1/13/2025 - AI Agents & the Paranet Protocol: Orchestrating the Future of AI

Transcript

1. 0:03[Music] d
2. 0:25[Music] [Music]
3. 0:55standing between you and H is [Music] insane standing too near you and a file makes it
4. 1:11[Applause] clear your trouble to [Music] meble can't you see leaning in
5. 1:28clothes smelling your perfume that scares me most leaning awayy it feel stronger every
6. 1:44day your trouble to me Rich can't you say good evening good people
7. 1:58[Music] she came on him like slow moving
8. 2:14C his be was warmer than a look in her eyes she said on a stool and he said what do you want
9. 2:29[Music] she said give me a love that don't freeze up [Music] inside he said I have miled it some
10. 2:44hearts in my time dear but to sit next to you will I shiver and shade and if I knew love well I I don't
11. 3:00think I'd be [Music] here asking myself if I've got what it t to melt your icy
12. 3:16blue I don't know that to me no whatever doesn't [ __ ] matter good evening happy [Laughter] Monday have you seen that AI learning
13. 3:30lab you know it's interesting he's got a dog that sings the dog sings beautifully he's always on key it's just a fantastic
14. 3:37situation well with the exception of the singer the the man part of it and but anyway so he s he starts the song and
15. 3:45it's like he's he's singing along and he's singing along everything's fine and then just like doesn't know the song
16. 3:50it's like well why would you play the song if you don't know the song it's just I don't understand could someone
17. 3:56explain to me why people watch this I can't either I'm right there with you listen we got it's a whole [ __ ]
18. 4:05situation it's a whole situation I you know I've got a dog that's under contract he has to sing so I can't not
19. 4:14sing um I know seven songs and so you know we we sing what we sing and then I know 150 almost songs [Laughter]
20. 4:31oh good lord good people there's so much we could talk about tonight I don't even know where to begin I don't even know where to begin
21. 4:42oh I wait so so first of all let me let me flip my little view over here on the Tik Tock hi ticktockers
22. 4:54now I'm talking to everyone at the same time um t to what what's the date 13th six more days I heard today R rumor today is
23. 5:05Elon Musk is going to buy Tik Tok Mr Wonderful's trying to buy Tik Tok Supreme Court's supposed to weigh in Wednesday Tik Tock is
24. 5:19it I have I have avoided the Temptation so far of listen this platform's going to go away and I'm going to miss all of you
25. 5:30so I'm gonna I'm gonna give another day or two see see if some things start playing out but uh a lot of lot of lot
26. 5:38of heartbreaking sadness right now everyone's running over to red note so red note if you don't know it is
27. 5:52the actual Tik Tock owned by the actual Chinese government in Chinese it's a available on the app store for some
28. 6:01reason uh apparently there's been a mass uh downloading red note is now like the number one or two app on the Apple iOS
29. 6:11app store because 170 million people were like [ __ ] it let's go to Red note you don't want us on a Chinese app why
30. 6:19don't we go on the one that's actually spying on us and people are logging in there going where's my Chinese spy I
31. 6:26want to meet my Chinese spy oh good lord so that's happening um let me get comments up and running here I I
32. 6:41am without a producer tonight so red notes not red noes red as in the Red Army red note paranoid people from the 1950s are
33. 6:57flipping in their grave right now what do you mean you're downloading the Chinese [Music]
34. 7:18app um if if you don't know me if you're brand new here God love you God love you you're just you're just tutored in a
35. 7:26little button my name's Kyle Shannon this is the AI learning lab we we do all sorts of things here if you're here on
36. 7:32LinkedIn you might think like well this doesn't seem very professional you're correct uh this is a this is a professional
37. 7:41uh cell phone on my part uh I should not be streaming this on LinkedIn but I have ADD and I'm lazy and once I got it set
38. 7:50up it's easier just to keep doing it than not so I figure people on LinkedIn who know of me as a upstanding professional
39. 8:02oh they'll get over that they'll they'll find out soon [Music] enough and for those of you not on LinkedIn if you're on the YouTube that's
40. 8:18going to be the primary home of all this [ __ ] um starting you know if if Tik Tok goes away um but right now I'd say it's
41. 8:27kind of 5050 anyway we're we're generally uh streaming live on Tik Tok over here and the other three platforms
42. 8:36X and Linkedin and Alan hilbeck is over there on LinkedIn holding down the fort for the professionals very nice I like it [Music]
43. 9:13all right Vicki's putting up some good stuff so don't if you're on YouTube couple of things we need there we need
44. 9:20plays and we need follows subscribers doesn't cost anything by the way I realized I've been asking people to
45. 9:27subscribe to YouTube and there's likey how much is that going to cost me you son of a [Music] [ __ ] it doesn't cost anything
46. 9:41[Music] wait in a Westerly [Music] Direction car is my train
47. 10:00driving I've been [Music] wondering what it is I'm running from [Music] again Cody x i play a different video
48. 10:12every night I may or may not be awake for it thank you that's that's getting the that's getting the play count up I
49. 10:18think we're pretty close actually but but thank you very much um I think what I'm also going to do I I mean here this
50. 10:26is part of the [ __ ] up thing about Tik Tok going away so I paid some decent amount of money to to download all of my Tik Tok videos and
51. 10:37push them over to Google Drive so I'll have them all and then I was gonna upload those to YouTube shorts but I think YouTube shorts don't
52. 10:46they have like a three minute cap or Max so I don't know so I guess I'll just do vertical videos as normal YouTube videos
53. 10:56which I think is [ __ ] stupid but whatever love your hair tonight oh boy what have I done I don't oh it looks all right
54. 11:04it's a little it's a little it's a little oh you know what I did I I took a little I did a little I did a little
55. 11:11eyebrow trim with the the eyebrows I I was looking mighty you know uh what's it called Nutty Professor
56. 11:20like so I did the eyebrows and I did a little bit of triming on the sides there I need a haircut so I just did a little
57. 11:26self self-own on the haircut Thank you very much the other thing I started today is uh I'm just getting too [ __ ] [Music]
58. 11:40fat so I let my I started my it's the holidays I'll just start eating more I started that like the week before Thanksgiving and then I have
59. 11:53continued that gluttonous eating agenda uh until this morning and I was like [ __ ] it so uh so I'm starting
60. 12:05intermittent fasting today and I think I'm just gonna have like a green drink for lunch and then just dinner and I'm
61. 12:12just going to do that until this body feels less ridiculous I know from out there because of all the Prosthetics and the makeup
62. 12:22that it looks like I'm [ __ ] jacked but situation is if you saw me stand up you'd go oh my you know it it would be
63. 12:32it could be it could be a shocking moment Corey Sandler hey Kyle some of my Tik Tock lives are seven eight nine
64. 12:40hours oh you're insane at The Potter's Wheel I know Corey what are we supposed to [ __ ] do with all this content and
65. 12:46I mean I guess I mean one of the answers is Let It Go Right sa La V we we got you know there's part part of it's that
66. 12:59you know the thing that that kind of baffles me this this really does kind of baffle me is that what is it that Tik Tok did
67. 13:12that is so different than reals and and Instagram whatever the [ __ ] they do or YouTube shorts and all those they're all
68. 13:22effectively the same thing [Music] I mean I guess it was the algorithm right just showing people [ __ ] that they
69. 13:33might like instead of their social [Music] graph is Blandon a green drink yes I will include it in
70. 13:50that it wasn't image based to start but video based yeah but YouTube shorts was that although I guess YouTube shorts
71. 13:57came later YouTube was kind of like you know old old people old people video I don't know I don't know and twitch C certainly has this for
72. 14:13gamers [Music] right so if you're wondering what we're
73. 14:29going to talk about tonight um we're going to talk about agents I think I think I might um do a little exploring with
74. 14:43[Music] uh with Runway ml with character Generation stuff um I've been working on a project for work and and digging a bit
75. 14:58deep ER on Runway ML and act one um and man it it takes some work but what you can do if you put in the work is pretty
76. 15:12stunning so we can go look at that um Brandon would be excited to know maybe we'll play with some jelly pod
77. 15:23which is a podcast maker that's kind of like notebook LM but you get to choose the the voices and then Vicky just sent
78. 15:31me another one that is similar to that that's open source um which looks interesting and what would be cool is if
79. 15:43I showed the thing Vicky sent me and not Brandon's Brandon can watch this later and get bitter about it and I think that's [Laughter]
80. 15:53funny thank you Lord digital Gods thank you for the croses the open source one is called um well it's on foul. a f.
