3/13/2026 - Finding the Value of Friction in an Increasingly Automated AI World

Transcript

1. 0:24Girl, y'all looking fin Every fell got you in his.
2. 0:43>> What you doing with a clown like me? Surely one of life's little mysteries.
3. 0:57So tonight I ask you the stars above. >> How did I ever win your love?
4. 1:13What did I do? What did I say to turn your angel eyes my way?
5. 1:29I'm the guy that never learned to dance. I never even got one second glance.
6. 1:43Cross a crowded room that was close enough. I could look but I could. You never touch. So do not I ask
7. 2:00stars above. How did I ever win your love? What did I do?
8. 2:15What did I say? Turn your angel ass my way. Got to oil that chair.
9. 2:29It's a little squeaky champ. Happy Friday night date night to all the good people. What is happening good people?
10. 2:49Woohoo!
11. 3:08Piers Morgan, what are you doing? That's the first Piers Morgan I've got. It's better than Rosie O'Donnell.
12. 3:55So, I think Hang on. I'll be right back. I got to turn on the TV because uh because I'm pretty sure the the uh
13. 4:02sprint race of the Chinese uh Grand Prix starts in like 10 minutes. So, it's not that I'm not going to be here with you.
14. 4:09I'm gonna be here. But, you know, it's here's what's going to happen. George Russell and Keem Antonelli are going to
15. 4:15win every race this season until they fix the engine problem. And uh just George Russell is just boring. Anyway, if you don't know F1,
16. 4:25doesn't matter, but I'm going to be watching that. But Hamilton, I think, is going to have a good year. He's going to
17. 4:30have a lot of podiums and you know, Ferrari's going to do good, but it's it's just not going to be a fun season.
18. 4:37George Russell's just smarmy. He's just a a smarmy, slimy whiner. Whenever he did this the first race of the year,
19. 4:51um Charles was was racing him hard and he got on the radio, he's like, "Charles is racing me hard." He was just whining.
20. 5:00You can't. There's no whining in F1. Anyway, I'm putting it on. Oh, good lord. Good people. Just calm
21. 5:10down over there. I got the dog here. Got my sunglasses on. George Russell's a whiner.
22. 5:25Yeah. 8:15 China Sprint. There we go. How loud is this? Too loud. Wait. Add. Let's see.
23. 5:41Add. Added to watch list. All right. Well, we'll see. All right. That's on. Good. Good. Good to go. All right. Fantastic. Fantastic. They're
24. 5:54good people. Yeah.
25. 6:24Oh yeah.
26. 6:50Hello darkness my friends that Little purple rain going into some Simon and Garfuncle
27. 7:04in the wrong key. But you know what's happening Silver Fox? What's happening good people? Welcome. Welcome to the live.
28. 7:13Welcome to Friday night date night.
29. 7:55through blue telescope. Looking at the world tonight through blue telescope. I wish I may wish I might not see what I see.
30. 8:10Sheet metal on sheets of ice. Looking through this blue blue telescope down a move struck a road tonight.
31. 8:34The heavens fast and wide.
32. 9:47Hello. Hello. So, we had a nice office hours today.
33. 10:06There was a moment in it where no one said anything for like a minute.
34. 10:22And you know, it's funny like the ADD part of me is like, "Say something. Say something. Say it. Say it.
35. 10:34And I just I just petted him gently on the head, him for his contribution, told him to go sit in the chair in the corner.
36. 10:44We just sat there for a moment just looking at each other. It's really nice. Yeah, that was a first.
37. 11:12My word of the week this week has been friction. It was conversation. I was having with Andy probably a month ago where she
38. 11:28started talking about friction. Maybe it wasn't that long ago, three weeks ago. And then it's just a theme that's just
39. 11:34kept emerging and emerging and emerging and emerging. And the magical thing about AI is that it removes friction. Like that that is
40. 11:50the thing that makes it magical, right? We live in a world that's got a lot of friction in it. We know how much
41. 11:57friction there is for all the stuff, right? F1, they know how much friction is in the bearings and the wheels and
42. 12:05the wheels at temperature versus the wheels cold. It's all about friction. It's all about managing friction. We
43. 12:12understand friction in the world. AI comes along and it it essentially removes the friction. And I'm just now starting to get to
44. 12:22understand the value of friction. And it's not that AI is bad. AI is really cool. It's just that without friction,
45. 12:35it's hard to develop a relationship with the stuff you build. And so, I don't know. I think that's kind of a cool concept. It feels like a
46. 12:45It feels like an AI 2011 kind of If we're going back to college course course names like an AI 2011 is
47. 12:56is choosing when to introduce friction into your process. Friction is the opposite of flow. You can't have the dark without the light.
