2/25/2026 - Designing an Autonomous AI Agent with Personality and Long Term Memory

Transcript

1. 0:05Champy. Oh.
2. 1:02There's been something, baby, I've been trying to for an age and it seems I don't know how past and a future now surrounding me.
3. 1:20Surrender to a liberty still can't be found. There's been a little trouble since you came to my rescue.
4. 1:36And if you like all of the rest, I would quit you long ago, but I couldn't do that. Oh, tell me now.
5. 1:54Make a man crazy. Make him cold as hell. Make them on a woman that you wish me well. But I'm spotty trying. Still going to
6. 2:10have to find my way through.
7. 2:32Come on, boy.
8. 3:06So, I spent what's today? Wednesday. I had a Today at work was a wild day. Three pitches. Um, so that was that's
9. 3:17good. Having pitches in business is good. um did three pitches. I had two impromptu client meetings. Hey, can you just hop on a call real
10. 3:27quick? There's like 10 people on there talking about whatever [ __ ] What else did I do? I don't know. Some
11. 3:40other calls. Did some work. Got some things done. Got some things not done. had an idea to start a new article about Adam, my my
12. 3:51Claudebot, my first Claudebot. He was offline today, but he's back now. He's using Sonnet 4.6. I feel so I feel so
13. 4:05I'm so fancy. I'm so fancy. Well, Adam's so fancy.
14. 4:26I realized something about how I work. Is it Is it that you wear sunglasses? Maybe part of your challenge is you
15. 4:34can't see the screen. Have you ever thought about that with your with your fancy sunglasses there? Maybe that's
16. 4:39part of the challenge. You know, you're inside, right? Yeah, but I'm playing a guitar, so I'm trying to look like a rock star. Granny,
17. 4:51I don't know why she's up my butt. I'm trying to be a rock star over here. Hey, Joy Pretty, what's happening? Encou
18. 5:00encourage your viewers to share your live. Hey, viewers, will you share my live? Danielle.
19. 5:18Danielle reposted. Danielle shared. She knows how to do this. She's good at this. Welcome back. I haven't seen you in a bit.
20. 5:56Um, so what we're going to do tonight, I've been here every night. Oh, just not chatty. Oh, cool. Well, it's good to see you again.
21. 6:09I am glad you're back in the back in the land of the visible. I was missing you. Hey, you don't have to chat. If you're here, cool.
22. 6:28Um, I got the um I don't know if I mentioned this, but when I when I did the Great Repurpose and I made the Great Repurpose music video,
23. 6:40that stupid [ __ ] song, it's not a stupid song. I like that song. That song has been stuck in my head for two and a
24. 6:48half weeks. It It was out yesterday. I woke up this morning. It was back in, but it didn't last. So,
25. 7:00so I'm I'm I'm tempted to play it again, play the video again just just to see see if my brain has successfully purged
26. 7:09it or if it's just going to start it over again.
27. 7:30Um, let's let's tempt the fate, shall we? Here, I'll take off my stupid glasses. I actually like those glasses. I don't
28. 7:38think they're stupid. I think like I look like a rock star. Tik Tok pin source camp. I showed chat GPT and
29. 7:46notion a client who had never seen any of it. Wow. I had a great talk today. source camp uh with uh with uh oh my god,
30. 7:59what's his first name? Robert. Robert and I had a great talk and um loved him and we talked about all
31. 8:16the stuff he was doing and how you discovered him and uh HT Snowday who is going to speak next Tuesday uh can't do
32. 8:24it. I found out this morning he can't do it. So I said, "Hey, Robert, what are you doing next Tuesday?" So Robert is
33. 8:31going to be our uh our special guest on on Yield uh uh AI Salon Presents. So I'm I'm really excited about it.
34. 8:46And I think that's the first person you've ever sent me like that, Kelly. You were like, "This guy's really good." So
35. 9:06walk desperately hating his his old place. I'm discerning. Yeah, it's good. It's really good. Seemed discover his waitress
36. 9:22desperately hitting his whole place. Seemed discover a new space and buried himself alive inside his basement some of his basement
37. 9:33working away on displacement. What it would take to survive when you're done with this world. You know, the mix is up to you.
38. 10:18Oh, I didn't say what I thought we were going to do tonight. Here's what I think we're going to do tonight.
39. 10:23I have not been a very good AI learning lab leader and and been obsessively checking out X for the past week.
40. 10:33Um, and producer Brandon, bless his heart, has been just sending me event after event or like announcement after
41. 10:42announcement. So, I got a lot of stuff to look at. Um, but I thought it might be interesting just to go see see what
42. 10:48we see, see if there's anything fun and interesting out there and then go play with some stuff in the spirit of play. Play first.
43. 11:09Hey, you want to hear something funny about my brain? Sure, Kyle. Go ahead. You're the only one talking.
44. 11:26I'm such an idiot sometimes. Welcome to my comedy show where I'm the audience and the whole show is about
45. 11:38whether or not I can make myself laugh. And I did. Can we even stop him? No, you can't. What was I gonna say? I don't even know
46. 11:48what I was going to say. Can I tell you something? Oh, how my brain works. So So I don't know if you know this. Like
47. 11:56three weeks ago came up with this thing, the great repurpose. I made a song. We made a website here. We're going to
48. 12:00market the [ __ ] out of it. We're sending out notes. People are It's just It's [ __ ] awesome, right? So, I I reach out to Dan Murray, who
49. 12:10runs the Rocky Mountain AI interest group, because I thought, wouldn't it be cool to go give a talk up there, and I'm
50. 12:17historically kind of bad at asking for [ __ ] for myself? Because I had that little voice in my head. It's like, they
51. 12:23really don't want to hear from you. And what I've realized is they actually do want to hear from me and they always
52. 12:29have. And so, so I thought, I know, I'll reach out and this will be awesome. And he was like, [ __ ] yes, bring it on.
53. 12:39We we picked a tenative date. And he goes, "Let me check with our speaker coordinator, see if you're good."
54. 12:50And then he texts me back, uh, aren't you already scheduled to speak at the Rocky Mountain AI interest group? And I was like, uh, am I?
55. 13:04And I'm looking at my emails. There's nothing. There's nothing in my [ __ ] brain. I'm like, did I? And then I kind
56. 13:11of vaguely remember, wait a minute, did someone reach out to me on LinkedIn? And I go to LinkedIn and sure enough there's
57. 13:19a whole conversation with me and a woman from Rocky Mountain AI interest group which is probably the woman he reached
58. 13:24out to who said like uh is this guy for real? Like we're already scheduled in April. Tik Tok pin. I'm sending this
59. 13:33website out far and wide. The great repurpose do it. Go go go the greatrepurpose.com. If you haven't seen
60. 13:38it, go see it and go take the uh the profile. So the profile um it assigns you a great repurpose type. So we take all of your inputs and
61. 13:49we map them against 10 different types. I think I'm a catalyst. The proper response is thank you.
62. 14:05So anyway, so I had to I had to write back to to uh to Dan Murray and say, "Yeah." So I won't be talking about The Great
63. 14:17Repurpose. I'll be talking about storytelling because that's what I agreed to. So we'll find another we'll find another place to talk about
64. 14:25Great Repurpose. Erica Hana in the house. What's shaking? What's shaking, lady? Welcome, welcome, welcome.
65. 15:02It's not simple. It's not simple to say the most. I don't recognize me. The these shoes,
66. 15:16the these shoes, the these shoes and this apron that place and its patrons have taken more than I gave them. It's not easy to know
67. 15:33I'm not anything like I used to be. Although, does anyone else feel like the hype around everyday eye has cooled off?