81. 16:11a and then it's called play AI so foul. a slm models here here's I'll do this copy
82. 16:26that so if I put this here and do
83. 16:41that and do that there it is right there on the screen on the big screen all right don't worry about Brandon he's a big
84. 16:54boy Vicki's already throwing him under the bus Brandon it's like your seat isn't even it's still warm it's still
85. 17:01this is like a Shakespearean drama over here all right um I W to I want to I made a uh I did a
86. 17:16Google deep research thing on AI agents as well so over on the AI Salon I shared that research document um which I think
87. 17:27it's pretty [ __ ] amazing Google deep research if you have not played with it or if you not have not experienced it it's pretty [ __ ]
88. 17:36insane [Music] um so so if you stick around I'll I'll give you the locations of the location of that thing hang on my glasses are like
89. 17:48champy took a dump on them what what is it with glasses like you don't touch them right you they're just on your desk and you
90. 17:58put them them on your desk and they're clean and then you pick them up and put them on and it's like someone took a
91. 18:03dump on the lenses what when does that happen are there like fingerprint fairies that just run around [Laughter] like get this one really
92. 18:19deep what the [ __ ] good Lord camcat kid all right what can you say about the AI diffusion AI diffusion rule from today I don't
93. 18:33know anything about that what did I miss what did I miss hang on hold please AI defusion
94. 18:53[Music] rule this thing Biden Administration proposes new rules on exporting AI chips is that from today
95. 19:08no oh it is well the horse is already out of the [ __ ] Barn did you see there's a new model
96. 19:23today is it a coding model or over the weekend it came out it was a coding model model that they trained on $450 of
97. 19:32compute that is like as good as 01 preview it's it's like benchmarking as good as 01 preview trained on $450 of compute the only thing that
98. 19:48that okay provoking an industry push back so I don't I don't know anything about this but it's like it's it just
99. 19:59seems to me it's like well okay so China's kicking our ass because our last our last exporting rule where we we
100. 20:09hobbled the chips that went to China forced them to innovate so they got better at working with shittier [Music]
101. 20:18chips like like at some point someone's going to recognize that we gota we got we've oh [ __ ] like I don't even know where to
102. 20:33start it's um the fact that we have zero Federal activity about AI
103. 20:48regulation they're pushing it all to the States they're just ignoring it federally and they're pushing it all to
104. 20:53the states and they're allowing the states like Colorado that went first on this to pass really shitty over Brad
105. 21:00rules that are regulating the tech and not the use cases which every time you do that it's a bad situation and the states because they are
106. 21:13like radically uneducated about what's actually going on with AI they're just passing laws so they can go we're out
107. 21:20there protecting you people you we it's good this this is what the politicians are like here ye here you leave fellow
108. 21:27citizens we understand that the AI robots are going to kill you and therefore and China's evil and except
109. 21:34for when they're not and but we're going to never give them the good stuff un but on a state-by-state level we will
110. 21:42prevent you from using AI so that it protects your interest unless of course you're an entrepreneur and then good
111. 21:49luck with that and by the way how do you determine whether a piece of AI technology is being used in this state
112. 21:56or that state well you as the entrepreneur can figure that out you're [Laughter] [Music] clever this is our [ __ ] legislation
113. 22:10we got the most profoundly powerful technology in the history of humanity and we're like ah will figure it out [Music]
114. 22:26sometime good Lord Lord good Lord people all right hang on he seems he seems a little aggressive tonight I don't understand what's going
115. 22:40on but he certainly does not seem stable uh Senator Ed Markley this is from Jim Walters various posts a few hours ago
116. 22:55Senator Ed Mary filed a bill to extend the deadline Tik Tok for another 270 days so you're saying we got a chance I believe in the finger
117. 23:09fingerprint Sprites very nice glass is one of life's great Mysteries summarize it in chat gbt oh that's a great idea
118. 23:16all right let's go do that we'll do copy here hang on we'll go share this with all of the world to see we'll put the
119. 23:23black bar back up because I'm a professional uh and then we'll go over uh wait um dang it there we go black bar back over here yeah that's
120. 23:40working 01 and o02 reasoning models are amazing except O2 is now called 03 but yes they retired the chat Bots I made
121. 23:48several months ago yeah I know it's pretty amazing uh and people that are are uh playing with um 03 or 01 Pro are
122. 23:57saying that it's quite a amazing that's the 200 Buck a month thing uh let's see summarize this article thinking I'm unable to access oh
123. 24:12links externally that's because uh let's see try now [Music]
124. 24:30[Music] let's see what it says here the Biden Administration has introduced new regulations to restrict the export of advanced artificial
125. 24:43intelligence chips aiming to prevent adversarial Nations China from acquiring Cutting Edge us technology [Laughter]
126. 25:01have you noticed that every Chinese model comes out about three or four weeks after the US model comes out and
127. 25:08it's the exact same [ __ ] thing you don't think they have access to our technology these I you know the more I
128. 25:17learn about politics the more I realize oh it's just all [ __ ] PR um these measures Target high performance semiconductors okay
129. 25:26Department of commerce's Bureau of industry and statistics bis will oversee these controls fine industry leaders
130. 25:33have expressed concerns that these restrictions could hinder Innovation and competitiveness they argue that limiting
131. 25:38access to International markets May reduce Revenue that's it fascinating so they're saying don't take re don't take
132. 25:45Revenue away from Nvidia or if we ever get our [ __ ] together affecting research developments and Investments well I understand the
133. 25:53reason to try to protect the US interests like we've got a a [ __ ] ton of elite and we've got a lead in um in energy generation as well I
134. 26:05think the administration contends that these measures are necessary to prevent the misuse of AI technology and Military
135. 26:12applications and human rights abuses by controlling the distribution of advanced AI chips it how about how
136. 26:19about passing some [ __ ] reasonable AI legislation at the federal level here [ __ ] idiots and I mean [ __ ] idiots in the
137. 26:28most positive possible [Music]
138. 26:44light is this a place I can rest my PO [Music] head together my thoughts in sweet silence
139. 27:01there this place where the feelings on De from an over exposure to violence where's champy champy already
140. 27:15sang champ sleeping he did his he did his Union minimum and he has checked out of the building oh [Music]
141. 27:30um all right so here's what I'd like to do um what are we going to do so here okay hang on just just bear with me bear
142. 27:41with me self- production here self-producing Clueless Okay if you've not gone to the AI Salon if you've not
143. 27:51joined the AI Salon you should do that go to the salon. scroll down until you see button that says join our community click that
144. 28:00button it doesn't cost you anything we did over the holidays you may have heard of this thing called AI FES of us AI for
145. 28:07the rest of us we did that we partnered up with Anne Murphy and the she leads AI Community we put on this two-day workshop Aon had 2300
146. 28:21people show up over two days on a holiday weekend um that's that's AIS Salon um [Music] and we just did some we just spent the
147. 28:37weekend Vicky Baptist and Leah and I and Eric Sanner showed up I'm pretty sure hang on a sec we're gonna go here we're g to go
148. 28:55here there we go all right so here's the AI salon now over here on the left hand side um there's a couple of things so there
149. 29:07there's a welcome area so when you first get to the AI Salon you land on the welcome there's a welcome video uh from
150. 29:13from Leah faston and I we co-founded the salon and then we talk about the journey to AI Readiness which is which looks
151. 29:21like play first mindfully create generously lead which I love that little this was sort of born of how this community acts and and what we
152. 29:33do and then there's a a little seven-step onboarding process you you sort of log in learn about us introduce yourself um check out our
153. 29:46values curiosity playfulness empathy bravery generosity what we're all about you can watch past meetings if you want
154. 29:54to see we've been doing this since December 7th 2022 so the AI Salon was founded the week after chat GPT came out
155. 30:03um you can RSVP for upcoming events you can go into the water cooler and if you go into the water cooler one of the
156. 30:10areas you can go is this thing called discussion threads and if you click on that the very first one is that research
157. 30:19document that I created using uh Gemini deep research so this is a list of AI agent platforms from no code to low code
158. 30:30to go build your own [ __ ] and with with it's pretty it's a pretty cool little research thing that that Google deep research
159. 30:38did um and then there's all sorts of other stuff and there's there's a news area there's a showand tell area where
160. 30:43you can show off your work it's called showof and toolbox where you can show off uh images you've made or gpts that
161. 30:52you've made whatever it might be that's really nice from CJ very cool so you can go share off show off your
162. 31:03work and then within each of these sections is a tool list of all the different tools that are worth checking
163. 31:12out and then we used to have what we called guilds but here's what we did some of the guild the guilds there were
164. 31:19too many of them they were confusing it we weren't sure where to put them and what we realized was that what the
165. 31:25guilds were affectingly acting as has is like clubs or or sub communities or hubs within within the space so so there's
166. 31:34now just sort of this showoff area where you show off your work and then there's clubs and hubs so there's AI Festivus
167. 31:40was you know that that event we did so there's a club for that there's a club for the Irregulars which is this group
168. 31:47these clubs are ordered by level of activity so for those that run clubs if you get more people to use your Club
169. 31:54you'll go up higher in the list which is kind of fun um and it just makes the the space a lot more simple but there's so
170. 31:59much going on here it's so valuable so go check all this out all right and then in the water cooler under discussion threads that's where
171. 32:11this this Google doc is navigating the AI agent landscape categories of AI agent platforms no code low code developer Focus
172. 32:24platforms then it describes them's benefits links to the site tutorials for that particular thing videos
173. 32:37insights now are these the right ones I don't know I just had Google deep research go do all this but it it went
174. 32:44out and it searched like 250 websites and then made this report so there you have it it's Google deep research is pretty [ __ ]
175. 32:55cool but anyway so that's all in there so if you want to go start learning about um autonomous agent platforms I I
176. 33:01think now's a good time to do that um okay so I want to talk about why I believe that is the case let me go find
177. 33:09something here let me let me while while I'm searching let me just do this um any questions any thoughts thoughts questions
