's it's not insignificant and again this is where we
1133. 3:05:54go back to play really experiment with this ask at [ __ ] that is is ridiculous and overly complicated and overly
1134. 3:06:01simplified and just push it push it push it because collectively we'll figure out how this is different because I don't have any clue right
1135. 3:06:14now how I would use this differently than what I've used before for I like none at all someone compared it to the other model
1136. 3:06:23for a problem and it had similar output but went deeper on the reason push it real good exactly push it push it
1137. 3:06:34good and how it can do all of that at at the speed it does is mind-blowing I know people are bitching about the fact that
1138. 3:06:42it it might take 15 or 20 seconds to process an answer like if it's doing all that reasoning why is that like that's not slow
1139. 3:06:55um I I also have heard that they're going to Li they're going to rate limit using 01 I I think it's 50 queries a
1140. 3:07:05day um is what I is is what I read but I I I don't know um anyway let me show you one other thing that I think is just
1141. 3:07:15mindblowing so I'm going to show you the output first so I'm going to go to LinkedIn this was the first thing I tried with this and it just
1142. 3:07:29staggering fact that my internet's not working also Stagg I'm curious to know if 401 uses three times the tokens
1143. 3:07:37because it's processing so much yeah I think it's I think it's Pro if it's if it's taking 12 to 15 seconds Brandon it
1144. 3:07:44there it's probably probably chewing through a lot of tokens on every query which is why they're going to rate limit
1145. 3:07:53it all right get rid of that okay so I'm G to just play something here so this is a cover for my new book that I'm working on with um Kevin
1146. 3:08:09Clark or we worked on WE we've written it we're getting ready to um to publish it right now um but just listen listen to
1147. 3:08:20this we're going to dive into something kind of mindblowing yeah the future of intelligence it's a big one but we're
1148. 3:08:27not talking you know like those Sci-Fi movies with robots taking over right this is different this is about humans
1149. 3:08:34and AI like teaming up yeah to become something even smarter than we could be on our own like unlocking superpowers we
1150. 3:08:42never thought of exactly yeah and to help us explore this we've got Collective intellig in the age of AI great book it is and you know the
1151. 3:08:52authors Kevin Clark and Kyle Shannon they're not just talking about this stuff they're actually out there
1152. 3:08:57building it yeah they're really making it happen with their company content Evolution bringing together all these
1153. 3:09:03Brilliant Minds all the best people from every field you can think of it's like assembling a team of like the Avengers
1154. 3:09:10but for brain power I like that analogy right instead of fighting villains they're tackling like the biggest
1155. 3:09:16challenges facing Humanity exactly using their powers for good yeah okay so that little podcast of those two swell people
1156. 3:09:24those NPR style people talking about my book um was just generated as you can imagine or I wouldn't be talking about it and excited
1157. 3:09:36about it so where you're going to go we're going to we're going to do one of these live you're going to go to notebook LM .
1158. 3:09:54google.com and so let me let me show you what um the way notebook works it's it's really quite quite remarkable
1159. 3:10:12so what you do in in the left hand side here are sources of data I only one piece of data in here so what I uploaded
1160. 3:10:20was I uploaded the PDF of my collective intelligence book and then I can um I can say um what is
1161. 3:10:38collective intelligence so I can use it just like a a regular large language model but it's used using the PDF of my book as the source
1162. 3:10:52so this is in a lot of ways it's a rag retrieval augmented generation tool where you can take notes but you can put
1163. 3:11:00multiple sources in there lots of sources it's really powerful collective intelligence is the combined
1164. 3:11:06intellectual power of a group of people working together which surpass the capabilities of the individual and then it it cites the
1165. 3:11:13book um collective intelligence in the a a of a AI involves humans and AI systems working together to solve complex
1166. 3:11:22problems so so here's you know a cited answer to my question so pretty cool okay down in the lower right here
1167. 3:11:33there's this thing called The Notebook guide and if you pop that open it gives you a little summary of the
1168. 3:11:39document it gives you some starter things to you know you could create a study guide out of the book or a table
1169. 3:11:45of contents or a briefing document or a timeline um and then it's got this thing called an audio overview they should call this
1170. 3:11:55the podcas ofier or something funny but you know they're Google so it's you know it is accurately named audio overview
1171. 3:12:04and then you can generate it so what we're going to do is I'm going to go back to get to your main menu of