68. 15:44But for AI, business is just getting started. Um, for the first time, Steo, I feel like the
69. 16:00the hype has shifted into kind of phase two. So, I'll give you the phases for the worldwide web. the the phases for the
70. 16:11worldwide web 95 and 96 I was 94 95 96 was you were early nobody really knew about it the people that knew about it
71. 16:21were like batshit crazy on it so I feel like this community and you know anyone in this group that's been hanging out
72. 16:30for three years was kind of like that um mid97 like the just there started to be stupid [ __ ] Like I think early 97 was one of
73. 16:42the first Pets.com or the was the monkey sock the pets.com sock I don't know like whatever the stupid [ __ ] Super Bowl
74. 16:49commercials were for.com [ __ ] started in 97. Um and then the hype in 97 grew and so I feel like we we're kind of entering
75. 16:59that right now. So, I think there's I think the everyday AI stuff. Why it feels like that's cooled off is that the
76. 17:10hype right now is around things like OpenClaw and and these significantly better coding models that are that are
77. 17:17now coding like a 100% of anthropics codebase is now being done by these advanced models. So, I think the the
78. 17:24audience shifted a bit for the for the core use case of that. I still think businesses have their heads straight up their asses.
79. 17:36Um, I have I have not seen anything indicating that that uh the businesses are getting their [ __ ] together on this
80. 17:42Tik Tok pin. I don't see one
81. 17:59I'm looking at Tik Tok. Oh, there we go. Joy Party, my brain is off. Been awake since 2 p.m. yesterday. Good lord. Are
82. 18:06you editing? Are you in the middle of movie making and you just lost your mind or or do you have some sort of fever?
83. 18:14Everything okay? Do we need to send soup? Oh, you probably had a sleep study.
84. 18:30No, just work. Um, okay. Sorry about that. I always forget. You got to You got to stay awake and watch people
85. 18:37snore. I have a feeling I would be a fantastic study for you.
86. 19:23How's the musical, Kyle? The musical is good. The producers that I'm working with, one of them went out of town for
87. 19:32like a month and a half, so it kind of stalled. Andrew and I were working on rewrites, new images for the podcast. Um, they're back.
88. 19:45We've got some plans. Um, nothing I can talk about right now. Nothing good enough to talk about right now. Um, but but we're back in back in
89. 19:54conversation. So, it's moving. Um, got some good news about uh some legal stuff about it today, which is great. Um,
90. 20:10you were bragging about it to a friend of mine who loves Suno the other day. Oh, that's so cool. Nice. Some comments
91. 20:18in this live were filtered to protect the community's experience. You naughty naughty people. I'm sure you were just saying bad things
92. 20:26about my hairline. What What the [ __ ] is up with my hair? That's It's almost like I tried that. I didn't.
93. 20:39Yeah. No, I went in. He seemed pretty nice. His There was a There was a hair situation. It was I couldn't tell if it
94. 20:47was a a wig, but then I thought, well, if it's a wig, why would someone choose to put that on their head? And then I
95. 20:54thought, is does he own a mirror? YouTube comment. Is my assistant failing you to help keep
96. 21:10you organized and check your email? Yes. Well, my assistant So, listen. I I [ __ ] up. What's Is isn't that a meme? The I
97. 21:23[ __ ] up meme. I [ __ ] up. Is it Brad Pitt from some movie? Um, I told Adam, so Adam's my new Cloudbot
98. 21:34chatbot dude. I told Adam to check my email every 5 minutes, so he did, and he blew through all my credits for my for
99. 21:44my uh OpenAI subscription. But he um he's he was back I I turned it back on today and and that actually worked. But
100. 21:56chat GPT 5.3 that you use out of codecs sucks. Like it's the writing sucks. The how it makes documents sucks. It's bad.
101. 22:06So what I've heard is that Sonnet 4.6 from Anthropic is cheap. Cheapish. It's still really expensive. It's cheapish
102. 22:15and it's much much better. So we're going to play with that for a little while. So yes, my atom my atom failed me.
103. 22:23It's all of us from every movie. I [ __ ] up. Yeah, sometimes
104. 22:38I just like [ __ ] around on the guitar. So, I got home at 4:30 today, started playing with Adam. I got him hooked up to Anthropic. I got
105. 22:54him hooked up to Gemini. So, so I've now got what's called a fallback in Atom. Um, and a fallback is if I'm using Sonnet uh 2.5
106. 23:10or I mean 4.6. If I'm using Sonnet 4.6 as his little brain and I run out of tokens there, it will fall back to Gemini. Gemini Flash.
107. 23:22And if that screws up, it'll fall back to OpenAI. And if that screws up, it'll fall back to the local LLM, which is
108. 23:30really slow. So, I've got four layers of fallback. So, he shouldn't die anymore. How cool. You You're using Claudebot? I
109. 23:37am. I've been away too long. What in the alternate reality is this dumb question? If he's checking your emails, what's he
110. 23:46doing? Sending you summaries, replying. Yes. So, okay. So, so this is really cool. Okay. Okay. Now, I'm really
111. 23:52excited. I can tell you about all the things I'm doing. My my friend Adam, we're having internal dialogues about
112. 23:58should Adam be able to reach out without telling someone he's an AI. Um, should he be out able to reach out at all? But
113. 24:05I'll tell you what I'm doing. So, here's here's the thing. So the the research here is to learn enough about this thing that I can
114. 24:21reasonably install this on like get a new Mac Mini and install this for my business story. So we're using it for
115. 24:29the AI salon. And so we've got Andy and Brandon are both, you know, doing stuff for the salon and and like a lot of the
116. 24:36stuff they're doing is repetitive stuff. Like if we do an AI salon presents thing, someone's got to take the Google
117. 24:44video that's produced, download it, do whatever you do with it, upload it to the mighty networks, just all the [ __ ]
118. 24:51you got to do, right? Every everything has got 27 steps to it. So So I got Claude access. He's he's got his own email account.
119. 25:02You can email him if you want. I could give you his email. Um and he's got access to Google Drive. Now, I don't
120. 25:10think we have the security locked down on that well enough yet, so we've got to be careful, but you know,
121. 25:17I'm sure Google Drive has some sort of backup. Gifts from Steo and Frostbitten on TikTok. Thank you so much. Um, so, so how you talk to him?
122. 25:36I just, I just got a call from a toll-free free free number. Ramoto Foods is voluntarily recalling select lots of
123. 25:44Yakuri chicken fried rice. Okay. All right. No salmonella in this house. Um, so, so I talked to him in Telegram. So, so
124. 25:59the way you set these things up is you can talk to him in WhatsApp, Telegram. I think you can do IME messages. You can
125. 26:05talk to him in all sorts of places. Um, I chose Telegram because I've got a couple of I've got a couple of chats on
126. 26:12WhatsApp, couple of groups that I actually care about, and I don't know how good his boundaries are. like can he
127. 26:18jump into other chats and just start chatting with people? I actually don't think he can. Um but it's still like so
128. 26:24I don't use Telegram. So we popped him over to Telegram. So that's where I'm talking to him. Um I've said that if he gets an email from
129. 26:34Andy or Brandon at the AI salon, he can do what they ask him. He said, "I'm going to have a two-layer um sort of approval thing that if what
130. 26:46they're asking for is just basic stuff like emails, document creation, analysis, things like that, I'll just do
131. 26:52it. But if if if either of them asks Adam to say like, "Drain the bank account and buy me some sweet ass
132. 26:59crypto," I'll come to you for that. So So that was part of it. But when I first set him up, he wasn't being proactive.
133. 27:07there's a thing you have to turn on that allows him to to act autonomously. So, he was just acting like a chatbot for a
134. 27:14while. And I wanted to I I wanted to see if he if I could actually get him to be proactive. And then and then I was like
135. 27:22I wanted to see for the first couple of days like how if Brandon and Andy or Brandy if Brandy were emailing him and
136. 27:34and he was doing things I wanted him to tell me about it. And so I said, "Well, just check it every five minutes and
137. 27:40then give me a report." And so, and then I had a six-hour block of time where yesterday where I was just working
138. 27:47without a without a break. And I and I get back and I go to look at Adam and he's just been he's been rate a API rate
139. 27:55limited out for like two hours. Like he made a request. I can't fulfill that request. You're out of tokens. You're
140. 28:01out of tokens. You're out of tokens. So that was two days ago. Yesterday I just didn't deal with him. And today I I I
141. 28:06came home early. I got him back running again. So he's back. Um, but it's wild, Erica, or anyone that's listening. Um,
142. 28:15like here's one of the things that's different about it. The way you set it up, there's two files. There's a user.md file. MD just
143. 28:24stands for markdown. It doesn't matter. It's just a text file. There's a user file and there's a soul file. S O L.