178. 33:25comments love Serena's graphic I do too One-Stop shop for AI thanks for this you're welcome you're welcome cuno yeah this
179. 33:40was one where um so so agents have become the talk dour and I
180. 33:59if we think the government is ill-prepared I don't listen I don't know why I'm laughing because it's [ __ ]
181. 34:09tragic like like you know there there should be a law that lawmakers can't be this [ __ ] clueless for something this
182. 34:17important but here we are um if we think that legislators and people are ill prepared for the current state of AI I I have a
183. 34:28feeling 2025 is going to be a uh a Fantastical popcorn eating movie watching uh Extravaganza cuno oh you had your
184. 34:42Minnesota meeting today I that's right um who was there it was it was Lori Ryan Chris vuno and someone else who was the
185. 34:50third person that was there Chris uh uh uh uh uh uh uh uh uh [Music]
186. 35:04oh kakei that's right that's right that's right khaki and khaki so I talked to the people that organized that event
187. 35:12and they said it was a hit were amazing um they said there's actually a follow on opportunity so great so so about
188. 35:23three weeks ago there was an AI education event for Oregon legislators that Anne Murphy went and represented
189. 35:32the AI salon and today there was one in Minnesota and what's what's amazing about these events is um the thing about the AI
190. 35:44Salon is not only is it really nice people and really generous people and really smart people but it's people that are are kind of
191. 35:55impressive and and I I you know I I sometimes forget that like like like we're just kind of hanging out talking
192. 36:04about Ai and we're asking good questions and things like that but we're doing it on a daily basis and and if you take
193. 36:13people who are really established professionals and over the course of two years they just sort of immerse
194. 36:21themselves in a topic and they take their professionalism and they sort of Route it through that topic you know
195. 36:27we're some of the most educated impressive badass people around this idea of generative AI not not AI
196. 36:34traditional AI so much but more like now that chat GPT is here and now that real people and real small businesses are confronting what these
197. 36:44tools are and and we're learning what's possible we're like you legislators need to [ __ ] deal with that and what's
198. 36:53been the result of of two out of two of these meetings this year is that AI Salon people have showed up educated the
199. 37:00legislators and the legislators have left those meetings going we got to get our [ __ ] together so so thank you to to
200. 37:08all three of you today for doing that the the this the spirit of the salon is also I reached out to these guys because
201. 37:15these things come up pretty quickly like they're sort of happening in real time I reached out to these guys seven days ago
202. 37:22and and all three of them said yeah I'm in let you know let me go do it so awesome I'm so yeah I heard it went
203. 37:28great and I just I didn't get a chance to get a note out so so thank you for for representing the salon that's that's so awesome so
204. 37:37awesome karuno we're decent at how to operationalize AI in various sectors and you know Chris it's not just how to
205. 37:43operationalize it it's not just that it's you can sit across a table and when someone is uneducated about what it even
206. 37:55is you can hear them not get defensive educate the room Elevate the conversation above fear into
207. 38:09possibility and then you can talk about and here's how you operationalize it right it's like you can do that too but
208. 38:15it's like there's there's a there's a much more fundamental skill right now which is there's a lot of Fear and Loathing around
209. 38:23Ai and being in the presence of people people that are like no no no you don't get it yeah yeah there might be risks
210. 38:31but there's there's all this other stuff no one's talking about it's opportunity it's optimism it's people Reinventing
211. 38:37themselves it's people coming out of retirement right that was the the other thing that the the the folks that
212. 38:45organized it said today which I thought was I thought was really cool is they were like you know like like all three
213. 38:53of them were like basically what they said was like gen xers like there were were no young people there I'm like I
214. 38:59I'm like I'm telling you Gen X is crushing it with AI like there's something about our you know ability to
215. 39:08be a pain in the ass our ability to to have have recognized that we saw this with the worldwide web and now we're
216. 39:14getting to see it again with with AI we're like oh this one's bigger and I think there's there's we we've just got a spirit of let's screw
217. 39:23some [ __ ] up here I'm really excited about that so anyway that's awesome all right Silver Fox I'm so part so excited
218. 39:32to be a part of the AI salon and learned so much it's a great group thank you so much that's awesome Jeff Flanigan
219. 39:40digging into future pedia today lots to unpack there too much to unpack there so much to unpack there khaki and how do we leave a seat
220. 39:49at the table and how do we leave a seat at the table for oppressed communities we got to get them we got to get them in the
221. 39:55conversation khaki I mean that's the that's the that's the deal it's one of the things I'm most excited about with
222. 40:02AI is that it gives everyone a chance to dramatically level up their skills no no matter what their access in including
223. 40:12you know a lot of accessibility stuff is is going to be bridged um so I think the tools are there but it's you know how we
224. 40:19get people at the table is we get people at the table um you know we we've got CJ Fletcher within the salon has got luck L
225. 40:29Cultura um where he's he's uh he's got a group within the AI Salon that are exploring um you know people of color
226. 40:38um and how they can how they can use these tools how they can express themselves how they can drive things
227. 40:44like that so I think it's like I think that is an active conversation how do you get people in the conversation you
228. 40:50get them in the [ __ ] conversation I don't think it's any different than you know what we do here I don't specifically Target oppressed
229. 40:57communities here um but I think it's I think it's what this activity is just aimed at different audiences right talk
230. 41:06about it demystify it figure it out have fun with it play explore experiment all that sort of stuff cuno the cool guy from the
231. 41:16governor's office was very engaged with cool use cases use cases are the bridge to get people to understand what's
232. 41:22possible getting involved right now is how to get at get to the table I mean that so so r rispy Wait ripsy ripsy
233. 41:32you've got it where where it's like we are so early right now it may not feel like we're early but trust me we having
234. 41:42been through one of these before in the 90s um we're early there are so many people that don't even know what's
235. 41:51happening like don't even have a clue what is possible today they don't have a clue what was possible two years
236. 41:59ago they have less of a clue what's possible today they have no idea what's about to hit them with with autonomous agents
237. 42:09um anything we can do to get anyone into this conversation right now will give them disproportionate opportunity and
238. 42:19have a disproportionately positive impact on whatever this industry becomes Rachel thank you Rachel wait
239. 42:31ripy same experience and I completely agree um what is possible today is so impactful and it's right here yeah it is it is here today like
240. 42:42the it's I every time I do something now like that that research thing that I just shared oh the thing that's on
241. 42:50screen right now hang on let me pop that down and go back to this this research thing it took it took Google deep
242. 43:00research about I don't know 10 minutes to process this but you know here's a 16-page research document on autonomous
243. 43:11AI platforms you know with citations like like you can just do that now well in this case you have to you
244. 43:20have to have a subscription to it and you have to use an individual Google account because if you've got a Google
245. 43:26work group account you're considered a pariah if you pay Google they hate you but like like this thing's amazing here let me show you
246. 43:39um where can I show you something here if I go to um
247. 43:54Runway we're working on a project for client generative session is it this
248. 44:17one let me change my sharing so you can hear this
249. 44:41[Music] um so this is like a virtual sales rep named
250. 44:55Michelle three 3 2 1 hello my name is Michelle your virtual sales rip it's great to connect with you our digital platform allows you to get
251. 45:07the latest updates on our products and services order samples and patient savings cards and access educational
252. 45:14resources to keep you informed and prepared for evolving Healthcare needs three two I mean like she doesn't
253. 45:26exist the voice doesn't exist that like I actually did the voice acting for this and the face acting for this and then swapped in her
254. 45:35voice you know we can do this today like this we can do today no special effects you know me me sitting at a computer
255. 45:43like hey welcome I'm your virtual sales rep it's just amazing so um all right I want to go find something so we can talk I want to
256. 45:55talk I want to to look love Runway ml to create 30 second promo videos for my clients that's Jim Ross from Self Storage from Three Mile
257. 46:06Storage uh Jeff Flanigan I'll be introducing AI to a h School nearby in a few days parent teachers fully on board that's awesome that's
258. 46:19fantastic um all right can you explain the $450 thing mentioned earlier oh yeah uh hang on
259. 46:35um search 01 $450 should do it yeah
260. 46:52[Music] sky- T1 open L large language model
261. 47:07beats 01 preview Sky T1 an open reasoning large language model trained with just $450 of large language model I mean I
262. 47:18assume that means trains with $450 of compute um to pate's point it is highly likely that this is a fine-tuned model based on
263. 47:33something else that costs significantly more than $450 right so this isn't just but um whatever they did here they've got a 32
264. 47:44billion parameter model which in in large language model SI oh wait I'm not sharing this hang on damn
265. 47:52it maybe share your screen Vicki needs a shot collar I tried jelly pod this weekend Brandon would be proud very nice
266. 47:59um yeah so here's the sky T1 beats 01 preview in in some benchmarks
267. 48:13right cost them 450 bucks to fine-tune this thing again I don't know I don't know a ton about this but a 32 billion
268. 48:22parameter model is small right big language models are 405 billion parameters 32 billion is is the size you
269. 48:32can you can start running these things locally on your hard drive right so imagine having something as good as 01
270. 48:40preview which 01 preview had a an IQ of about 120 I think 03 has an IQ of about 157 um so imagine having a you know 120
271. 48:52IQ problem solving super knowledgeable entity running off your local hard drive you don't have to go out on the internet
272. 49:00anywhere so that's that all right okay I guess that shot collar didn't arrive in time it didn't yeah that's impressive isn't it
273. 49:14trench I know it's it's Bonkers okay let me go find something else here I want to go find I want I'm trying to think which
274. 49:21one I want to read first okay I know which one I want to read first there were two two posts I read today backto back
275. 49:34that I've kind of known what's coming with autonomous agents but they they really solidified for me that something different is we're