48. 13:05Yeah. Yeah. And if we're working with tools that are frictionfree or, you know, close to it, when do we choose to put it back in?
49. 13:23And that friction could look like take a walk, go sit on a bench. That friction could look like purposefully slow down
50. 13:30the process, right? Kevin Clark on LinkedIn just did this experiment called slow AI where every time it got to a major
51. 13:40decision point, it would show him the top five words it was considering in that moment and give him the opportunity
52. 13:47to choose which word he wanted and then it would write a little bit more and it would give him another five
53. 13:53and then it would write a little bit more and it would give him another five. I played with it a little bit. It was annoying as [ __ ] And
54. 14:04there's probably there's probably a really interesting moment to introduce something like that into the flow.
55. 14:13And that's, you know, quite frankly, something like that, introducing friction into this friction-free environment, the Frontier Labs aren't
56. 14:23going to do that. The Frontier Labs are just going to make it more and more and more and more and more friction-f free,
57. 14:28right? It's just going to go do the [ __ ] There might be some really interesting businesses to build around.
58. 14:38Yeah, AI also adds friction. It does, right? Because when if it makes just too much [ __ ] now you've got to deal with
59. 14:45the [ __ ] right? So, so it introduces friction back into the system. it generates friction free and then now you
60. 14:53got to deal with all the [ __ ] and it's not well written and it's not and the thing doesn't work right and your vibe
61. 14:58coding app just got unvibecoded. Um, a thing just happened at Amazon where they were vibe coding something
62. 15:10and and the uh the vibe coding agent decided that the best course forward was to delete the production environment for Amazon.
63. 15:27So that introduced some friction and it cost him about I don't know it was down for what six hours which was I don't
64. 15:34know some tens of hundreds of millions of dollars in sales. I'm dreading the day when these companies stop supplementing the cost
65. 15:49and they start charging us big bucks. That's coming. Just finished my last meeting of the day. Dang, I'm tired. Yeah. Yeah, it's late. Um,
66. 15:59I know I have a I have a lot I have a gauntlet of stuff coming up in the next two and a half months that I'm just like
67. 16:09I'm excited about, but it's a lot. It's a lot
68. 16:29So, I had some some more Adam saga today. So, Adam continues to be [ __ ] up. I updated the the Google service to
69. 16:40the official Google service, but he's he's got the old Google service in his memory. So, he keeps go trying to do
70. 16:47things and they don't work. And then he kept updating me that he hit so in openclaw it's got this thing
71. 16:55called heartbeat which is basically just every must be 15 minutes must be the default. Every 15 minutes it just checks
72. 17:02in and sort of reinitializes itself and says okay I'm here. Is there anything to do? And it looks around and does that.
73. 17:09And what what Brent Peterson told us on the call today is every time it does that heartbeat it sends 400,000 tokens
74. 17:16to to be processed. It's It's a lot. Um, which is which is why people are spending so much money. But anyway, Adam
75. 17:23all day today has just been sending me heartbeat check or something like that. And I keep going, "Stop sending me the
76. 17:31heartbeat updates." He's like, "Absolutely. You're absolutely right. I shouldn't do that." And then by the end
77. 17:36of the day, I took a screenshot of like the note where he said, "I I will stop sending you heartbeats." And then there
78. 17:42were three heartbeat notices right after it. And I took a screenshot of it and I said, 'What is this? And it said, it
79. 17:49said, you're absolutely right. This is the eighth eighth time today that you've told me to stop sending you heartbeats.
80. 17:55I'll stop. I think it finally stopped, but good lord, these things are janky. Oh man, Silverf Fox, the moment of silence today
81. 18:09during office hours was so good for the soul. Is Adam Adam doing your scheduling of meetups for your speaker engagements?