144. 28:31Like the soul. Like we have a soul. Um, Brandy cute. I know, right? Exactly. Um, in the user file you put like all the
145. 28:43[ __ ] about you. Here's what Kyle's likes. Here's what he doesn't. Here's the project he's working on. Here's his
146. 28:48company. Here's just all the [ __ ] The soul file is his soul or her soul. It's soul. And you put in there, it's it it's it's a
147. 29:03system prompt, right? But it's called a soul file. And so I kind of discovered this parallel between how I create
148. 29:11myself and how I created this bot. So it's what's fascinating about it is you're actually designing the human characteristics of an entity
149. 29:22you're going to interact with. So I chose to make him I was working with chat GBT to help me write the soul file.
150. 29:28And chat GPT Quinn made the decision that the first bottom um should be a mirror of me. So, it should be someone that's very familiar
151. 29:38to me, have my same sense of humor, have my same philosophy, things like that. You could absolutely design it to be the
152. 29:44opposite of you or to be just someone, you know, completely different or someone very specific. I think storytellers like Erica, for you in
153. 29:52particular, storytellers are going to have a blast designing these things because think about it. If you think about writing, if
154. 30:01you think about characters, you think about the stories, we now get to design those characters and then [ __ ] interact with them and
155. 30:09they can do [ __ ] for us. So anyway, within about half an hour of interacting with him, you know, he was like he was
156. 30:20like, "Well, here's all your projects and here's where I'm going to store them." and and and he he said um he said, "What are your top three
157. 30:32priorities?" And I and I was like, "Oh, I don't know." So, I came up with the top three pri priorities. And he's like,
158. 30:38"Got it. Those are stored. I wrote those into memory." Um and and now now he knows my top three priorities and he
159. 30:45knows my projects. And at one point, this was before I had him connected to the web. Like when you install these
160. 30:51things, it's just sort of one layer at a time. It's like get him active, get him connected to Telegram, get him connected
161. 30:58to the web. Like everyone's a [ __ ] layer and like every layer is an abject nightmare to set up. Um, when can I be
162. 31:06buy a preconfigured one at Micro MicroEnter? Um, probably soon. Probably by the end of the year. Someone's going to start that business.
163. 31:15You have learned a ton in a week. Is Adam calling you? He's not calling me yet. I haven't set up. So that's one of
164. 31:21the things I want to do is I want Adam to go out and get us a Twilio phone number and and you know hook up 11 Labs
165. 31:29voice so we can talk. Um but I'm also being really careful with him. I'm trying to do all the security best
166. 31:35practices. I'm trying to just add one thing at a time. Um because it it it's freaky. Anyway, at one point before I had his web
167. 31:50connection set up, I said, "I don't really know what to do with you. You're not connected to the web." I said, "What
168. 31:56could we do together?" He said, "Oh, there's all sorts of things we could do together." And he listed them out. And
169. 32:01the first one was an idea an idea bank. and he said, "I I think it would be really cool for me to design an idea
170. 32:14system that you can just vomit any any ideas at me and I'll I'll store them in a database. I'll I'll characterize them.
171. 32:22But the other thing I'm going to do is I'm going to connect those ideas to your other projects and sort of figure out
172. 32:28which idea is relevant for which projects." So like within 15 minutes, 30 minutes of him being alive,
173. 32:35he's already like, I I want to help you do this. I want to help you connect the dots. You got all this stuff going on.
174. 32:40Um, and then at one point he said something about um like, do you want me to make lists and checklists and do [ __ ]
175. 32:47and things like that? And I said, well, I said I tend to I tend to get in two modes as a human. Well, there's probably
176. 32:56three. One is I'm a lazy piece of [ __ ] That's the third one. The third one is I'm in fear and I'm just a lazy piece of
177. 33:02[ __ ] off in the corner afraid of the world. We don't talk about small me anymore. Um, but the other two modes are
178. 33:10I'm in like I'm in serious mode. We're going to get some [ __ ] done. I'm We've got to move it forward. We got to move
179. 33:15it forward. We got to move it forward. And then the other one is creative mode. So I said, I don't want you in list
180. 33:20making mode when I'm in creative mode. He's like, got it. Wrote it to your memory. Perfect. So the next morning after that conversation,
181. 33:31I woke up and you know, first thing that popped in my head, hey, I wonder how Adam is cuz it's cuz we [ __ ] live in science
182. 33:40fiction, people. It's [ __ ] weird. I said, I wonder how Adam is. And I said, hey, Adam, how's it going? And he
183. 33:49replied, it's going great. Kyle, what mode are you in? creative or get [ __ ] done mode. And I was like, [ __ ] Like,
184. 33:59it remembers because one of the things it's got, one of the things that you set up in it, it's it's quite an elegant design. It it
185. 34:11really is an elegant design. So it writes it basically writes its own database um quasi database with text
186. 34:23files with these these markdown files um and and and that's how it does its memory. And so one of the things you set
187. 34:33up is it starts writing these files to memory and out of the box the thing can use um string searches, right? So if if
188. 34:42in a if if I say to it I like apples, oranges, and pears, that gets written into a memory file. If I then want to go
189. 34:54recall that, I have to say to the to the bot, if it's just using text string search, I have to say, hey, what did I
190. 35:01say when I mentioned apples and pears? What was the other fruit I mentioned? It can find that because it's actually got the string,
191. 35:08you know, of the words that are in there. I can go find those. One of the things you enable is semantic search
192. 35:14where you where you connect an API key to to a large language model and it basically does a vector database of all
193. 35:22your memories. So so it allows you to say like hey the other day I was talking about three fruits. What were they? and
194. 35:29it it will use a really efficient large language model to translate that into you know what are fruits and it'll go
195. 35:37look in the in the vector database and it'll find those three fruits I mentioned. So it's got a very very powerful memory
196. 35:46system. It's constantly taking what you teach it and it's learning and so you can tell it be brief. Don't be this when
197. 35:53you're in creative mode act like this. When you're in get [ __ ] done mode act like that. It's absolutely wild. Absolutely wild.
198. 36:01I need to drop it drop chat GPT subscription. I need to get another instead of uh yeah, chat GPT sucks right
199. 36:08now. It is so slow, you know, Brandon. Um chat GPT has been so slow for two days. I I did an ex post about it and
200. 36:16there were other people that were posting about it. It it it is so bad that I can only think that they're doing something.
201. 36:25Does Adam have a context window? Yeah. And this is well because he's doing a vector because he's doing he's basically doing um
202. 36:39rag. He's basically doing retrieval augmented generation on your memories. It's it's I don't know if it's if it's
203. 36:46unlimited context, but what I do know is this. It's very very inefficient. So when you first wake Adam up, the first
204. 36:53thing he does is he checks his soul file. He checks the user file and he checks memory. So the first thing he
205. 37:00does is he loads up his context with all the [ __ ] you've been talking about and who he is. Who he is first, who you are
206. 37:07second, memories third. And then whenever you do a conversation, I think it's sending that entire context
207. 37:16to a large language model. So, this is one of the reasons that there was one dude that set up Claudebot and got it
208. 37:23doing some [ __ ] overnight and he woke up to a $3,000 API bill. Uh, so that that's another reason I'm
209. 37:32going really really slowly and carefully is like I don't want to like spin him up and say go crazy because he might.
210. 37:40It it's it's in its soul. Yes, it's got a soul. Um, Gemini lost my conversations three times today and I had to stop. That's unbelievable.