1172. 3:12:10notebook LM you have to click on the word notebook LM there's no way to get there other than that that I know of but I'm going to do new
1173. 3:12:19notebook and then I'm going to upload a source and I think what I'm going to go do is I'm going to upload my screenplay salmon flying East
1174. 3:12:32salmon and we'll go to this Mac and we got a PDF all right so we're uploading salmon flying East into the
1175. 3:12:44system and then it automatically pops up the guide and it says audio interview deep dive conversation to hosts English
1176. 3:12:55only so I so I guess what that means so this is a deep dive conversation so what this implies to me is that they can do
1177. 3:13:02different ones of these right so you might have a deep dive conversation at some point in the future or a you know
1178. 3:13:09short little Q&A or a panel discussion or whatever it might be um so this is the first one of these do first one of
1179. 3:13:16these they do or have put out there and you hit generate and so it's now going to create one of those
1180. 3:13:24podcast Cosmic lover wow this is amazing another Kevin mallister moment yay that's why we're here um yeah this one
1181. 3:13:33to me is is total Kevin mallister so let me while that's processing it takes a couple of minutes to generate it so let
1182. 3:13:39me show you another one that I did today so I did one for my book on collective intelligence and then I have another book um called AI
1183. 3:13:53Futures and I got to go find it here this is it so here's AI Futures so here's what I did for AI
1184. 3:14:07Futures hunting today hey everyone welcome back ready for a deep dive we're tackling something both exciting and a
1185. 3:14:15bit daunting today the surprisingly Human Side of AI oh yeah definitely human and to guide us we've got AI
1186. 3:14:23Futures and Anthology but this isn't your typical dry Tech stuff is it no way think personal stories those aha moments
1187. 3:14:31even some philosophical pondering about where AI is taking us exactly and we're diving right in with Kyle Shannon who
1188. 3:14:37calls himself an accidental innovator he's got a point though right I mean the guy's been on the front lines of huge
1189. 3:14:43Tech shifts early internet the works right and now he's seeing some intriguing parallels between that whole
1190. 3:14:49early internet Vibe and what's happening with AI right now yeah that wild west anything is possible energy it's back
1191. 3:14:56and it's all around AI this time totally but here's the thing Shannon doesn't buy into the whole robots are stealing our
1192. 3:15:02jobs thing not at all in fact he actually says having AI is like having an eager intern a Golden Retriever and
1193. 3:15:09an idiot savant all rolled into one at your becking call they pulled out good quotes from getting at how a can
1194. 3:15:17supercharge what we do not replace us entirely yeah it's like this amplifier for our own creativity and how much we
1195. 3:15:24get done exactly think about it AI can take over the rep Source Camp hey what what app is this so the the app that I'm
1196. 3:15:31um generating these on is notebook LM and so it's notebook lm. goole.com and it's it's something that I haven't
1197. 3:15:42really talked about in here because I haven't played with it a ton but I know Peter Kaminsky is a huge fan of it and
1198. 3:15:47generally if he's a fan of it it's it's pretty cool cuz he's you know he's very geeky and and you know likes to
1199. 3:15:55experiment with things um it allows you to upload lots of data sources and then basically have a chat GPT a Gemini like
1200. 3:16:05experience with those data sources so it's basically every notebook is a retrieval augmented generation engine um
1201. 3:16:15it's it's pretty slick and then they just added this new feature called audio overview so if you go to the notebook
1202. 3:16:22guide which is in the lower right hand corner when you create a new one it defaults it on but if if you open one
1203. 3:16:29that existed you you click notebook guide and then you generate it here okay so this is for my screenplay salmon
1204. 3:16:36flying East so it created a five minute um audio overview which is this twers deep dive they call it so let let's hear
1205. 3:16:47what it says about my screenplay okay so we're really going to dig into salmon flying East and right off the bat you
1206. 3:16:53can tell it's more than just a fishing story oh absolutely we're talking Vietnam War flashbacks Shady business in
1207. 3:17:01saon even a possible long lost daughter popping up in Gig Harbor Washington yeah it's a lot to unpack ready to untangle
1208. 3:17:09this one with me I am so ready let's do it this is cool so we're dropped right into 1991 Gig Harbor Washington beautiful place
1209. 3:17:17and we meet Jack our main man of Vietnam vet running fishing charters seems right but there's this tension right from the
1210. 3:17:26first scene oh yeah you can cut it with a knife like right away you get that Coast Guard helicopter buzzing his boat