276. 49:46we're about to embark on something quite different so let me go find the one so I'm on David Shapiro's at Dave
277. 50:04shap sh a pii Shapiro just Dave shappy there's Ethan mullik there's that one is this no that's not it it's this one is this it
278. 50:20yeah all right so I'm going to push this in close so you guys can read
279. 50:36okay so um Kim anismus chubby who's a who's a AI commentator and she said you know the
280. 50:51news that meta wants to replace mid-level Engineers is spreading we're taking six we're talking six figure
281. 50:57salaries here people and meta wants to replace them completely and this was from Zuckerberg probably in 2025 we at meta
282. 51:06as well as other companies that are basically working this are going to have an AI that can effectively be a sort of midlevel
283. 51:15engineer that you can have at your company that can write code so that's Zuckerberg just talking about you know what they're doing at
284. 51:23meta all right can everyone see do I have everything set up correctly I think I do I'm on a sec and then David Shapiro responded to
285. 51:34this and I want to read through his post because it there's a level of clarity in it that I really appreciate um but it's but
286. 51:46it's it's a big deal it's a big deal if you don't know David Shapiro he has been an AI researcher for years he he has
287. 51:55been you know ay that does you know highlevel systems troubleshooting for his day job and wrote five books on AI
288. 52:03he's done a number of uh cognitive architectures open- Source projects to do AI cognitive architectures and about three or four
289. 52:14months ago he quit Ai and he didn't quit AI like um I'm just done with it I'm out of here screw you people he quit I like they've
290. 52:28figured it out that based on his knowledge and knowing what's coming that he realized that him doing anymore research on this was pointless
291. 52:45because they basically figured it out and they've gotten it right and if if you listen to the engineers coming out
292. 52:50of open AI right now they're no longer talking about AGI they're not they're now talking about ASI right Sam Alman basically said we we now
293. 53:01know we're not speculating we know how to get to AGI and what we're up to now is getting to ASI and if you don't know those two
294. 53:10acronyms AGI artificial general intelligence where the AI is as good or better than most people at most economically viable jobs that's
295. 53:20AGI ASI is super intelligence where these things become smarter than all the humans you you know the best possible human in every possible
296. 53:30category super intelligence right so so we are on a path right now for that so here's what here's what Shapiro wrote in
297. 53:41response to meta is going to replace midlevel and Below Engineers with autonomous agents so David Shapiro
298. 53:51let me lay this out fresh because after years of building cognitive architectures I've learned something fascinating about
299. 53:58intelligence versus agency everyone's hyping up AI agents like they're the next breakthrough but they're missing a
300. 54:04crucial piece of the puzzle that took me way too long to figure out here's what threw me for a loop raw intelligence
301. 54:12isn't the same as knowing what to do about it think about it chat Bots were relatively straightforward because they
302. 54:18were just waiting for humans to start the conversation give them a prompt they'll Dazzle you with their smarts I
303. 54:25call this the genius in the jar phase but true agency that's where things get spicy okay so we're in the genius in the
304. 54:34jar phase with current llms right they're super smart but they require us to deal with them and I and one of the
305. 54:41things that I hear unable to verify hang on sorry I'm having H I'm having uh Tik Tok is making me do physical dexterity
306. 54:51Mind Games okay I did it great he quit because he's suffering from from an auto autoim immune condition as well
307. 55:01but AI helped fix it yeah exactly exactly well it's funny Danielle he quit AI but like all he talks about now is AI
308. 55:09so he's moved from researcher to commentator uh which I think is that makes sense to me okay but true agency that's where things
309. 55:19get spicy the real Cha challenge hit hit me one day on building cognitive architectures you're sitting there with
310. 55:26this this incredibly powerful system that can think about literally anything and you suddenly realize wait how does
311. 55:32it how does it decide what to think about it's like having having infinite processing power but no executive
312. 55:40function as a card carrying ADHD member with no executive function I I resonate with this so current ai ai is like add
313. 55:50right all this capability no way to focus it that's hilarious um uh okay the human brain handles this naturally but programming it that's a
314. 56:05whole different ball game you'd think that solving the what Tod do problem would be straightforward with enough
315. 56:11computing power right wrong I've had to dive deep into ethics and philosophy just to create a basic decision
316. 56:19framework utilitarianism uh theology deontology um thrown uh throw all that in and you've barely scratched the surface because then you
317. 56:30need to build out the entire model of what an agent can do what it can't do what it should and shouldn't do and most
318. 56:38importantly why that's the agent model layer okay that's important for the next thing I'm about to show
319. 56:47you it's got to know what I am and how do I work before it can start even using its digital hands I call this the
320. 56:55constraints capabilities in context or 3 C's of the agent model um and just when you think you're getting somewhere you
321. 57:03hit the broader context problem your agent needs to understand its environment available tools specific
322. 57:09tasks access levels and we haven't even touched on goal tracking and T task prioritization yet so again this is a
323. 57:17guy that's built these systems right it's like trying to build Jarvis from scratch and realizing why Turing never
324. 57:25cck crack this part of the puzzle Allan Turing the Turing test you know I I'll know uh computers have reached intelligence when I can have a
325. 57:34conversation with one and not be able to tell the difference if it's a human or if it's a machine right that Turing um
326. 57:41this is why Turing never cracked this part of the puzzle which is the how do you use you know this intelligence
327. 57:49resource utilization keeping track of what you need and have and how close you are to solving the actual problem llms
328. 57:57can do each of these steps no problem but coordinating it all a bit harder that's what makes today's llm so
329. 58:07interesting let let me see see there's comments here Shapiro pioneered the hierarchical framework well for me in
330. 58:13which the agents have an upstairs office that deals with the human in the loop yeah exactly yes AI can write code but
331. 58:20that AI that writes code cannot do my job it doesn't have the ability to insert itself agentically into what
332. 58:28already exists exactly okay um what else do we have here any other comments okay that's what makes today's llm so
333. 58:41interesting there's still narrow AI in their operation but they're packing some serious general intelligence so when he
334. 58:48talks about there's still narrow AI what he's talking about is classical AI you would take a narrow task like beating a
335. 58:58chest Grand Master or diagnosing one specific kind of cancer and you would train an AI on that it would just have
336. 59:05this narrow expertise what David's saying here is even though today's large language models have this general
337. 59:13knowledge capability they're narrow AI because they still require us to figure out what to do with them okay um there's
338. 59:24still Nar narrow AI in their operation but they're packing some serious general intelligence features under the hood
339. 59:30World models reasoning planning problem solving it's all baked in but true agency that's still the Holy Grail we're
340. 59:38chasing true agency these things being able to do it on their own that's the Holy Grail we're chasing and it's not
341. 59:45necessarily just the model it's the rest of the body so to speak okay think about it if your brain was just floating in a
342. 59:52jar somewhere no eyes ears or hands you couldn't really do anything either to make matters a bit more confusing AI
343. 1:00:00agents exist in cyberspace where what they see and hear is just too different from our meat space context right and
344. 1:00:09their hands are API calls so he's basically saying like we don't know how to use we we don't know how to give
345. 1:00:15these brains in a jar bodies right okay and then here's how he closes it up look here's the punchline that most
346. 1:00:24people haven't wrapped their heads around yet we cracked this problem this is one of the reasons he
347. 1:00:39quit AI here's the punchline most people haven't wrapped their heads around yet we cracked this problem those years of
348. 1:00:48research into cognitive architectures and agent Frameworks weren't just academic exercises we were building the foundation for some
349. 1:00:56exponentially bigger than anything we've seen before the real kicker once you've got one working agent system you don't
350. 1:01:05just have one you've got the potential for billions so take take it in so what he's saying is we've got this brain in a jar
351. 1:01:18what we haven't figured out well the hardest problem to figure out is how do you give that brain in the jar the body
352. 1:01:26the ability to go do [ __ ] that's the part he saying we've cracked and he says if you crack it for one agent you've
353. 1:01:34effectively cracked it for all of them okay um the same agent perfectly replicated operating at scale across
354. 1:01:44every digital system on Earth copy paste repeat we're not just talking about replacing a few mid-level Engineers here
355. 1:01:52and there we're talking about a tsunami of artificial EX executive function that's about to reshape every
356. 1:02:00industry every workflow every digital interaction um the people tweeting about 2025 aren't being optimistic if anything
357. 1:02:11they might be understating it just how fast this is going to move once it starts the dominoes are already
358. 1:02:18following and trust me you're going to want to be ready when they hit every we should all print out this paragraph and like hang it by our
359. 1:02:31desk they've already cracked the problem and once they release it and if they do it for one agent they've
360. 1:02:42effectively done it for for all the agents okay questions thoughts it feels like the universe just got a lot bigger yeah I we'll wait till
361. 1:02:54you see what I show you next next I have convos with GPT all the time it knows a lot but it doesn't provide
362. 1:03:02insights that we can well don't so listen sacred is profane don't assume that that's going to be the case for much
363. 1:03:14longer like that's exactly what he's talking about here his whole point is that these things are smart but they're limited
364. 1:03:23because they require us to make that kind of decision making and insight and understanding that's about to change and
365. 1:03:31and it's about to change kind of instantly and overwhelmingly it's not like you're going to get one agent and then three
366. 1:03:39months later you'll have 10 agents and then no this is going to be like once it happens it's just they're going to start
367. 1:03:45to proliferate and then as those agents start figuring out the other kinds of agents that need to be built and the
368. 1:03:50other kind of Frameworks that need to be built they're going to prol proliferate even faster okay got it cool right it's a big deal
369. 1:03:59it is a big deal Jeff Flanigan it's a big deal okay all right let me go find this other thing I want to
370. 1:04:17find this is Robert scobble
371. 1:04:33he forwards a lot of [ __ ] it might take me a second to find