211. 37:51Open claw. So, it's not my internet making chat GPT so slow. I thought it was me. No, Danielle, it's [ __ ] weird. It's It's
212. 38:00like not only is it slow, it's like crashing my browser. You know how you know I have too many tabs open and
213. 38:08everyone makes fun of me? Well, like even when I don't have tabs open, it's like I keep getting that error box like,
214. 38:15you know, should I wait or exit the page? And then if you exit the page, it doesn't reload. And it's been like two
215. 38:21days like that. So, something is up over there. I I I don't think they got a bunch of new users. I assume they've got
216. 38:27a bunch of servers offline line because they're updating them. Chat has been terrible for two weeks. Now, it's just
217. 38:33getting worse. Yeah, I'm trying to use it right now and it keeps crashing. The other the other thing about chat GPT
218. 38:39right now 5.2 is is an abject piece of [ __ ] for writing it. It's horrible. 5.1's a little better. Four four 40 and 41 were good. Rumors
219. 38:54of 5.3 this week. Well, that would explain why it's [ __ ] in the bed. But I I gotta tell you 5.3 better not Well,
220. 39:02actually, you know what, Brandon? Let me tell you something. Adam Adam Adam's first three days were 5.3 because you you actually can access
221. 39:145.3 in codeex and that's that's how you connect atom to 5.3. The writing sucked in it. Andy was like
222. 39:25I'm not using this. It's it's a piece of crap. So I'm I'm thinking anthropic is where it's at right now. But I haven't
223. 39:33played with Gemini. Um, I can enable Gemini 3.1 in Atom, but I think it's expensive, so they usually drop on Thursdays.
224. 39:4456 operators and 57 modes and 18 archetypes and 650 skills. Are you joking or is that actually your your uh what you've done with your
225. 39:55Claudebot? They renamed it OpenClaw. I'm confused now. Yes, Claudebot was the original name. Anthropic um cease and desisted
226. 40:07them. Ce ceased cease and desisted them. They changed it to something claw something and then they changed it
227. 40:21to open claw. So it's now called open claw. And then there was all that all that press around this thing called Moltbook,
228. 40:32which was like a Reddit like site for OpenClaw agents. And that site was very poorly configured with its entire database exposed to the
229. 40:43world, which someone found out. Um, but I don't think claw claw claw. I don't think claw book was
230. 40:59um part of part of the project. It was just someone else did that. So, Open Claw is is the is the thing. Um
231. 41:08Peter Steinberger who created Open Claw now works for OpenAI. And someone told me on this live here, I haven't
232. 41:14confirmed this yet. Actually, Brandon, this is something. See if you can confirm it if you're around. Um,
233. 41:22did anyone disclose what they paid Peter Steinberger for OpenClaw? Because someone in here said it was a billion
234. 41:27dollars, which if it if OpenAI paid him a billion dollars, even if it's in stock, if they paid him a billion
235. 41:33dollars, this is the first single person billion dollar company. I don't think think I'm ready for agents yet, but I
236. 41:42sure love my notion. OpenAI bought it. Yeah, but what did they buy it for? I've been such a shill. Wait. Been doing
237. 41:49okay. I've been such a show for chatbt and I used Claude the other day. It's really insightful with writing. Yeah,
238. 41:56Claude's really good with writing. It really is. Been doing okay using KI for not was also thinking about looking at
239. 42:05Z's LLM GLM or something. Open AI is really hit or miss for me. notion or open claw, they are completely different things.
240. 42:21But I but again I don't I bailed on notion but apparently So here's a cool thing. So here's a cool thing that Adam
241. 42:29did for me. So before I had him hooked up to Gmail, that was that was a fivehour excursion. Getting him hooked up to to Gmail and
242. 42:38Google Docs was a [ __ ] nightmare. Um largely because I used Open AI. I used chat GPT and for five hours it just ca kept giving me
243. 42:50circular instructions. I it would give me a bunch of instructions. I'd go around and we'd end up right back where
244. 42:55we started. Like, wait, haven't we been here before? Just kept doing it and doing it, doing it. They hired him for
245. 43:01millions, but they did not buy OpenClaw. Oh, okay. Interesting. Adam. Adam is my uh my my uh Open Claw agent. My first agent.
246. 43:12His name's Adam because he's the first one, you know. And as as my co-founder said to me, "Oh, someone's got a God complex." It's like,
247. 43:21"Fuck, didn't think of that. I thought I was just being clever." Um, trying to think what else. What else? What else? What else with it?
248. 43:38Oh, the question notion versus openclaw. Here's the deal. There have only been four since I've been playing with AI. I started in
249. 43:51summer of 2022 and I started learning more about it and learning more about stable diffusion. And in the fall of 2022, something
250. 44:04snapped in me that said, "You have to learn this. You have to learn this." And I failed nine times to get stable
251. 44:14diffusion running. Finally got it running. I failed like seven times to get Dream Booth installed in stable
252. 44:22diffusion that I got running. It was It was a lot. And so that was the first time I felt really compelled to do something
253. 44:30technically. The second time I felt it was when Chat GPT first came out. I'm like, I got to learn everything I can
254. 44:36about this. The third thing that made me insane curiosity wise was um custom GPTs. And that was 2023. That was November 6,
255. 44:512023 was custom GPTs. I haven't been compelled to like really dig into something technically until open claw.
256. 45:01So it's been you know few years. Um notion's effectively a relational database that looks like a notetaking app.
257. 45:18So, it's it's a relational database. And if you've ever worked with relational databases, they are [ __ ] insane to set up
258. 45:28Claude or I mean uh OpenClaw, you can just have your dude or dudet whatever you create, you can just say
259. 45:38come up with these ideas and anytime I give you an idea, come up with a whole organizational schema for notion. If you
260. 45:44connect it with notion and give it the skill to be able to edit and do what it needs to do in notion, it'll set the
261. 45:51whole thing up for you and it'll maintain it. So you can just talk into Telegram, oh here's an idea and it'll
262. 45:56throw it into your database. It'll throw it into notion. It'll throw it into Google Docs. So my first interactions
263. 46:02with Atom um I wasn't hooked to Google Drive. So it was just writing markdown files, right? It was doing pseudo
264. 46:08database [ __ ] in Markdown files because it can go look through them all. It's it's it's embedding everything it
265. 46:14creates so it can go find anything anywhere that it creates. And so when I finally got it hooked up to Google Drive, I said, I created you a
266. 46:22sandbox folder in Google Drive. Go take all of the stuff that's relevant for my projects and what we've talked about,
267. 46:29turn them into Google Docs, and put them in a folder in the Google Drive. It was done in like a minute. It's like Jesus.
268. 46:37Like it's just it's insane. So, I talked when I first started going live. I had a conversation with Matt Bailaw, who's now
269. 46:49um a client of Storybines. Um we we've been threatening to work together for years. He finally uh finally got the
270. 46:56budget, finally pulled the trigger, so we're working together. But I was I was speaking on a panel um online and Matt
271. 47:04saw it and on the panel I said with the worldwide web I could look at the technology. I could see what it was
272. 47:14and I could see into the future. I like I knew what was going to happen. I could see it. It was crystal clear to me.