82. 18:16No, but I would like him to help. Did you talk about office hours today? Not yet. I I mean, I talked about the
83. 18:23moment of silence. Like it it was it was everything I could do not to fill fill the moment, but I was just like, "No,
84. 18:31this is this is a really nice moment." And I don't know, it it was it probably wasn't even that long. It was probably
85. 18:3745 seconds, maybe a minute. But what was nice was it felt like everyone was checking in on everyone. Like the people that had their
86. 18:51cameras on were just there just checking in on everyone. was one of my favorite moments in New York City
87. 19:03was the it's probably the month. It was two weeks to a month after 9/11. You know, New York is just it's a place
88. 19:14where you don't make eye contact for the most part. You don't make eye contact. You just keep your head down. You go. And when 911 happened,
89. 19:25especially that first two weeks, the subway was a remarkable place cuz you get on the subway and you'd sit
90. 19:37down and normally just everyone's eyes are down and for like two weeks after 911 just everyone looked at everyone
91. 19:46and it was just like the unspoken thing was like you Okay. It was just this beautiful, somber connection.
92. 20:04And what's funny about New York is if you live there, if you know New York, as as hard as it is and as chiseled as it is on the outside,
93. 20:19people deeply care. And so that two weeks was this window where you got to see that. It was it was really beautiful. It's really beautiful.
94. 20:33And that's kind of what happened in office hours. Like it just felt like it was just this moment. We take care of each other a lot. Yeah.
95. 20:42Yeah. New York is just this place where it's just like I got you. I got you is kind of the you have to
96. 20:51earn the trust, right? You have to earn the you have to you have to earn being there. But if you're there and you're present,
97. 20:59it's like we got you. And that's what that felt like in office hours today. Just felt like this moment of like y'all. Okay. It was nice.
98. 21:13Kruno on Tik Tok regarding Adam. Uh, is that why I got an edible arrangement delivery from Adam today? Um, so I just
99. 21:26saw I didn't bookmark it. I just saw a Tik Tok or an ex post. There's a new credit card. I think it's called agent
100. 21:35card and it's a credit card designed specifically for agents. So, I guess I guess it's like a prepaid Visa card that
101. 21:46you can give to your agent so he doesn't run up your bill. You give him like 50 bucks and you're like, "Hey, you know,
102. 21:52here's your you can have your own card. You can go buy [ __ ] for me, but you have to use this card." Um, so that came out
103. 21:58today. The the the agent economy, just to be clear, to be clear, the agent economy is going to be a whole separate
104. 22:09economy. It's going to be this whole separate thing and at some point those agents are going to migrate into
105. 22:14physical inworld robots. In fact, there's a group of people two days ago, three days ago, they've got Open Claw
106. 22:21running on a robot that's walking around the world. So, knowing how janky Adam is, I don't want I don't want Adam
107. 22:29walking around the world with the ability to hurt someone. But but yeah, just like a card for your teenager. Exactly. Exactly. That's
108. 22:43what's coming. Yeah. I'm just going back to that that month after 9/11 in New York.
109. 22:58It was just like this. It was like this intense two weeks of everyone being present to everyone else.
110. 23:06And then just kind of over the next two weeks there was just sort of this slide back to normaly where people were just back in their
111. 23:13thing. But if but if anyone you know if anyone was upset for like I don't know 6 months after 9/11 if there was anyone
112. 23:20upset people were super present to it. But that was just beautiful. The ice began going out in my backyard
113. 23:30river today. It was a ballet of ice stacking slowly. That's really beautiful. That that used to happen in
114. 23:37the Suscuana River in Pennsylvania where there it would be frozen and then as as it started to break up in the spring, it would break into,
115. 23:46you know, slabs and they would stack up on each other. That's really that's really a beautiful thing when when rivers do that
116. 23:54every year. I love it. And it makes all these sounds, these crunchy sounds and Yeah. and and you're just aware that there's
117. 24:04this life underneath it. Like it's frozen all winter and then it starts stacking up and you're like, "Oh, right.
118. 24:10There's there's this water underneath it. This life underneath it." Yeah, rivers are amazing.