372. 1:04:53this is it did he do it as a [Music]
373. 1:05:10okay this one okay if you don't know who Robert scoble is he he's been an apple Fanboy for years he's a San Francisco based you
374. 1:05:22know uh technology reporter and and and you know he's he is deeply entrenched in the scene in Silicon Valley he was there
375. 1:05:34for the first I think he said it Siri was launched in my home um I live streamed the first ride in the first
376. 1:05:41Tesla ever right like this is a guy that's connected so what he writes this will be the most important video I've ever
377. 1:05:53done this is a guy that's been making videos and making posts and and and being on the Forefront of what's coming
378. 1:06:01from a technological standpoint for like 35 years that first sentence that doesn't land lightly on me because he's not a
379. 1:06:11bullshitty kind of guy like if he thinks something sucks he thinks it sucks now when he goes all in on something like
380. 1:06:16he's all in on on uh on the holck and you know and Google or Apple Vision Pro and all that sort of stuff so there's
381. 1:06:24things he goes all in on that that you know you may or may not agreee are going to be a big thing but he's not a g guy
382. 1:06:30that says stuff like this too often so this will be the most important video I've ever done and again in the context
383. 1:06:36of I had just read the David Shapiro post and then I read that sentence and I'm like what the [ __ ] is going on
384. 1:06:44History will be the judge but I'm pretty good at figuring out the future each of my seven books properly predicted decade
385. 1:06:52Long's Trends my first on social media was written before Twitter X even started keep in mind Siri was launched
386. 1:06:59in my home I have streamed the first ride ever of of the Tesla of the first Tesla um I've had many other firsts but
387. 1:07:06this is the beginning of a new operating system the guy that created Amazon's Marketplace which is the main reason
388. 1:07:14Amazon is profitable company today this will be a hugely important technology and you could only learn about it here
389. 1:07:23the Accenture booth at CES was not open to the public you had to be invited in I am the first to take you behind the
390. 1:07:30scenes and Jensen announced Nvidia was partnering with Accenture on robot orchestration this is it so CES report
391. 1:07:39video 28 so what I'm going to do here is I'm gonna change my sharing so that you can hear this and we're gonna we're
392. 1:07:46going to listen to this video and if anyone's still here you don't have to still be here if if you're not here I understand
393. 1:07:57um I think this shit's really important you know to the to the David Shapiro thing be prepared like how do
394. 1:08:05you be prepared I don't know you watch [ __ ] like this and you try to get your head around it like I can't quite get my
395. 1:08:12head around this the guy that we're about to meet not this guy but the skinnier guy he talks to in a second is the guy that wrote dos like
396. 1:08:21the dis operating system that Microsoft bought and turned into Microsoft that's the guy who started this the company
397. 1:08:29that they're talking about right now that guy right there that's him all right so here we go we're backstage at the Centra this is
398. 1:08:42the Cs that you never usually see how the Nerds are running the world and uhler I'm uh one of the lead
399. 1:08:52nerds x Microsoft Amazon and I'm the COO of uh autona and we do the paranet which does all the orchestration for the demo
400. 1:09:01that you're going to see a little bit later so so hold on a second because perinet I I got a little I drove to
401. 1:09:07Seattle to see pinet and understand it this is a new network protocol that arranges or organizes orchestrates
402. 1:09:15drones robots autonomous vehicles and human beings and human beings we pull human beings and you're going to see
403. 1:09:21that in the demo later where Allen will actually fault in for a robot the robot will fault in for Allan seamlessly so
404. 1:09:28this is a distributed asynchronous operating system Jim Hardy our founder over here who you seen on a present
405. 1:09:35video I met with him at his house Philip over here is the brains behind the operation yeah Brian's running all the
406. 1:09:42robotics today on our side of the house yeah we've got our great Partners agility robotics over here yeah the
407. 1:09:48coolest robots you're ever G to see this is I badass on air but I just did now the the the thing we're g to see out
408. 1:09:56front is a warehouse of the future right it's a work cell in a warehouse of the future and then we're going to also see
409. 1:10:03the real life version of that being orchestrated then we'll see an Nvidia Omniverse version where that starts to
410. 1:10:09zoom out and integrate multiple warehouses yeah this has impacts on not just Warehouse we were talking about the
411. 1:10:16fires Malibu that just burn down a community right sorryy and you guys still Brian has a GEOS background we did
412. 1:10:26a bunch of aerial Imaging we used to fly uh for FEMA and other First Responders and so what we're going to be partnering
413. 1:10:33with coming up in the next year or two is drones and drone manufacturers to be able to do that distributed get there
414. 1:10:40where you need them yeah at the time you need them same with hurricanes we used to fly hurricanes right after the
415. 1:10:45aftermath and you've got no infrastructure so how do you get infrastructure quickly right starlink is one answer but then starlink to drones
416. 1:10:53trying to help the first respond so the conversation going on right now on Tik Tok about you know where the
417. 1:11:00women nerds are asking the same thing where the where are the women at yeah the Nerds just want to date right um I
418. 1:11:07was just thinking how would women influence this scene so he here's what one of the things I am most excited
419. 1:11:16about with the AI Salon community and this community here um is the generative I seen so so you got to realize these
420. 1:11:27guys are sort of well you'll see here this thing that they're talking about is at the protocol level like at the IP
421. 1:11:36protocol level it's just it's more of a bro scene do I wish there were more women in it yeah I wish there were more
422. 1:11:42women in it but the generative AI thing that requires empathy to be good at it is where I think women are kicking
423. 1:11:53ass the fact that we had three women representing the AI salon and educating our politicians about AI today makes
424. 1:12:01like makes my heart really warm but these are very different layers the layer that we're talking about here is
425. 1:12:06this fundamental protocol level we we'll get to what this is about in a second it's only 15 minutes so we'll we'll they
426. 1:12:12they get into they're just kind of shooting the [ __ ] right now but they start getting into what the paranet is
427. 1:12:18here in a second but uh yeah we need more [ __ ] women in it but anyway I like I I think that because generative
428. 1:12:28AI is a is a language and empathy like if you are more empathetic you're actually better at generative
429. 1:12:37AI than people that are not and I think this is an opportunity for people that have good empathy skills to really step
430. 1:12:45into their [ __ ] power and that that I'm super excited about so so let's let's distinguish that a bit but yeah is
431. 1:12:52is the tech industry still bro [ __ ] riff yep it sure is this par yeah turn up the volume on this when I start again
432. 1:13:05sure though has impacts on everything from a music festival that's going to need autonomous cards to get to it a
433. 1:13:13warehouse where you're going to need to control a bunch of robots and humans working together what what else does
434. 1:13:20what does paranet do that is really magic here and what why am I here because I'm here wait Shane Shane saw
435. 1:13:28said I love working with creative women then he drew an arrow pointing to himself Rel reluctant empath I'm telling
436. 1:13:35you man the the the again it's like revenge of the liberal arts Majors re like if you're good at if you're good at
437. 1:13:41people you're gonna be good at this stuff okay it's more like CIS rific Yeah well yeah that too um us wom women are
438. 1:13:52infiltrating it okay here we go so now he's going to talk about what the paranet is yeah paranet p a r a NE T and
439. 1:13:59think internet paranet you you'll you'll hear in a second this is way down at the protocol level is what these guys are
440. 1:14:06building but once you understand what the [ __ ] they're talking about combined with what David Shapiro just said all of
441. 1:14:12a sudden you go holy [ __ ] like this is coming and it's coming fast okay here we go wait who's the guy on the
442. 1:14:22salon ton skier who's the guy on the salon talking about discovering ai11 uh I don't know he he he left there was he was he he he
443. 1:14:33was not uh he was not living the values of the salon so he's no longer there I don't know what he was talking about and
444. 1:14:39he was not willing to share it so that that's that's no longer uh an active conversation see orchestrating all of these B thing it
445. 1:14:52does is it can orchestrate in a heterogeneous as almost anything so this is the critical piece right it can orchestrate in a heterogeneous
446. 1:15:04fashion almost anything okay go on humans robots computers Enterprise workflow when you're at that you know
447. 1:15:13Coachella and youve got bottlenecks of people trying to come through a facility right well why wouldn't you just
448. 1:15:18automatically open up other Gates that actually can have like CV security and just monitor the flow and move it back
449. 1:15:25back and forth instead of bottlenecking the way it happens to right so so he they'll get more into this but
450. 1:15:33like what he's talking about here is imagine if all of the systems of the world that includes humans and systems
451. 1:15:42that don't normally talk to one another what if those systems could all talk to one another dynamically that's what they're talking about
452. 1:15:52here okay I'm not a big Coachella I don't look like no I went and you know they have 3,000 Uber cars driving people
453. 1:15:59around with humans someday soon there's going to be autonomous networks right U whether it's or or Tesla your roboxing
454. 1:16:07right I'm imagine sech pricing except for you're not your goal isn't necessarily to cost you know charge more
455. 1:16:13but to in fact get more people through yeah make more people happy what what is so what what is it actually doing and
456. 1:16:20why is this new kind of network protocol really it's a smart net turn it to Jim for hey Jim come on over what why what how does this work
457. 1:16:31and why do we need a new network protocol for this new robotic autonomous ah that we're heading into so this
458. 1:16:38there's like 9,000 languages right yeah it's been documented there's 4,000 protocols um so 9,000 computer languages
459. 1:16:474,000 protocols so think tcpip HTTP those are all protocols 9,000 languages 4,000 protocols what's missing has been this this
460. 1:16:57protocol stack called The OSI stack with seven layers on it forever and we know that internet really is insecure we have to kind of plug and
461. 1:17:06patch things and buy a bunch of software it's $9.5 trillion dollar worth of cyber crime in the world which is just going
462. 1:17:12to grow with AI so we saw this as a problem we saw that it's not just going to be humans using the internet anymore
463. 1:17:19it's be machines using the internet P Mo how many of those are in use 9,000 is silly well 9,000 is silly
464. 1:17:28Pate until you've got like some obscure language that's running some control system at a nuclear power plant um that
465. 1:17:37needs to be interfaced with that that's that that's what he's talking about is the ability to interface th the ability
466. 1:17:45to coordinate things across domains um that that currently doesn't exist and do it in a secure way machines are using