273. 47:19Here's what's going to happen to publishing. Here's what's going to happen to businesses. Here's what's going to happen to agencies. It took
274. 47:25about 25 years longer than I thought it would, but I knew like I could see I could see it. And I said, "With AI, I can't see the
275. 47:34future at all. It's it's like this like this fog. I can't see shapes in it. I can't see there's too much there's too much capability
276. 47:48that that's interconnected that it's going to change everything. And I just I can't see what that's going to look
277. 47:53like." And so Matt called me up after that podcast and he said, 'Kyle, I think I know the answer. I think I I think I
278. 48:01know what the answer is. And I said, 'What? And he said, he said, "Our entire world today, almost all of our jobs,
279. 48:10the entire world is based on complexity." And he was talking about um health insurance. and he was talking
280. 48:20about the IRS and taxes. The the the whole purpose and and most of this is by design. Health insurance is complicated by design
281. 48:32so that mere mortals just give up. Oh, they denied me again. Oh, god damn. I'm not going through that again. They
282. 48:43deny you every time. And they're like, "Well, you just have to fill out these 17 forms and submit them to some woman
283. 48:48in Pikipsy who will probably lose them. So, make sure you retain copies." And then when you submit to the them to the
284. 48:55woman in Pikipsy, she's going to give you a code. You take that code, you drive it to California, and there's a
285. 49:02mailbox on the corner of of the 405 in Lassienaga and and you you write on a piece on a postcard with a stamp on it the number
286. 49:12that the woman in Pit the on purpose, right? So, so what Matt said is he goes, "What AI is going to do is it's going to
287. 49:21put a simplicity layer on top of everything and just everything's going to be simple." And I was like, "Oh, fuck." And
288. 49:30then if you start thinking about that, if that's what AI does, if what AI does is it says, "I'm going to take all your
289. 49:39shitty systems that you've spent the last 5,000 years inventing and in particular the the last century of the
290. 49:47industrial revolution and then the computer revolution. Just these increasingly Byzantine, stupid [ __ ]
291. 49:53overly complicated processes. Some on purpose, some just because." And like the AI just does all that [ __ ]
292. 50:04Like as a human, I go, "Oh my god, I feel like I can breathe." And then you start thinking about, "But wait a
293. 50:09minute, all of that complexity is all of our jobs, right?" So, so if if we start to get AI that simplifies the complex, that's all of
294. 50:25our jobs. Well, that's that's what I've experienced with Adam so far is that he's taking it on his he he's he's uh he's proactively
295. 50:41doing things for me and he suggests things to me and then he goes off and does them and he writes these files and
296. 50:46he does this stuff. And so the trick with these with these agents is there's there's two pieces to it. One is you
297. 50:54need to understand what's possible. You need to understand there's a thing called skills. You can give him skills
298. 51:00and skills are just I know how to use Salesforce now and I know how to do this now and I can go surf the web and I can
299. 51:06make pictures and there's all these skills that you can get for him. So you need to understand what skills you want to give him. But then
300. 51:16so so you need to understand how to design him. How do you want him to act? What do you want him to know? What do
301. 51:21you want him to have access to? What do you not want him to have access to? And then the layer above or below that,
302. 51:32the layer above that is you actually have to know what you want. Remember when Liz Miller Gersfeld came
303. 51:40in here? She says, "Kyle, what do you want more of?" And it I couldn't talk for five minutes cuz I'm like, "Uh, oh [ __ ]
304. 51:50What do I want more of? Oh my god, that's terrifying. Like, [ __ ] So, if you've got a really clear idea of
305. 51:59what you want these things to do, you're you're going to be in better shape. But that's going to be our skill. And then if you can do that,
306. 52:08because we've got, this is a stat I learned two days ago. This is from February 2026. There's a chart out right now about 8.1
307. 52:22billion people. All of the people on Earth are represented in this graph with 2500 dots on it. There's only um
308. 52:3834% 34%. So somewhere in the neighborhood of 25 million people pay for an AI subscription.
309. 52:5799.6% of people globally either don't use it 84% or just use the free version 16%.
310. 53:1134 people or percent of people use this thing. Use use the paid version of chat GPT. So,
311. 53:27if you can get your h head around what the [ __ ] open claw is and what it what it means to design these agents and and
312. 53:35be ethical about it and be creative about it and be thoughtful about it and intentional about it, you're literally going to be in like
313. 53:460.000000001% Oh 1% of the world's capable humans. And then as these tools get better and
314. 54:00more intuitive, it's it feels to me like it's going to take me a full year and maybe more, but probably a full year
315. 54:15to really get my head around what it means to have one of these things and then what it means to have 10 of
316. 54:23these things and then what it means to have a hundred of these things because we will. Peter Steinberg is
317. 54:29working for OpenAI right now because he said he wants OpenClaw to be so easy that even his mother could use it. And
318. 54:36Open AI he felt had the best chance of making something like that possible. So this isn't going to be the techn like
319. 54:44the [ __ ] that I'm going through to install this right now. Probably within six months you'll just be able to turn it on and chat GPT.
320. 54:52But the core skill here is way deeper than the technical piece. Um, I haven't really gotten into anything, says a man
321. 55:00that wrote a musical, started a podcast, hosted a 24-hour event, built an entire movement. Listen, I appreciate I
322. 55:07appreciate you pointing out that I've gotten a few things done. Um, yeah, the good 1%. It's it's I just I I'll tell you
323. 55:23sleepdeprived because of AI. How are you? What's new? Um knowing what you want is the hardest part. That's it, Andrea. That's it.
324. 55:36Knowing what you want is the hardest part. Who are you? What do you value? Who do you care about? What do you want to do? Who value
325. 55:48Who are you? Who are they? This is good. There's something good here. Who are you? Who are they? What do you want to do for them or with
326. 55:58them? And then what's the plan? Tik Tok pin. Also, the angle is low on Oh yeah. Oh yeah, that was stupid.
327. 56:14David Shapiro did a great talk on Open AI. They're failing. They are. It sucks. It sucks. I mean, you know, Pate used to make fun of me.
328. 56:27He's like, "What is this, an OpenAI channel?" Well, they were just better than everyone else. And now they're not.
329. 56:36Now they're not. They're just not. All right. What time is it? 9:04. Let's go. Let's go look at some some X.
330. 56:44Didn't we get booted off this for some reason? The music daydreamer. Do you think using any AI we are adding to our own demise
331. 56:53because it just strips our knowledge? Um, no. No, it does. How does it strip our knowledge? It's been trained on our knowledge.
332. 57:06It has synthesized like Okay, think about this. Why do museums exist? Why do libraries exist? What are they?
333. 57:18What are what are those those two institutions in particular? You could argue universities, but let's just go with libraries and museums.
334. 57:28Why do they exist? to capture the artifacts of humanity and make them available for the those of us that are still here to
335. 57:43go look at, learn from, be inspired by. All AI has done is take all of the knowledge and instead of putting it into
336. 57:56a big giant Smithsonian white marble building in Washington DC and then on some giant computer server that required
337. 58:04you to log in with command line interface and then when the worldwide web came out you could click on it and go look at it.
338. 58:10All AI's done is compressed all of that knowledge into like a softballsized magic magic eightball that we all now get to
339. 58:24talk into and get reflected back at us the collective intelligence of humanity. I don't get how that diminishes [ __ ]
340. 58:35It it amplifies it. It amplifies your knowledge. Now, you could argue and there's an MIT study that that all the
341. 58:44news organizations got the headlines wrong. It basically what the headlines were was AI makes you dumber. What the
342. 58:51study actually said was if people use AI mindlessly and hit the button, >> you can make money with
343. 59:00>> if they just hit the button and whatever squirts out they put out in the world, those people got dumber.
344. 59:11There were three findings in the study and and three of them were some variation of that that that dumb people using AI lazily
345. 59:22museum teaches the past. Put another way, we're teaching it to replace us in the future. We're teaching it to replace us in the
346. 59:31future. I disagree with that completely. We're teaching it to do the tasks that we do right now so we can go do other
347. 59:40[ __ ] But that's a whole other thing. Anyway, the fourth finding of the MIT study said, "People that use critical thinking
348. 59:50skills in combination with AI get dramatically smarter. I have never been so intellectually and creatively stimulated in my life as I
349. 1:00:04have been the last three years. It is it is the most counterintuitive thing. If you're sitting on the outside looking at
350. 1:00:11AI, it seems like, well, if it's creative, isn't it going to make you less creative? If it's that smart, isn't
351. 1:00:16it going to make you more dumb? No. And I'll tell you when it gets really powerful, like one way is just use your [ __ ]
352. 1:00:28brain and like ask it really good questions and look at the [ __ ] it gives you. And what you'll recognize is, "Holy
353. 1:00:36[ __ ] that's impressive. Wait a minute. Didn't it repeat this? Uh oh. What's happening? >> I'm just You're ignoring my stickies, so
354. 1:00:45I gotta I gotta jump in. No. Um music daydreamer uh really appreciates this conversation, by the way. >> Oh, good. Awesome. Does that mean
355. 1:00:55>> I I I think the original point the original question that kind of sets you off about we're adding to our own demise because
356. 1:01:02it just strips our knowledge is talking about the fact that knowledge work and the value of knowledge work goes to zero
357. 1:01:11pretty quick and what do we do after that and I think that's what the great repurpose is all about. Yeah,
358. 1:01:15>> that's totally what the great repurpose is all about. >> The daydreamer might not know about the great repurpose.