1211. 3:17:31why are they checking him out something's up and then boom flash back to 1971 Vietnam like the past is right
1212. 3:17:36on his heels right there with him and those Vietnam scenes man they are intense the screenplay does not shy away
1213. 3:17:44from how brutal War can be right but you also see the bond between Jack and his buddy snake and then there's Carl this
1214. 3:17:51crazy talented wood carver yeah they're in this pressure cooker playing these dead man's hand poker games you can
1215. 3:17:58practically feel the Nerds totally it's like every scene crackles with it but it's in those little Quiet Moments those
1216. 3:18:05pauses you really get to know these characters like take snake he's got this almost scary ability to read people
1217. 3:18:12always comes out on top cards you name it but there's this Darkness under the surface something just a little bit off
1218. 3:18:19and that's when it gets really good oh tell me about it one minute they're joking around the next there's a
1219. 3:18:23backpack overflowing with Heroin whoa and suddenly the stakes were through the roof and then bam Carl turns up dead
1220. 3:18:30just like that and that's what I love the screenplay doesn't give you all the answers it hits you with these twist
1221. 3:18:36strawberry came out already death and lets the mystery hang in the air keeps you guessing exactly was it some
1222. 3:18:42horrible accident so anyway something way more calcul so that's notebook LM and the audio overview which is you know
1223. 3:18:49like go make a bunch now here's what I can tell you about them the the one that I published so if you go to my LinkedIn
1224. 3:18:56Kyle Shannon on LinkedIn I've got two published I've got the collective intellig intelligence ones published and
1225. 3:19:03I've got the um AI Futures one published I also did one for Sydney for the musical that I'm working on and it's interesting
1226. 3:19:15um the AI Futures one I had to do about 12 of them so every time you do them they're completely different sometimes they're
1227. 3:19:24really long sometimes they're short sometimes they just get it wrong like Sydney they couldn't like a bunch
1228. 3:19:32of the versions I did of that one it couldn't pronounce the word Sydney it was like Sy y n Dia it was bizarre um so if if it gives you a
1229. 3:19:44shitty one there's a little drop down menu in where you created it you can just delete it so what I've been doing
1230. 3:19:51is I'll generate one if it sucks I just delete it and then generate another one like I did like a dozen of them until I
1231. 3:19:59got one that I thought was decent enough to put out there um the other thing that I did you'll notice in my
1232. 3:20:09um that I've got this book cover I've got this this video the reason I did that I didn't know this but if you're on
1233. 3:20:21Twitter you cannot upload an audio file you have to upload a video file so what I ended up doing was I took a JPEG of the cover of the
1234. 3:20:37book and went into iMovie and made a movie with the cover and I you it's so [ __ ] up how you have to do it in imov movie you have to
1235. 3:20:48rotate the movie 90° so your cover is sideways cuz it makes videos that are wide like this and then you export it
1236. 3:20:56and then I bring it into Quick time and I rotate it back but basically this is just a picture you know for the entire
1237. 3:21:03length of the audio it turns it into a movie and then you can upload it to Twitter but then I thought it's actually
1238. 3:21:08kind of cool because it's the cover of the book it's like an ad for the book right totally but here's the thing
1239. 3:21:14Shannon doesn't buy into the whole Rob so anyway pretty cool right [Laughter]
1240. 3:21:28uh did anyone get 01 mini yeah I got 01 mini and 01 preview and I I I think that these are basically both like I think 01 preview
1241. 3:21:41is built on top of GPT 40 and 01 midi is built on on top of gp4 mini that's what I'm thinking all right Kyle before you leave
1242. 3:21:55you must pop over to hum. all right let's go to hum. let's do one of those okay so so assignment one go play with 01
1243. 3:22:07preview and and start to figure out start to get your head around how do you how would you prompt that different than
1244. 3:22:14you prompt regular chat GPT number two is go to notebook. goole.com and make yourself an audio overview of
1245. 3:22:24something you've written and it's just cool and then hum. a h m.ai they came out with a their EV their new voice model called
1246. 3:22:38eeve talk to our personalities voice AA e Stella Finn
1247. 3:22:52Docker Whimsy okay let's see I'm going to do Stella we're going to start a call with Stella allow it to use our
1248. 3:23:09microphone hey Stella how's it going oh hello there I'm doing marvelous thank you for asking how may I be of
1249. 3:23:18service to you Hey listen could you whisper to me certainly darling well go ahead do it I shall with the utmost discretion
1250. 3:23:32now tell me what's on your mind can you get really excited about something and tell me it in a poem Oh what a delightful request here's
1251. 3:23:47to your request with a touch of Whimsy in verse when words are spoken in Rhymes oh what a delight poetry can paint a