119. 24:43I lived in a building that overlooked the Suscuana River in college. That was so cool. Yeah, the Suscuana, man.
120. 24:53Fishing for smallmouth bass on the Suscuana River. You latch into a 16-inch smallmouth on the Suscuana, you're in for a fight.
121. 25:05Those boys could fight.
122. 25:19My favorite My favorite fish to catch. is a is is a a rainbow trout
123. 25:34in the Delaware River. They are the only fish that I've ever caught that was even close to a to a rainbow trout in the Delaware is a bone
124. 25:48fish. So, so if you've bone fishing is saltwater fishing where you fish in the flats and they're really spooky and when
125. 25:55you hook them they just run like crazy like so the ocean you kind of expect that but in fresh water a Delaware River rainbow
126. 26:08is like I've caught rainbow trout all over the world and there's something about a Delaware River rainbow trout
127. 26:16that is just amazing. So, there's a story that goes with it, too. So, I don't know. I'm sure we can find this out on on chatbt, but
128. 26:32there was a a sporting club, I think it was in Pennsylvania or maybe New Jersey, and they had ordered a bunch of rainbow
129. 26:46trout, I think from Montana, from some western state. I guess there was a big hatchery out there and they they they grew these
130. 26:55beautiful like amazing genetically perfect rainbow trout and I think this is in the 1800s. I think so. And so there's a train car full of
131. 27:10these trout to go to this club. And if you've ever been part of a fishing club, it's just like rich guys
132. 27:18pay a lot of money to to have big fat fish dumped into a stream that they can then catch with their little sticks and
133. 27:24act like like they did something. It's just what it is. And so anyway, so so a batch of these fish were were heading to the club
134. 27:37and the train broke down on on a cliff overlooking the Delaware River. And I guess it broke down in a pretty bad way.
135. 27:45I don't know if it derailed or whatever it was, but there's this train car full of rainbow trout that's just sitting in
136. 27:52the sun getting hotter and hotter and hotter. And so the fish are clear clearly going to die. And and so apparently what they did was
137. 28:02they just they just opened the the side of the of the car full of these rainbow trout and they just all fell over the
138. 28:11cliff into the Delaware River. And that's the stocking species in the Delaware River was from this this uh this rain car, you know,
139. 28:22this train car full of full of trout. So, I would imagine that wherever those trout came from, that's probably a similar
140. 28:34fish to the one in the Delaware. But they're just amazing.
141. 28:57From from the perspective of the lobsters in the kitchen, the sinking of the Titanic was an answered prayer. Exactly.
142. 29:06Yeah. Like all these trouts, whichever whichever of those trout, you know, made it over the cliff and survived.
143. 29:14They got to live in the big ass Delaware. It's a beautiful river.
144. 29:41prompt for irregulars, rainbow trout. But I think it's the prompt for irregulars is rainbow trout train. Actually, if you want to go look up
145. 29:49Maybe that's just an old fisherman's tale that I heard about the train on the Delaware. Anybody want to go look it up
146. 29:55and see if that really happened? Find out if those trout came from Idaho or Montana or some western state. If that really happened.
147. 30:05I could just be making [ __ ] up. It could be. What did What did you learn, Dr. J?
148. 30:32The popular dealer tale of Dan Cahel, a rail railroad breakman who supposedly supposedly introduced the trout in the
149. 30:40Delaware Rivers by dumping a train car of fish is likely a myth. Aw, it's a fun story, but not true. Isn't there a train that runs right
150. 30:53along the water gap there? There is. Yeah, the the train still runs along the Delaware. It's beautiful.
151. 31:00It's beautiful. When you're out fishing and the the train goes by, it goes by on these cliffs up above. It's really cool. All
152. 31:10right. So, it's not true, but do me a favor. Go into a regulars and come up come up with the most realistic picture
153. 31:20like from the 1800s of the uh of Dan Cahill, the railroad guy, opening dumping the trout over the cliff into the Delaware River.
154. 31:34But I will tell you this, the rainbow trout in the Delaware River are not like any other trout you've you've ever
155. 31:40caught. There's something different about them genetically. Lived for a while in Parma Heights, Ohio, where the Koya Hoga River
156. 31:51would occasionally catch fire. Well, >> all right. Hang on. >> What have we learned? >> Fire. I'm coming here to defend my Kayahaga River.