467. 1:17:51the internet how are they talking to each other so what if you create a new new protocol but a new layer so what we
468. 1:17:58did most is what most people don't realize is when the web came out the internet had been out since 1983 tcpip
469. 1:18:04right and we had email SMTP all these protocols when the web came out HTTP it allowed you to transfer HTML from a
470. 1:18:12server to a client and the rest is history there's billions of people using these these things and all kinds of
471. 1:18:17other Protocols are out there for doing you know iot and modbus mqtt and so forth and those are all good protocols
472. 1:18:24they all nothing changes so the IP infrastructure and the internet as it exists today nothing changes but it's
473. 1:18:31broken from security and Trust standpoint so we said let's create a new overlay an overlay like the web is an
474. 1:18:38overlay when you use zoom it's an overlay only you and your friends are talking in Zoom no one else gets in why
475. 1:18:43can't we create a network overlay such that machines can be up there so we did that but then what are these machines
476. 1:18:52but we actually had the yeah this gym is the DUS uh create something like the original operating system like dos you know yeah
477. 1:19:00which you were involved in yeah and um a couple years after that uh Andrew S tanon who wrote the book computer
478. 1:19:08networking also wrote a sacred is profane this is stepping on a lot of toes it's it is stepping on a lot of
479. 1:19:15toes but agents are going to step on a lot of toes as well and and yeah I I agree what what what you're talking about is
480. 1:19:25there there are massive Industries built on top of the insecurity of the current Internet and if you all of a sudden
481. 1:19:36throw in autonomous agents that can go take action on everyone's behalf that's where that security vulnerability starts
482. 1:19:44to become completely just unimaginable how bad it could be and that's that's what he's trying to attack here a
483. 1:19:52textbook called distributed operating system there aren't any distributive operating systems so what we needed well something
484. 1:19:59like this could be used to start World War II I think what he's talking about is this could also be used to prevent
485. 1:20:05was this huge history of the internet IP infrastructure and for somebody to come along autona and the paranet to build
486. 1:20:13that overlay so now we can have distribut operating system what that means is you have to have new machines
487. 1:20:18we're not taking new CPUs or gpus like that we're talking about a new machine it's called an actor and this stuff goes all the way back to
488. 1:20:25MIT um and the early pure thinkers of of AI they thought about actors and planning languages so we took the idea
489. 1:20:34of actors and said okay what if every machine every software every API every robot every web page every database
490. 1:20:40every Everything could be an actor like Paul was talking about with you know World Management systems and robots and
491. 1:20:46so forth he was referring to we need to get all these things on this new secure trusted Network again it's the Internet
492. 1:20:52it's IP infrastructure but it's the paranet paranet now is this overlay and machines only talk to each other through
493. 1:20:57this protocol now even though they're written in C or python or whatever doesn't matter we can quickly aoriz them
494. 1:21:03put them on a network and when they're on that Network they can securely talk to we can quick quickly aoriz them so
495. 1:21:08basically what they're talking about is creating like a metadata wrapper that points to something doing something down
496. 1:21:15at the at the a lower level in trust we had had a b did a bunch of things around security that we can talk about another
497. 1:21:22time but yeah you want to keep Iran so if if you're all if if anyone's new here I see Jason came in who this
498. 1:21:30and what what have we been talking about tonight so what we're talking about tonight is David Shapiro's basically said that
499. 1:21:37that the the main problem with agents is how do you take this smart brain this llm this brain in a jar that's not
500. 1:21:46really connected anything and right now currently requires our expertise to make it work and how do you solve the how
501. 1:21:54does it decide what Tod do problem and what Shapiro argues is that problem has effectively been cracked and once that's
502. 1:22:01released on the world once you solve that for one agent you've effectively solved it for all agents so that was the
503. 1:22:08first thing we talked about this is this is a guy that has been in the business for years effectively proposing or has
504. 1:22:17built a protocol lay layer called the pro the paranet that sits on top of the current Internet technology stack and is
505. 1:22:27a coordination layer that coordinates between objects and functions of everything that's happening on the
506. 1:22:34internet so that's that's what he's talking about out of your network yeah but but yeah we want to problem with the
507. 1:22:39whole internet right now if something gets on your on your network if it gets through zero trust security you're done
508. 1:22:46right you have to start over right the paranet says if you're using a protocol if a no gets infected the network
509. 1:22:52realizes it's bad and gets rid of it so we call that so that's that's an important thing right there right in in
510. 1:22:57in the current Internet if you're hacked and someone gets into your system it's done what he's saying is if you're
511. 1:23:03operating at this paranet level and someone grabs a node the system will be smart enough to know that that node is
512. 1:23:09[ __ ] up and it'll actually just shut it out of the the the Swarm negative trust security so to answer your question why do we need a
513. 1:23:19network because trillions of machines are coming yeah right we're going to be machines right and they have to talk to
514. 1:23:26each other and they all have to work together but the only way to do that is through an overlay and a exclusive protected secure trusted uh
515. 1:23:35collaboration protocol and this protocol is pretty cool because it actually does um Q&A so machines are doing Q&A with each other
516. 1:23:44so everything's getting get more intelligent py don't out every single thing he's saying so so here's the thing
517. 1:23:51this is likely a private company he's trying to spin up a protocol layer but this is clearly a commercial entity
518. 1:23:57right so whether or not this thing ends up being a protocol whether or not this thing ends up even launching I don't
519. 1:24:06know like I don't know where this is the reason I'm showing this is that there is fairly significant work
520. 1:24:14being done on the orchestration layer so if the current um Frontier Model companies solve the how do we use this
521. 1:24:23smart brain we have an ajar problem and you have something like this that can start to coordinate it things are going
522. 1:24:29to accelerate very fast he's spouting nonsense Kyle I have actual credentials for this type of thing yeah but it's
523. 1:24:35it's not binary just because he said dos was the original operating system I assume he meant for for PCs rather than
524. 1:24:43mainframes and and Minis and things like that I get your point but don't throw the baby out with the bathwater because
525. 1:24:49it's it there's this isn't necessarily just this specific guy and this specific company there are probably lots of folks doing this
526. 1:24:59but one of the questions I've always had is if you have if you introduce billions of things onto the internet with lots
527. 1:25:07and lots of different protocols how do you coordinate all that and so that's what that's okay so whatever if you if
528. 1:25:17you want to listen to pate and what I'm showing you is completely useless all the AI and agentic Frameworks all that
529. 1:25:23stuff you hear by agents they all need some place to talk robots need a place to talk but the weird thing about robots
530. 1:25:29is that they they are in the OT side of the world not the it set right OT is where all the hard engineering goes the
531. 1:25:35electrical engineers they now need to place well how do I get how do I put this in your Warehouse how do I put this
532. 1:25:39in your factory yeah well you got to do Network you security no you don't just put on the paranet it discovers
533. 1:25:45everybody else out they're all secure and they run so that's that's this is like air traffic control for the
534. 1:25:50internet well it's like air traffic control above the internet like yeah it's not a bad it's not a bad thing
535. 1:25:57where you've got you know planes landing on the internet right so you've got robots and software and all sorts of
536. 1:26:04things doing all this stuff down at the internet level and this is saying how do you get those things to talk to one
537. 1:26:11another it's the paret in a nutshell wow this has deep implications on the future of everything doesn't it well
538. 1:26:19because because as humans just like s TP had exactly our original thoughts SMTP basic protocols but start with security
539. 1:26:32everything is secure from the beginning to the end sorry Jim email was never secure uh the web wasn't really secure
540. 1:26:39until they put s on it but the horses left the barn so bad things happen but you need more than just encryption you
541. 1:26:46need a trusted Network that's intelligent you need you need security actors on your network that are looking
542. 1:26:51at all the other actors doing the work so you need to be able to have have a programmable Network that's intelligent
543. 1:26:56that can know what's going on one of the biggest problems in networking particularly in data centers is just setting up routing
544. 1:27:03tables there's millions of hours every day being spent setting up your router tables to your switches your network on
545. 1:27:10the paret and this new protocol of course you still have routing tables you have routers and your switches but
546. 1:27:14you're on this new semantic layer now now all the actors these new machines they don't go through routing tables
547. 1:27:19they just say here's my skills here's what I could ask and then all the different nodes with all the actors
548. 1:27:24skills they all discover each other and no one gets to ask a question they're not they're not able to ask yeah and
549. 1:27:29that's how you create security and Trust going forward now I'm I'm here because I'm trying to get ahead of Elon
550. 1:27:35Musk yesterday I met a a voice company and he immediately announced a new roock with multi-party voice to respond right
551. 1:27:45and uh it's hard to get ahead of that man right um let's say he's building a paranet of his own to run his autonomous
552. 1:27:53car now and his robot yes so he has a vertical why does he need to throw all that away and come and see you no one
553. 1:27:59needs we're not seeing through any overlay augment yeah so why would he need to come and see you and learn about paranet
554. 1:28:10learn about this new network protocol to integrate with the the Tesla system he's building so Tesla is um he's all about autonomy right so
555. 1:28:21he's doing humanoid robots Optimus he's doing you you came here in a in a robot car I did my robot drove me here so that
556. 1:28:29that's on a network but the thing is nothing that humans build is truly autonomous it has to be on some sort of
557. 1:28:36network with information orchestration collaboration workflow planning and observation right yeah so we built a
558. 1:28:43generic version of whatever he built that has those five properties again orchestration workflow collaboration planning and observation
559. 1:28:52so we build a network that you can do so you could build the Tesla Network on the paranet you don't need to because he's
560. 1:28:57already done it but if you need to build an autonomous system for your factory for your Warehouse yeah someone just