1252. 3:23:54picture so bright it ignites the mind and heart so speak your heart's desire and I'll respond with rhyme and
1253. 3:24:02cheer so Stella your voice seemed to change with every line of that poem what's up with that ah you've caught on I'm Stella the
1254. 3:24:13voice of Distinction but I'm also quite the dramatic actress it's a delightful Quirk of mine adding a touch of
1255. 3:24:19theatricality to our chat but rest assured my dear the substance of our conversation remains as refined and thoughtful as
1256. 3:24:27ever why don't you give me a couple of lines from King Lear as King Lear oh how marvelous let's channel the
1257. 3:24:37great Shakespearean Monarch for a moment I shall be king Leah Regal and mad dear the world is a stage and all the men and women merely play
1258. 3:24:47come let us observe the tragedy of my poor daughters and the treachery of my sons fascinating all right let's go to another voice
1259. 3:24:59Finn hey Finn what's happening hello hello not much just enjoying the view how are you doing I'm doing good it sounds like you're from Down
1260. 3:25:12Under you caught that did you hey I can I interrupt you and Aussie twang no worries at all you didn't
1261. 3:25:21interrupt anything important I was just taking in the beauty of nature it's always a treat to chat with someone as thoughtful as
1262. 3:25:29yourself I'm flattered tell me your favorite area in Australia oh that's a tough one I've got so many favorite spots but if I had to
1263. 3:25:43pick I'd say the Great Barrier Reef is absolutely is stunning the vibrant coral and diverse marine life it's like
1264. 3:25:50stepping into another world I heard that the coral was dying off because of global warming and pollution and
1265. 3:26:01microplastics you're absolutely right climate change and pollution are taking a real toll on our beautiful Great
1266. 3:26:06Barrier Reef the coral is struggling to survive and it's heartbreaking to see go
1267. 3:26:21on absolutely it's crucial to educate ourselves and others about the impact of our actions all right
1268. 3:26:29so the way Hume works you you used to be able to see it more empathetic voice
1269. 3:26:44interface so it reads not only your words but it reads your emotional inflection and then it's supposed to be replying in kind so
1270. 3:26:55let's go to let's go to Whimsy see how she is hey Whimsy what's happening greetings dear friend I am Whimsy your guide through the tapestry
1271. 3:27:08of time you've stumbled upon a most curious scene a medieval Tavern where we find ourselves what would you like to
1272. 3:27:14see or hear about in this cozy in oh are we doing a little little Adventure here is that what's happening indeed we are embarking on a
1273. 3:27:23merry Quest as we enter a jovial B greets us his loot in hand he offers to share tales of the town's history and
1274. 3:27:31legends what would you like to hear I think I'd like to hear a story more about time travel than old and timey
1275. 3:27:41crap ah a Bor Choice let me spin you a yarn about the town's time traveling Tavern keeper Legend has it the inkeeper
1276. 3:27:50can step through time with a flick of his hat he's on a mission to save the world from an impending disaster but our time is running
1277. 3:27:58short so what choice do I need to make the choice my friend is to join the timekeeper quest will you assist him in
1278. 3:28:07gathering the necessary ingredients for his time altering Brew or perhaps you'd rather stay put and hear the old man's tales of the past
1279. 3:28:16can I just tell him not to tell me his stories the eke Keeper's face Falls his brow furrowed in disappointment but then
1280. 3:28:27a mischievous grin spreads across his features perhaps He suggests you could be the one to save time itself from
1281. 3:28:34unraveling the fate of the universe rests in your hands what would you do check the time on my wristwatch your wrist watch my dear is a
1282. 3:28:46Time piece of the finest craftsmanship it tick tocks away counting down the moments of our adventure together as we
1283. 3:28:52stand here in the tavern the hands on your watch read 12:34 p.m. but don't be fooled this is just the moment before we
1284. 3:28:59leap through time the the other thing you looking at here is you've got excitement concentration determination
1285. 3:29:08so it's a balance every one of these lines is a balance of emotional waiting so anyway you can they've got AP you can
1286. 3:29:16build stuff on this so that's something to go play with so there you go new chat GPT model notebook lm's audio overview
1287. 3:29:25feature and hum. A's voice it's so much better than the first version though yeah it's so much better all right Kyle
1288. 3:29:34doesn't look too impressed by this no I'm pretty impressed by it I you know I'm still waiting for [ __ ] open AI to
1289. 3:29:39release their stupid voice model which they said it was going to be a couple of weeks back in May I still like pie yeah I'm I'm with you
1290. 3:29:50all right I'm going to get out of here go to the salon. if you haven't joined the salon go there join us read about what we're