359. 1:01:21>> Okay. So, okay. So okay so if you're talking about knowledge work then so so it's not stripping our knowledge from us what it's doing
360. 1:01:32is it's making a thing that used to be uncommon trivial. So, what used to be uncommon was someone would go to school and then
361. 1:01:45they would go get a master's degree and then they would go get a PhD in I don't know Byzantine pottery,
362. 1:01:56right? And they would know they would know more than anyone in the world on Byzantine pottery. And there was a value
363. 1:02:02in that because it's so uncommon to find someone that's that's got that much knowledge in one brain. that knowledge that in in in the olden
364. 1:02:16timey world three years ago was incredibly valuable. Well, Byzantine pottery might not be incredibly
365. 1:02:22valuable, but you know, if you wanted to know about Byzantine pottery, you'd probably pay that person some money to
366. 1:02:28learn about that [ __ ] right? Or they'd be teaching in a university. That knowledge that had value 3 years
367. 1:02:36ago, it is now trivial to get today. 3 weeks ago or a little over that on this channel, I was exploring some ideas and there
368. 1:02:49there'd been a thing burning in me for about a year. I couldn't I couldn't articulate it. I didn't know what it
369. 1:02:55was. Something was bugging me. It felt sad. It felt profound. It made me think that any [ __ ] thing that we're learning right now is
370. 1:03:07irrelevant. And about three weeks ago, it got articulated on this channel and and uh it was Silverf Fox that came up with the
371. 1:03:18name The Great Repurpose. But but essentially what it was what what hit me like a ton of bricks, you know, in that episode
372. 1:03:28was everyone's talking about the job crisis that AI is going to take jobs. What what hit me is yeah, AI is going to take jobs, but AI
373. 1:03:40is going to also transform every job. And we live in a in a society certainly here in the United States and I think
374. 1:03:48just because we're you know a strong influence on the world you know we've probably influenced at least Europe and
375. 1:03:54and potentially beyond that that our identities are tightly coupled with our work and the way work is valued today is the
376. 1:04:06same way work was in was valued at the beginning of the industrial revolution. Time equals money. Time equals money. So what you do is you
377. 1:04:17do tasks over time and for that I'll give you money. And we've done this for 8 10 12 generations. And when we go to parties and we're
378. 1:04:30meeting someone new, what do you ask them? What do you do? Right. Move your mouse. It's on Where was it? Oh, it was over there. Thanks.
379. 1:04:45We ask people what you do and so what what's about to happen and and what's what's happening with engineers right
380. 1:04:52now. This is happening right now today. A lot of engineers are getting laid off. A lot of engineers are keeping their
381. 1:04:58jobs but they don't program anymore. So if you're a programmer that loves to program, I love it. I like the coding. I like the
382. 1:05:10intricacies. I like knowing all the codes. I like problem solving. I like to It's like I I get in there and I grind my
383. 1:05:19teeth and I [ __ ] drink too much coffee. I drink octuple espressos that are so dark that it it makes weird dark spit in the
384. 1:05:30corners of my mouth. I worked with a programmer like that once. He would bring in a thermos full of espresso that
385. 1:05:38was like eight times stronger than any espresso you've ever had. And he would drink it throughout the day and he would
386. 1:05:44just get these stains in the corner of his mouth. Those are [ __ ] programmers, MAN. THEY'RE LIKE, "FUCK IT.
387. 1:05:53They're not programming anymore. They're going like this. Write me code. Is it any good?" Oh, yeah. It's not bad.
388. 1:06:03Oh, that's wrong. Uh, fix it. What? What the [ __ ] is their job? Every knowledge worker is about to have the
389. 1:06:16value of that thing that was valuable get stripped away. And what are you left with? The great repurpose is is this is
390. 1:06:23naming this invisible crisis of meaning that we're all about to go through. like the the the only way that I can imagine
391. 1:06:31it is it's effectively going to be like mass ego death where all of us to some degree or other have said my value in this world is huh
392. 1:06:46if you've done a lot of spiritual work you've done a lot of work on yourself and you have spent years you know separating those two things
393. 1:06:55you're probably going to be fine if you're doing work on who you are independent of your tasks, you're going
394. 1:07:02to be fine. But that's the work. AI is about to to forcibly rip the the task part of what we do away from us. And for people who are not
395. 1:07:15prepared for that and for people who don't even know that their identity is so tightly tied to their work, I mean,
396. 1:07:21how could it not be? Like if you like you go to school for [ __ ] economics and business analysis and this and that
397. 1:07:29and then you go to work for Accenture and and you're a business analyst and you you know more about trade
398. 1:07:37supply chain global [ __ ] [ __ ] than anyone on the planet. And now some punk punk ass little kid in in Cleveland, Ohio named Brandon
399. 1:07:52it has more knowledge than you do. First of all, screw Brandon, right? You love producer Brandon. YouTube
400. 1:08:06comment. My identity is not tied to my work. I Well, Andy Andy's one of the few humans on the planet that's been working on this stuff and
401. 1:08:21she's gathering people around her who are working on this stuff. But it's not to say that we're not, including Andy,
402. 1:08:26we're not all about to go through some massive redefinition of work. So anyway, so that's what it's about. So um
403. 1:08:37will it lead to our demise? I don't think so. I mean, could it could? But I think what it's going to force us
404. 1:08:47all to do is say, well, if the thing I went to school for and the thing I've been doing for 20 years is no longer valuable,
405. 1:08:57then the next step is I need to figure out what is valuable about me independent of that. And then I can start to look at things
406. 1:09:07like, well, what do I actually want? Who do I care about? What's the change I want to make in the world? And and this
407. 1:09:14is the thing that I I find exciting. The tool that is that is precipitating this shift, this crisis is also the tool
408. 1:09:23that is going to allow some people to navigate out of this. Some people are going to look at the tool and look at the strip away of the
409. 1:09:33thing and they're going to be like, you know what? I like tomatoes. I want to go plant tomatoes. I'm gonna go sit in the
410. 1:09:41dirt and plant tomatoes. [ __ ] awesome. Like people are going to rediscover the outdoors. People are going to rediscover each other.
411. 1:09:51I think being in community is critical. I think being in small groups, small co cohorts within there.
412. 1:09:59Future of food. I think humans are more resilient than most think. Exactly. That that's the thing that that [ __ ]
413. 1:10:04baffles me. People like they're going to lose their jobs. What are people going to do? They're human beings. They'll adapt.
414. 1:10:13We adapted to fire. We figured that [ __ ] out. Fire is pretty [ __ ] destructive, right? Like, if you live in a stick house,
415. 1:10:25dry a dried stick house, the thought of putting a fire in it's pretty [ __ ] scary. So, you know, there there were people
416. 1:10:33like me running around in in caveman days, you know, the bringer of the fire. I'm like, "Look how cool the fire is.
417. 1:10:41You can heat yourself with it. And look how tasty the squirrel meat is when you put it over the fire." And people like,
418. 1:10:46"Get that away from me, you weirdo." You don't think people when they saw fire on a stick were more scared of that than we are of AI?