561. 1:29:02made a comment about beta VHS versus beta yeah I mean this is so so here's the thing this is a guy that's that's
562. 1:29:12talking about you know building the protocols to be able to make VHS right and Tesla's doing the same thing he's
563. 1:29:20done over at or Elon Musk is doing this over at Tesla those are going to be competing protocols does this guy end up
564. 1:29:28winning I don't [ __ ] know Does Elon Musk end up winning he'll probably at least be in the game right um but
565. 1:29:35someone has to crack this layer and and what what I was surprised by is how far how far down the road this particular
566. 1:29:44version seems to be that's that's a generalized protocol now one of the hardest things to do in the world is introduce a new protocol
567. 1:29:54right how do you get people to adopt your protocol well you know that's VHS versus use us there's GNA be a lot of
568. 1:30:00failures use tless Network yeah I don't know what he's using for his humanoids but he probably should use something
569. 1:30:06like the paranet yeah because we built it to allow robots become actors in the network the thing that people don't
570. 1:30:13assuming he has a paranet himself because he's building stuff like that to have his robots work together and his
571. 1:30:20autonomous cars work together and figure out where they're supposed to be to pick up a ride or stuff like that can he hook
572. 1:30:27his paranet in to your paranet and now they could ours level ours is Lay layer eight okay I'm just going to say it when
573. 1:30:35you do the layer above seven it's eight here we are we're at this layer he's not doing layer eight he can certainly plug
574. 1:30:40into that yeah right so and that explains how a Ford could join their robots into the Tesla Network into the
575. 1:30:48paranet right working with the robots I'm about to see up in front of the H right yeah and so then you know it's
576. 1:30:55just not going to happen overnight yeah so we're partnering with Accenture right so so here's the hey we're just a humble
577. 1:31:03startup trying trying to write a new protocol this is not going to happen overnight means that this is somewhere
578. 1:31:09between vaporware and something right and if you listen to all the stuff that c c talking about CEs it's all about
579. 1:31:15autonomy and Ai and trust and people working with AI systems together and agents well that's the parent I mean
580. 1:31:22that's what it is so that's why we're partnering with them working through this and it's very early like we just
581. 1:31:27you know yeah we don't there's only a few thousand human ey BR that's there will be million many millions yes soon
582. 1:31:37if you ask if you yeah so his company I don't I forget what his company's called his company is uh has partnered with
583. 1:31:45Accenture so this is this is backstage at CES two days ago where Accenture did this Warehouse of the future demo
584. 1:31:55that all of the robots were Co coordinated on this paranet um protocol yes the musk it's even more than that I
585. 1:32:03I I did ask bus I asked him for his spreadsheet actually predicting how many robots he's gonna sell and he said you
586. 1:32:11don't need a spreadsheet it's gonna be so [ __ ] many of them because we were talking about people with bad needs and you know just
587. 1:32:20that one little tiny Market 60 million people without bad needs they need a robot to help them out right just that
588. 1:32:26one little sliver so I wanted that's FR he has to predict all the markets right because he has something like that he
589. 1:32:33has he has grock you know that knows everything at least does know it yet but it soon will oh as soon as I post this
590. 1:32:41video grock will know all it needs to know yeah soon as you put it up that's an talk to Gro what does Gro need to now
591. 1:32:49you just needs to ask hey what's what should what should I be doing with the baronet needs to know what it needs what
592. 1:32:55does grock need to know to to hook into the parad yeah so what what are the API calls that grock needs to
593. 1:33:03know more to come we publish a dev kit grock will be all over yeah and uh perinet is we're on the web with grock
594. 1:33:13fan when we publish a dev kit so they don't have a dev kit yet so this is I assume this is really well thought out
595. 1:33:21probably some papers probably some stuff that not thought out to P's Point um probably spitting duct tape at this
596. 1:33:29point so they so I don't know how how far in they are paranet you'll find it at pet. you know and aon.com autonoma
597. 1:33:39doccom paranet doai and autonoma Doc O N O ma a okay and does this cost money to integrate right now what's your
598. 1:33:53business model right now you're in the back room on the Cs right you're in the back like this where you kind of this is
599. 1:33:59the dirty room make they never like to let the press back here because it's dirty not supp talk see he's even
600. 1:34:06shaking his head we're working through various models right now right so what we want to do is some combination we
601. 1:34:11don't know our pricing open source and Licensing such that we make it easy for the world to use this one of the things
602. 1:34:16we did was we we went and we got arpanet doai and so we want the researchers of the world all the universities to be able to use
603. 1:34:24freely and arpanet was really the beginning of sharing information across the world arpet will be distributed
604. 1:34:32Intelligence on a paranet network that's run by researchers so we're going to make that free everyone and then um for
605. 1:34:42paranet it's really kind of like a licensing world with certain amount of bandwidth on the network haven't worked
606. 1:34:49out that perfectly yet but basically we have a distributed Network operating system every not is out there if you're
607. 1:34:54using it commercially that'll be a license and then how much bandwidth how many actors with how many skills are
608. 1:35:00being you know interacting at any point in time kind of like a network fee so still working through that still early
609. 1:35:06we haven't got our day kit out we'll have that in a few months and then you can put that in a Gro yeah few months
610. 1:35:11from a dev kit right so so these guys are early so so because they have a good team they partnered with Accenture they
611. 1:35:18they might even have partnered with Accenture to help build some of this [ __ ] so um so anyway that's what's going
612. 1:35:25on all right enough enough comedy jokes so there you have it I think those two things back to back you know on the same
613. 1:35:38day they were significant for me here here's the here's the implication independent of whether or not perinet as
614. 1:35:45that guy described it is his thing or some other thing one of the things that I have not been able to wrap my head around is how do you
615. 1:35:55introduce millions or billions of autonomous agents into an insecure environment like the internet without it being absolute [ __ ] chaos
616. 1:36:07um it's it's got to be solved and I you know I think individual companies are going to solve it on their own so open
617. 1:36:14AI is going to have their version grock's gonna have its version um you know or you know Tesla and all that are
618. 1:36:22going to have their version is going to have their version the Chinese are going to have their version how do all those
619. 1:36:27things talk to one another this is the first thing I've heard that I start to say okay I can at least see conceptually
620. 1:36:33how you could create a layer above what exists right now which feels like an abject [ __ ] nightmare um all
621. 1:36:46right Windows is still basically the dis operating system uh uh sacred is profane you may be splitting
622. 1:37:02hairs paint for better or worse that's my whole stick I mean listen paint is it's it's all about definitions and Clarity I am
623. 1:37:18about muddiness and what's possible so I think that's why having p here's a really good voice a really good balance
624. 1:37:25to my to my optimism and and fuzziness but um here's the point my the point and the reason that I shared let me go back
625. 1:37:37and show you where it is the reason that I shared this document so if you go to the
626. 1:37:52AI salon I put this in in I put a post of this in um the code showand tell section I put this in Pates um in Pat's uh club and I put this
627. 1:38:08in the water cooler conversation so this is a document that I did today using goo Google deep research of um AI
628. 1:38:20agents broken into three categories no code low code and um and developer focused platforms um you don't necessarily have to go
629. 1:38:34build agents you don't has have to necessarily um dig deep into this but I I feel that right now just like building custom gpts was a
630. 1:38:46good augmentation to understanding prompting and chat GPT in general understanding what agents are as they are today and understanding what
631. 1:38:56their limitations are today is going to allow you to be as prepared as possible for the thing that David Shapiro described which is
632. 1:39:07basically once one shows up The Dominoes fall very very quickly and it seems to me unless everyone's [ __ ]
633. 1:39:19lying that that first Domino is going to fall in 2020 and it it there might be multiple dominoes that fall in 2025 it could be
634. 1:39:302026 but even if it's 2026 That's mighty [ __ ] close for for a series of Technologies being Unleashed
635. 1:39:39on the world that can do many people's jobs do the jobs not just do parts of the jobs that require the human to be
636. 1:39:48there but do their jobs so so I kind of feel like 2024 was the year 2023 was the year we got excited 2024
637. 1:40:01was the year we got to take a breath and sort of take in what was Unleashed on the world 2025 feels to me like the year
638. 1:40:08where shit's getting serious and so why I wanted to talk about this tonight is I want people to um I I would strongly encourage you if
639. 1:40:19you haven't been paying attention to agents just start paying attention to them um do your best to sift the wheat from
640. 1:40:26the chaff and if you've got a little curiosity and geekiness in you come play with this list and just go look at what
641. 1:40:32these platforms are and what they claim they can do right even if you just see that just what do they claim they can do
642. 1:40:39there's also a bunch of Open Source things out there's there's there's one that was just talked about I think it
643. 1:40:43scrapes Reddit and does some [ __ ] for you that's been open sourced in the past three or four days um they're they're
644. 1:40:50starting to get pretty interesting and if you're one of those people that knows how to turn loose an agent as soon as
645. 1:40:57the platforms are are available you might get a competitive head start on you know whatever you're trying to do in
646. 1:41:04the world so all right with that questions [Music] thoughts I on China the impression I get
647. 1:41:20is gpts went nowhere generally speaking I've still not made a custom GPT I'm introducing a hypothetical agent on
648. 1:41:35Wednesday oh are you for real CEF or or is that just a joke because right now I'm in hypothetical land but I here's the thing
649. 1:41:44I don't think we're GNA be in hypothetical land that that much longer [Music]
650. 1:42:00all right fantastic 2025 the year of disruption yeah it's like yeah to say the
651. 1:42:15least Kyle as I said earlier can we all hide in a blanket fort in your living room for a while Daisy May like that is the appropriate
652. 1:42:25sentiment the appropriate sentiment from where I sit is like listen a lot of what I'm talking about tonight still lives in the realm of chat
653. 1:42:38TMZ right which drives Pate [ __ ] crazy a lot of what I'm talking about is speculation but there's enough indicators
654. 1:42:49that this might not just be hype there's enough indicators is that yeah in 2025 some shit's going to drop that