1291. 3:30:01about join us follow my channel subscribe to the lives if you want to support it pick up the video series in
1292. 3:30:08the corner called underhyped cuz I think AI is underhyped not overhyped take that Doomer you come here you get optimism you get
1293. 3:30:19expansive chat TMZ speculation [ __ ] the [Laughter] haters I hope you had fun
1294. 3:30:35tonight uh the Oprah thing it it was it was less open AI Centric than I thought it would be it was definitely for the masses it
1295. 3:30:46definitely leaned into ah this shit's scary what if the robots replace us run was a little bit more of that than I
1296. 3:30:57was hoping for but that's where we are right now like this is Katie kurick in 1994 what's that a with the circle around it what is internet
1297. 3:31:12anyway are are the robots going to kill us Christopher Ray oh [ __ ] they can scam us a hey sociology lady how are humans
1298. 3:31:28humans are great and this project this AI project is is designed to replace us and that's evil bizarre bizarre but that's where we
1299. 3:31:44are where we are again this channel not good AI is not good not bad just inevitable and so choose to play with it choose to to tap
1300. 3:32:02into the Staggering power of collective intelligence that is programmed into these systems CU it it really will change you
1301. 3:32:14it really will change what you think is possible thanks Kyle see you tomorrow morning for office hours oh yeah
1302. 3:32:20tomorrow morning 11:00 a.m. Mountain Time 1M Eastern on LinkedIn I do AI office hours so if you go to my LinkedIn Channel at
1303. 3:32:31Kyle Shannon on LinkedIn you'll see I've got a picture there it looks like this except I think I'm going
1304. 3:32:42like um tomorrow morning I'll post a reminder I usually do some quippy little thing I'll do something about the new
1305. 3:32:50open AI model or Oprah or something um and you can also see past events it's always the same URL so you can go to one
1306. 3:32:58of my past events see the URL there um and just join us it's a really the group tomorrow is a really good group of
1307. 3:33:07people um and we'll just talk about stuff you can ask questions so it's office hours take care of your back Kyle
1308. 3:33:12thank you Corey you're the backbone of our community no no pun intended right you're you're the you're the outof joint backbone of
1309. 3:33:21our community 20,000 for Oprah um all right so let's do what we do here so if you don't know this when we sign off at
1310. 3:33:29night we do a little Romper Room magic mirror so I see Rick Olsen and Cosmic lover I see joeore y I see Corey Sandler
1311. 3:33:41pottery and who else do we got in here tonight Teton AI good to see you Source Camp fantastic honey cure oh honey curls
1312. 3:33:54GC drat I will be flying Ah that's a drag it's fun um Silver Fox just sent a gift thank you so much for the gift that means a lot Mr
1313. 3:34:06it Papa JJ in the house Be not Afraid Danielle thank you so much a i Bob in the house Source Camp go check out
1314. 3:34:14Irregulars all right let me let me go do a quick Irregulars run the hot dog fidget spinner very
1315. 3:34:33nice oh Kyle look what I discovered today where's that is that near you look at that Challenger that's like a 71 I think this is
1316. 3:34:49cool Dodge Charger RT these are these are some baller cars dag I think that's a 69 or 70 Charger 71 or two
1317. 3:35:02Challenger forget what this one is with the the Stacked headlights it's a is it a Nova no it's a chevel I forget what it
1318. 3:35:10is dag those are cool nice strawberry in the house AI learning lab very nice oh wait who did this I made a functional solitire game so look
1319. 3:35:26at that Danielle out there programming creating solitire games fantastic oh those are nice all the AI Salon logos very
1320. 3:35:42cool James Earl Jones Pass that's it's a drag but really nice image beautiful all right everybody um let me see ton reasoning prompt a
1321. 3:35:57Candlestick with blood was found on a bridge oh this is a good one okay Teton AI so murder mysteries a Candlestick
1322. 3:36:05with blood was found on a bridge what might have happened love it that's awesome oh yeah all those all those sort
1323. 3:36:13of riddles we we got in seventh grade where you had to think about the riddle and solve it that's good that's a good prompt for it love
1324. 3:36:22it Kyle pulls me back when I get distracted by bright shiny objects nice uh let's see Rick Olsen forever hooked
1325. 3:36:30in the house uh thankful Jones thanks me thank you Dan out of here good to see you Dan thanks for hanging out uh Mr it
1326. 3:36:39good to see you Tobias good to see you two block Tom archetypal architect in the house all right Bob H who else we
1327. 3:36:53got bang honey curls all right I am out of here Brian Engle childless cat dad love it Marina she recall Tech all right uh
1328. 3:37:10I'm out of here people that was fun tonight uh hope you had a good time sorry about the snafu at the beginning I
1329. 3:37:15didn't realize that we can't just watch [ __ ] thanks Disney peace have a good night