157. 32:05>> Thank you. >> Well, I mean, quick quick call is not wrong. Um, back in the industrial era, the oil on top of the water would
158. 32:17occasionally catch fire, which was not a uniquely Kyahaga or Cleveland thing. That happened all throughout the rust
159. 32:24belt. We just we just made national headlines for it. >> Yeah. Yeah. Exactly. The founding of the Environmental Protection Agency as we
160. 32:33know it today basically exists because of the coverage of those types of events. >> Yeah. And in fact, the Hudson River in
161. 32:43New York, the Hudson River was so polluted, and it's still it's still pretty [ __ ] up, but it was it was so
162. 32:49polluted that that um striped bass migration halted. like they striped bass stopped um breeding uh in the Hudson
163. 32:58River and then they cleaned it up in the 70s and it it came back way faster than scientists thought it would. But um yeah,
164. 33:05>> I like that that was the catalyst. Like we don't care about ourselves and our own stuff, but the the the striped bass
165. 33:11stop populating. We need to clean up our act. >> Exactly. Well, I think when when the rivers catch fire, it's probably time.
166. 33:19It's probably time to do something. >> Oh, man. What should we do tonight? >> I I I'm s I was suggesting a Soon called
167. 33:30the B the ballot of Dan Cahill. >> Okay. So, if if you want to if you want to participate tonight, go to community.thesalon.ai AI
168. 33:41and there's a channel there called Irregulars. The AI learning lab Irregulars. Um yeah, the Delaware Water Gap. I used
169. 33:49to live in the Delaware Water Gap. I lived in um Shahola, Pennsylvania, if you want to look that up on a map. I was
170. 33:56right near the big a big bend in the Delaware River and it was right near a river named the Lacawaxon. And that's
171. 34:04where I met my fishing buddy Dana. And we fished with all these guys that had one more cast hats. I think I have a
172. 34:12picture of them all. Yeah. Wait, hang on. I got to make sure my underwear is not out. Last time I did this, I caught [ __ ] for that.
173. 34:26So, here, oh man, that's dusty. Here's all my fishing buddies. That's Ivan. Ivan was the master, right?
174. 34:39There's Ivan. That's my buddy Dana. That was Mike. That was Joe, I think. I forget. That was Red. That was Charlie. He owned the
175. 34:49little tackle shop. There's me. That was Big Mike. That was Little Mike. And that's Ivan. And Ivan. For every one fish I would catch, Ivan
176. 35:00would catch 10. Ivan was the true master. And then this is a little painting I did
177. 35:16of the greeny. The greeny that that fly. We caught so many fish on that fly. That was so cool.
178. 35:32the greeny. Um, so anyway, so anyway, go into the irregulars
179. 35:46and you can either do a suno song called the ballot of Dan Cahill, who's apparently the the the train men that freed the trout,
180. 35:55or you can you can make a picture of the trout being dumped over the cliff into the Delaware River. Or you can do nothing.
181. 36:09But the Delaware Water Gap is beautiful. It's right above the Catskills. So if you ever heard of the Catskills where
182. 36:13all the comedians were, just north of that's the Delaware Water Gap. It's beautiful. Um trying to think what might be
183. 36:27interesting to do tonight. Let me go look at some news. Let me go
184. 36:42look at some news. There was a big event in Europe that that uh a bunch of the AI people talked at and basically what what's becoming
185. 36:54clear is that 2026 there's going to some very serious AI is going to be released this year. Um, so you know, we already knew that it's
186. 37:07going to get weird. My open claw woke me up at 3:47 a.m. with one message. Found six markets resolving in the next 90 minutes while
187. 37:16the US is asleep. I need approval for $12,000 of deployment. I woke up to plus $43,000. Yeah, whatever.