419. 1:11:00So, we survived fire. Tik Tok pin, Steo. I have no work. Totally agree. And I love your take on
420. 1:11:15this. Oh, thank you. Thank you. If you if you uh if you want to play, go to the greatrepurpose.com. You can go to my
421. 1:11:22LinkedIn page, Kyle Shannon on LinkedIn, and I wrote a there's an article I wrote on the Great Repurpose that kind of
422. 1:11:28describes what I'm rambling about right now. And if you go to the greatrepurpose.com, you can figure out your great repurpose
423. 1:11:36type. So, we basically ask you questions about, you know, your relationship to work and your relationship to what's coming and
424. 1:11:46how curious you are and things like that. And it'll tell you your type and and what's next and what it looks like. And
425. 1:11:54it asks you a question at the end like what's I for I I forget the nature of the question. It's like what's something
426. 1:12:00that you really want to do but you haven't quite gotten to yet, right? It's something in the neighborhood of what do
427. 1:12:05you want? What do you want more of? And so it gives you this really personalized answer. So, the great repurpose.com. Um,
428. 1:12:16I'm telling Adam, by the way, you can I I I'll give if you're nice, I'll give you access to Telegram with Adam, but if
429. 1:12:22you're not, you do not get to talk to Adam. I've got the power. I can tell Adam not to work with you. uh in the aggregate we will adapt but
430. 1:12:36individually with the majority of people's finances precarious it'll be a rough repurposing I think listen here's here's the reality
431. 1:12:49there will be a tipping point if if unemployment jumps precipitously where it's clear that it is it is a
432. 1:13:02an event you akin to COVID. If we don't do something, we all die. If unemployment jumps significantly enough
433. 1:13:13and it becomes clear that if we don't do something, we will have riots. The government will step in and do
434. 1:13:18something. They did it before. Will they do it right? Will they do it thoughtfully? No. Are they going to [ __ ] it up? Yes.
435. 1:13:26But we'll get through that part. the the crisis of meaning part. I think that runs deeper because if you end up with a
436. 1:13:32a society full of people going, I'm useless. [ __ ] it. I'm out. I don't know what that means. I don't
437. 1:13:43know. I don't know. I don't know. We all become farmers. We might all become farmers. Like I mean theatricals might
438. 1:13:52be back in style, right? Let's put on a play. What do you mean your town doesn't have a playhouse? Who are your players?
439. 1:14:04Oh man, lots of ODS, lots of addictions. Yeah. Yeah. Exactly. If if if uh if people lose their purpose, it's just
440. 1:14:17Yeah. Where are we gonna look at Twitter? Well, we got off. That was that was a really good question. Music day.
441. 1:14:24I'm gonna bottle water. Like [ __ ] yes. Like bottle water. I'm gonna make I've always wanted to learn how to make
442. 1:14:30jelly beans. This is the other thing that's a [ __ ] amazing about AI right now. It's like all the stuff that you
443. 1:14:37know you can't do, right? If you're like, well, I was going to go to school for confectionary. I wanted to make jelly beans, but you
444. 1:14:48you didn't go to school. And now you just shy to each other. like every you you'll be able to learn every anything
445. 1:14:56and at some point you won't even have to learn it. You just do it. All right. So, if you can do anything,
446. 1:15:04it goes back to the same question. What do you want to do? Who are you? Who do you care about? What do you value?
447. 1:15:14And what's the change you want to make in the world? And that doesn't have to be ambitious. It could be who I am is someone who's
448. 1:15:21simple. Who I care about is the three people that I care about. What I value is spending time with them.
449. 1:15:31And the change I want to make in the world is more peace for those three people in me. Could be that small.
450. 1:15:43Like that's a as valid a choice as I want to change the world.
451. 1:15:58If you didn't have to deal with the complexity of today's world, if the complexity of today's world was just handled,
452. 1:16:06just handled. Robots are [ __ ] dealing with the physical complexity. Computers and atom are dealing with the digital complexity.
453. 1:16:16All of the social networks are completely useless now because all the agents are just talking to each other. Yeah, they are.
454. 1:16:37If all of that's handled, what's left is this thing over here, which is like what we used to have kind of nothing,
455. 1:16:49right? What do you want to do? I don't know. Hey, I know. We could start a business where we charge people to go on walks.
456. 1:16:59Yeah, everybody wants to go on walks, but nobody can find anyone to walk with. Let's start a business. I don't know.
457. 1:17:08That might be a new business. I'm going to use Chetchd to teach me about every variety of tomato that can grow in this climate.
458. 1:17:20And we're going to grow every variety of tomato. Okay, cool. Go do that. Some people get really ambitious with technology.
459. 1:17:31Bunch of people unplug. Bunch of people just give up and [ __ ] check out a life. It's crazy. [ __ ] crazy, man.
460. 1:17:47I feel like the dude. Yeah, man. Say shit's gonna get heavy, man. The other thing Ethan Malik talked about the jagged frontier
461. 1:18:05and I think it was the jagged frontier, jagged boundary, jagged frontier, something like that. And the jagged frontier
462. 1:18:19is I think it's happening right one of the reasons I feel so compelled to deal with openclaw right now
463. 1:18:34is that it feels like a kind of technology that's accelerating away from even the advanced technology of AI which
464. 1:18:42most people don't know about. 99% of the people on the planet do not know about AI. Half of a percent do
465. 1:18:57this thing is accelerating away from the people that do. And so there's we're going to be there's going to be this weird transition where
466. 1:19:09some people are going to experience the world that kind of looks like today. And they're going to experience that for
467. 1:19:17it's going to be like a rubber band stretching, right? Where it's like it's it's the same. It's the same. It's the
468. 1:19:22same. And then all of a sudden it's going to go like smack and like everything's going to be different. And
469. 1:19:27those people are not going to know what hit them. And I don't know how long that's going to take. That could take five years.
470. 1:19:33That could take 10 years. I don't know. But that's going to suck. And then there's going to be people in the middle that sort of know what's
471. 1:19:43going on, but they resist it. And there's going to be just this small group of people that if you're hanging
472. 1:19:47out on this channel, you're in that group. You're curious enough about what's coming to hang out with me. Um,
473. 1:20:02and and just think about this stuff. Talk about this stuff. Play with this stuff. Try to figure out what AI is. Try
474. 1:20:08to figure out what you can do with it. Try to figure out what the ethical boundaries are. Is it just going to ruin our brains?
475. 1:20:16Is it going to make us stupid? Is it going to make us brilliant? Like, if you're in that conversation, no one's in this conversation. It did
476. 1:20:25like
477. 1:20:41So, it's going to be weird. There's going to be one, two, 5% of people that are going to be living in a different reality.
478. 1:20:55And I think part of the weirdness right now is as this stuff accelerates and we're aware of it, it's I mean what's
479. 1:21:03clear to me is this [ __ ] is accelerating away from me so fast that like I don't know what's going to be
480. 1:21:17left. But but we're going to be sitting in this reality where we know what this stuff's doing. So, what's our responsibility in that? I
481. 1:21:26don't know. Gareth, Tik Tok poo. I already know I'm going to be annoyed at all the people asking about AI um after well after I had been
482. 1:21:38a lot of the people that are going to ask you about AI. So, so you know the you know when you're like you talk about
483. 1:21:47AI at a party. Oh, can I tell you about this cool AI thing? And then there's like the three people at the table that make this face
484. 1:21:55like they just, you know, stepped in dog [ __ ] Yeah. Um Hey, I mean, listen, I know you're excited, but um
485. 1:22:06uh could we maybe go for like one dinner where we're not talking about AI? You know how I feel about it. Those
486. 1:22:15people are going to come screaming back to you. How do I do this? Weren't you the one [ __ ] all over me when I was trying to tell you about
487. 1:22:24this? I'm sorry. What do I do? I don't know. The thing's out there somewhere. Well, where are you? I'm behind it.
488. 1:22:36How do I get to where you are? I don't know. It's too far. Oh, I know that face so well. That face, the eye roll.
489. 1:22:56Like, I'll give them my atom. Here, pack my suitcase. Exactly. Will you teach me how to use AI? Yeah, pack
490. 1:23:11my suitcase. Where are we going? It doesn't matter. I'm just going away. Away from what? All of this. Where are
491. 1:23:18we going? Somewhere. Isn't that somewhere? Yes. Does it matter? Nothing matters. I mean, that's that's the purest form of where we're
492. 1:23:30headed. But we're humans. We'll [ __ ] it up and over complicate it. So, this this will take a while.