655. 1:42:58fundamentally changes the game and yeah I feel like an appropriate response to that is I want my doie give me a hot
656. 1:43:06chocolate my doie and a warm blanket I'm just gonna sit out the next three years I think that's an appropriate response
657. 1:43:13um and if you want to be prepared if you want to not be blindsided by this stuff then I think it's I think it's worth
658. 1:43:21putting time into this so there you go Jim Ross this makes my head hurt but it's amazing I know but Jim Ross so so
659. 1:43:29you know okay so where you are right now with lovable dodev so so Jim Ross is if you don't know Jim Ross awesome dude
660. 1:43:40he's got a self- storage business called thre Mile Storage where he helps people run storage businesses right where they
661. 1:43:47own storage unit businesses um and in the past week he sent me two different Publications that he's written with
662. 1:43:54lovable dodev where he's like oh my God like the one thing you said was you wanted to do this a year or two ago and
663. 1:44:01the dev that you met with said it would be a minimum of $5,000 and you did it in a couple of hours right or an hour or
664. 1:44:09whatever it was and then and then you sent me the next application you wrote like the next day right so so this agent
665. 1:44:17stuff right now is head spinny we don't quite get it kind of stuff but at some point there will be some version
666. 1:44:23of an agent spinner uper that'll be like lovable dodev like we will be able to use these things um quite easily like it may take a year
667. 1:44:35or two to get the the systems or you know to get the right company to to introduce the right interface for it
668. 1:44:45um but I think that I think that it's it's it's coming really fast and you know like for your business Jim it could be
669. 1:44:53you know rather than having to hire a manager to go manage all of your different properties you could have you
670. 1:44:59know a suite of Agents go manage all of your properties right so you could literally become a oneman show that's
671. 1:45:06managing you know dozens of properties around the country is possible oh good lord all right appropriate response push the pink bow
672. 1:45:19button you can make money with my gpts are absolute obsolete already um where are you all posting your music but beyond the salon um if
673. 1:45:35it's good enough I put I put Quantum Cipher on Spotify I just put it up there as a song one man show that's the plan exactly
674. 1:45:49exactly looks like Cosmic lovers on a on a roll to get someone hired which is awesome gonna hire from the salon fantastic love that love
675. 1:46:01that yeah I would say you know posting your music on uh on uh on the salon's good I know that there's at least one
676. 1:46:12there's probably like five now I would go on futurepedia doio and see if you can find like is someone creating a Spotify for AI music
677. 1:46:24I personally think that's a limited model because I don't think I don't think it's going to be that much longer that we can actually
678. 1:46:33distinguish between AI music and music like it's e it's either good music or it's not um but I think short term there
679. 1:46:40are probably going to be platforms that are like post your AI music here and we'll create the you know the Spotify
680. 1:46:47for AI music I'm sure there's a bunch of those out there in fact I kind of remember there being one or two of them
681. 1:46:53um all right I'm putting my music on United Artists there you go it goes to a ton of platforms that way I don't think people can tell very
682. 1:47:05well today um have I have I watched megalopolis I have not watched it yet I've been listening to other users music in
683. 1:47:17soono oh um oh what's his name um [Music] Jay um he was in our prototyping meeting today he's using XML in sunno here I'll
684. 1:47:35just show you this quick um where did he send it I think I've got it frostbitten
685. 1:47:55so here's this song look at the look at the prompt for this song or the lyrics can you see that yeah you can here let me make it
686. 1:48:06bigger you can't really see it too good but that's all XML but it's like interval four 4101 drop outro so he's sort of setting
687. 1:48:21up different modules and then he defines the patterns so interval one has those musical patterns interval two has those those musical
688. 1:48:34patterns number one feel Soul interactive module conform and then in the lyrics of the song he's dropping in
689. 1:48:44when he wants different patterns and he says this works that that it's generating songs that basically follow
690. 1:48:50the XML recipe up above so I'll share this in uh in the salon he he might have already shared it there but I think it's
691. 1:48:58kind of cool so I'll go to um music and audio and then I'll go from
692. 1:49:13at frostbitten his zany XML music prompt
693. 1:49:28check out the song and in quotes lyrics and then ask him how the hell
694. 1:49:42he's doing this I had I had chat GPT analyze the XML and it's it's not based on anything that that I know so might be that he
695. 1:49:53just came up with this idea and is having chat GPT create a custom XML schema but that's in the music and audio show in there you
696. 1:50:04go all right people we don't need to notify everyone it's getting late I think it is time time to get
697. 1:50:20rolling rolling rolling rolling any final thoughts questions our students will be loaded up
698. 1:50:35and securing AI certifications through a new AI course in high school very cool Silver
699. 1:50:50Fox you are you are welcome welcome thank you AI machine learning dude frost bitten can't wait to see yeah it's it's
700. 1:50:59cool I I I want to get my head around it but I also you know what's amazing about the frost bit's approach to sunno is
701. 1:51:07he's taking a full-on programming approach to it and it's working I take this sort of just vomit words at it and
702. 1:51:16just keep vomiting words until I get it right like the breadth of how you can get results out of large language models and
703. 1:51:26and these tools it's it's so wide like any way your brain works so so long as you're treating it like this
704. 1:51:33collaborative interaction with these tools you can get you can get it to work there's there's kind of an infinite
705. 1:51:39number of ways to do it oh I've vomited everything at it oh that's funny frostbitten says that's awesome um
706. 1:51:47Brandon hey I just landed what did I miss we spent the whole hour on uh jelly pod no we didn't I call suo a curation process oh
707. 1:52:03my God that thing I did today in Runway trying to get those those agent you know those those avatars looking like
708. 1:52:11something not from a [ __ ] nightmare I mean it was probably I don't know 30 or 40 Generations per little segment I did
709. 1:52:23um I am excited for these tools to be less janky but you know the opportunity is is the the biggest opportunity is
710. 1:52:32when the tools are the jankiest let me just leave you with that if if you're frustrated with the Jank of AI right now
711. 1:52:39that means that you're in in a good place because it means you're early and your opportunity is bigger than you realize all [Music]
712. 1:52:50right yeah the people that make fun of AI music because it's not real as silly I the people that make fun of AI music I
713. 1:52:57can promise you have never made AI music right if you look at it from the outside you can watch someone just go
714. 1:53:06push a button and out comes music and what the what the doomers um contend is that that's what everyone's doing everyone's just being
715. 1:53:18lazy most people are not being lazy that are trying to make music that's worth listening to they're putting a lot of
716. 1:53:24time into it they're putting a lot of effort into it they're putting a lot of production into it um it's just how you
717. 1:53:30do that production is different than how we currently do it but didn't we have the same [ __ ] when we went from you know
718. 1:53:37eight track tape recorders to digital audio workstations and midi and sampling I mean [ __ ] sampling in the 80s that's
719. 1:53:46not music that's just stealing other people's music right like we've been through this before it's the same [ __ ] [ __ ]
720. 1:53:53it's like but whatever this is fancier this is fancier sampling they're the same type of people that were saying Electronica wasn't
721. 1:54:05music 30 years ago exactly exactly you know the other thing that that absolutely blows my [ __ ] mind is
722. 1:54:13that the same people that are afraid of AI replacing them are like well then there's you know if it's this good then
723. 1:54:21there's clearly no place for artists it's like [ __ ] the whole point of an artist like what is an artist an artist is
724. 1:54:31someone who looks at the same world that you and I see and they see it differently right so artists are going
725. 1:54:39to look at the same AI tools that you and I see and they're going to look at them differently and they're going to go
726. 1:54:45oh what if I did this with it and then what if I took that and combined it with this and then they're going to make a
727. 1:54:51thing that the rest of us are going to like go wait how H what is that I've never seen that before right because an
728. 1:55:01artist put their point of view into using these tools that doesn't go away like like I just I like the [ __ ]
729. 1:55:11Simplicity I just feel like the political dialogue because it's so binary has kind of set us up with this
730. 1:55:18binary AI good AI evil it is much more [ __ ] gray than that right much more [ __ ] gray than that so anyway and
731. 1:55:29artists are more important than ever so anyone that tells you they're not clearly isn't using this [ __ ] all right Corey
732. 1:55:40Sandler artiste extraordinaire looking at things a bit differently a different box exactly I mean you should have seen when
733. 1:55:49we were all off building custom GP s Cory Sandler's gpts were [ __ ] insane well they still are crazy glasses
734. 1:55:58designers and just like you know artists are artists artists will do things differently so and artists are still
735. 1:56:07required it's about thinking outside the box yeah it's about looking at the world and having the confidence in who you are
736. 1:56:15to know what you want and then just go get these tools to do that for you all right Kyle please yes what what am I let's see Kyle my boss
737. 1:56:31ask teenage Engineers I don't quite know what you're asking Cosmic lover just DM me offline we'll talk about it tomorrow
738. 1:56:39um all right cool I am gonna get on out of here can Mr ASMR make another appearance in Tik Tock no I think well
739. 1:56:48so here's the thing I made that video the the uh hey are you scrolling because I get this same woman like I don't know two times a week hey
740. 1:56:59are you scrolling same woman same video and I assume that everybody got that video but they don't like so I made that
741. 1:57:07video and I think most people didn't understand it unless you're an irregular check teenage engineering sampler okay I'll check it out
742. 1:57:17archetypal architect if you like carpentry got an AI replacement I'd be like cool a new tool yeah exactly exactly all right
743. 1:57:34cool all right beautiful let's see I got a [Music] video I blocked [Music] it all right everybody I'm going to get
744. 1:57:49out of here I will see you tomorrow so wait tomorrow's Tuesday nothing going on it'll be same bat Time same bat place so
745. 1:57:548:00 pm tomorrow I don't think there's anything going on but I don't have a producer telling me all right all right um Cosmic lover we'll do
746. 1:58:06that tomorrow and because I don't want to start something new especially if I haven't seen it before I don't want to start
747. 1:58:12something new and run down a 30 or 45 minute rabbit hole right at the end here um all right peace out I will see you guys uh
748. 1:58:23tomorrow trying to think if there's anything else I need to talk about no not right now all right peace out everybody