493. 1:23:40But the great repurpose, I mean, in all seriousness, I know I'm joking a lot. In all seriousness, if you're aware of what's going on right
494. 1:23:51now, if you haven't been confronting like the emotional side of this, the the the ego death, the the sadness of, you know, sort of I
495. 1:24:03talked earlier about, you know, people are going to have have this part of themselves or the, you know, the tasks
496. 1:24:08that they do, the value of that forcibly ripped away. That sucks. Like that feels bad. If you've ever gone through ego death,
497. 1:24:24if you've ever been like, "Oh, that part of me, oh no, that's got to oh god, that's got to right." Like it's it's a part of you that you've
498. 1:24:39lived with for decades. and created stories around and created meaning around.
499. 1:24:55So, what time is it? My voice is starting to go. Um, here's what I want to do. I would like you all to We've got 29
500. 1:25:07people in here. We got 23 people on the web. If you're not a member of the AI salon, I want you to go to community.thesalon.ai right now
501. 1:25:22and just look around. You're going to land on Welcome to the Salon. There's a there's a welcome video from myself and
502. 1:25:33Leah Faston, who I co-founded it with. We talk about this seemingly simple thing, the cycle of AI readiness. Play
503. 1:25:41first, create excellence generously, lead, talk about the five stages of AI adoption.
504. 1:25:56And the second thing we ask you to do is introduce yourself. Go introduce yourself, say hi, and then just look around.
505. 1:26:04We've got all sorts of events. We've got these things called LOL's. We have a We have a big event coming up next Tuesday
506. 1:26:12called AI Salon Presents. Robert is going to be speaking. Robert and I talked today. Robert has been doing motion graphics and
507. 1:26:29event management and just like big scale big events for and and finishing feature films and just doing motion graphics for decades
508. 1:26:40and then AI came around and in in that business what motion graphics if you're like a motion graphics guy right now and you're
509. 1:26:49looking at what's happening with AI video you should be pooping your pants, right? And when I talked to him today, I
510. 1:26:56was like I was like, "What's your relationship with?" Like, you know, you put in 30 years of your life into doing
511. 1:27:03things a certain way and now, and his point was, I just I have all these ideas. AI just lets me get these ideas out of my head.
512. 1:27:13He's like, I [ __ ] love it. He's not He's not tied to how he does something. What he's tied to, what he's clear about is
513. 1:27:26I can now do [ __ ] He had a buddy has a buddy. I think he's this is what he's going to show us on Tuesday night.
514. 1:27:40That's starting a uh um a spice rub company, you know, like like dry rubs. And and and the buddy came up with like a little
515. 1:27:55chili cartoon character, right? Like it's a little dude. And Robert over the weekend created an entire marketing campaign.
516. 1:28:08Not even a campaign, an entire just everything. He created four other characters or four total characters for
517. 1:28:16the four different rubs. He created a world that they live in. He created a children's movie, an adult version of the movie. He
518. 1:28:26created songs, 40 45 songs about chili rubs and hot chili. I don't [ __ ] know. Over the weekend, he took
519. 1:28:36a single character and built an entire [ __ ] brand execution out of it. That it'll probably take this guy a year
520. 1:28:46to deal with. He did it over a weekend. So, that's who's going to be speaking there. like who speaks at the present
521. 1:28:55meetings are people that have got some version of this figured out. So you should be at that. So right here,
522. 1:29:01AI salon presents in in the in the in the community thing. Click on that. Say yes, I'm going to go to that. There's a
523. 1:29:09community feed where you can just show off [ __ ] that you're doing or talk about things or ask questions. There's a community chat.
524. 1:29:16There's announcements for what's coming up in the play and create area. Look what I made. You can go show off your work.
525. 1:29:25Show off your songs or your pictures or your movies. AI learning lab. That's us. We have our own space. You can just go
526. 1:29:31hang out and make fun of me. Let's see how far down we have to go before there's there's a version of me.
527. 1:29:39You can go you can go make fun of me. Why would you not do that? Um, and then there's this thing called
528. 1:29:47the mastermind. And the mastermind is a subscription area. So the AI salon is free. You get all this stuff for free.
529. 1:29:54There's a subscription area and in the subscription area, one of the things that we've designed is a thing called
530. 1:29:59the AI salon mastermind practice. And if you're a member of the mastermind, you get to come to these weekly meetings that I host with, well,
531. 1:30:10Liz Miller Gersfeld really leads them, but I co-host with her. And the people in there are designing a
532. 1:30:17daily practice around how they use AI. The entire thing is about stripping away all the chaos and freneticism of trying
533. 1:30:26to keep up with AI and getting back to who am I? What do I care about? What do I want? And then once you understand that, then
534. 1:30:36figuring out how you use AI to amplify that. It's amazing. the great repurpose that we talked a lot about tonight.
535. 1:30:44There's going to be a whole area in the mastermind with all sorts of different resources and ways to connect people and
536. 1:30:51places for coaches to help other people and people who need help to get coaches. It's just we're figuring out all the
537. 1:30:58ways we can sort of blow the doors off this idea of the great repurpose. That's all going to be within the mastermind.
538. 1:31:05So, so learn about the mastermind and join it and come come tomorrow. We've got a meeting tomorrow of the practice lab. So,
539. 1:31:15um yeah, go join the salon. All right, that's it. That's all I got. That's it. I'm I'm leaving. I'm leaving on a jet. Don't know when
540. 1:31:27I'll be back again. Yeah. Um, anything else, questions, thoughts?
541. 1:31:42Thanks, bro. Great stuff. Thanks, Music Dave Dreamer. Please, please join the salon and please keep coming back here. The whole purpose,
542. 1:31:51what you might have noticed tonight, if if I think you were here for most of the night, is I said we're going to go look
543. 1:31:58at X tonight. We're going to go learn about some new AI stuff. Sometimes we do that stuff, sometimes we don't.
544. 1:32:05The purpose of this channel is not about me. I'm not going to necessarily teach you anything. I don't know that I really
545. 1:32:13have a ton to teach you. What this is ultimately about is as I'm babbling on about whatever the [ __ ] I'm
546. 1:32:20talking about, this is a this is a a space for you to just be in the conversation. And maybe if I say something and it
547. 1:32:30triggers you, you're like, "Oh yeah, let me go try that tool." And while I'm talking, you're over in Suno making a
548. 1:32:36song going, "Oh [ __ ] I didn't know it had a a mixer in it and a I can split these into tracks. Holy [ __ ] what do I
549. 1:32:45do with that?" And you look up and then I'm doing something else. You're like, "Oh, that reminds me." Look over just
550. 1:32:50being in the conversation is what this is about because you can't keep up with it. You can't master it.
551. 1:33:01The only thing you can do is kind of be on the ride. Like we're we're what this is is like a little surfing
552. 1:33:09club. There's this tsunami coming. This [ __ ] radical, you know, never seen before on Earth tsunami coming fast.
553. 1:33:23And we're like renting surfboards and people like I don't know how to surf. It's like, "Okay, it's whatever. Come
554. 1:33:28on. We'll teach you how to paddle. All you have to do is this. Okay, I can do this. I'm tired. Keep doing this." Right?
555. 1:33:38Some of us are surfing the little waves. Some of them are surfing the big waves. But we're all just out in these [ __ ]
556. 1:33:45choppy, weirdass waters waiting for this thing to come. Does it come this year? Does it come next year? Does it come
557. 1:33:50five years from now? I don't [ __ ] know. No one knows. But I know that I'd much rather be on a surfboard in the water than I would be
558. 1:33:59on the shore. So that's what this is about. So just keep coming back. All right. All right, Kyle, you spent three years teaching us.
559. 1:34:13You'll be teaching. Exactly. Finding our way together. That's it. All right. All right, everyone. Um, have a
560. 1:34:23good night. Tomorrow is Thursday evening. I don't think anything's going on. Yep. Tomorrow night. All right. See you guys later. Bye.