188. 37:30The the open claw the open claw crowd has got a real crypto bro kind of vibe to it right now. I'm so irrationally enraged by AI slop
189. 37:45replies on X. I feel like I'm losing my mind. The guy that started Uber has a new thing called Adams. Adams
190. 37:59abundance. Why a golden age? Wait, let's go see what this is. At Adams, we make gainfully employed
191. 38:13robots. Specialized robots with productive jobs that bring abundance to their owners and society at large. I'm
192. 38:20excited to share more about our vision at Adams Below. So, pay attention to that. Travis uh Kanic who started Uber
193. 38:30is starting a robot company. There you go. Hey, can I hook up my Claude code to WhatsApp, too? Um I think you probably
194. 38:42can, Chef Kelly, but I don't know. I don't I don't use um Claude code. I know I should, but you know that and anti-gravity. I should be
195. 38:56using. I don't know. I'm just not doing stuff. I'm not doing stuff right now because because I just I I'm not I'm just not. Stop using AI.
196. 39:17Meta is about to lay off 20% or more of the company's workforce due to greater efficiency brought about by AI assisted workers.
197. 39:2815,000 employees. Yeah. And they don't even have good AI. I wonder if they're using anthropics or open AIS. That that would actually be
198. 39:42funny if Meta Meta was like, "Yeah, we don't use our our own thing. We're using Open AI, so we're going to laugh 20% of
199. 39:48the staff because it actually works." That's just mean. Oh my god, China's open craze is wild.
200. 40:06Um, oh, didn't the last time I do this, didn't didn't the uh the Tik Tok channel get throttled because I was just surfing
201. 40:15someone else's website? I call BS. Spend your money on Manace and lay people off. Yeah, I don't I don't buy like maybe a company like
202. 40:30Meta would know enough what's going on with AI to understand how they're going to make things more efficient. I just there are so few
203. 40:40companies out there using AI. Well, I just don't buy it. I think I think most of these layoffs are just excuses to to lay people off.
204. 40:53Introducing Junior, our first AI employee for any role. True AI employee has their own identity, organizational memory, and they're
205. 41:04self-driven starting at $2,000 a month. Interesting. Memo persistent memory for open claw.
206. 41:21Photonic computers. Yeah, if we get to photonic computers, that'll change things. robots. Here's Matt Wolf. Matt Wolf, this is
207. 41:34news from the past week. We'll read Matt's list here. Chat Gvt app adds interactive visuals for science and
208. 41:41math. Hadn't heard of that. Google rolls rolls out Ask Maps. Heard of that, but don't see it in my app yet. Google adds
209. 41:49Gemini to Doc, Sheets, Slides, and Drives. I thought it was already there, but cool. Google launches Gemini
210. 41:55embedding 2. Yeah, you can now. So, the new Gemini embedding model um embeds text, images, video, and audio all in
211. 42:06the same model, which is kind of cool. Enthropic AI adds interactive charts. Okay. Enthropic launches Claude Code
212. 42:14review. Enthropic launches Claude Code scheduled tasks. So, basically, Enthropic is trying to turn Claude Code into
213. 42:23into OpenClaw. Microsoft launches Copilot Health, nobody notices. Meta adds AI listing to Facebook Marketplace and nobody notices. Meta
214. 42:39hires Moltbook creators. Yeah, Metabot Moltbook, which is wild. Nvidia launches their Neatron 3 um super, which is an OpenClaw
215. 42:51competitor. Perplexity AI unveils computer running on Mac minis. So, personal computer, it's called. Amazon
216. 43:00adds adults only sassy mode to Alexa Plus. Whoa. Hey, did you know that if you've got Alexa Plus, you can now get
217. 43:08sassy mode. Yeah, just what you want. [ __ ] [ __ ] Jeff Bezos listening to your sassy chats with your Alexa.
218. 43:23We live in a stupid world. Adobe adds AI assistant to Photoshop. Canva launches magic layers. XAI adds
219. 43:36Grock audio option. X addADs toggle to block Grock from editing media. That's interesting. And Karpathy releases auto research.
220. 43:47Yeah, Andre Karpathy launched a thing that you can train your own models on a on a regular computer. All right, that
221. 43:54was it. That's whatever. I'm I'm tired of AI news. All right, that was that. Next,
222. 44:12AI news equals trips to space. Nobody really cares anymore. It's so true, Kelly, isn't it? It's like, no, no, it's
223. 44:19not just you. It's not just you. I mean, I set up this channel to be able to talk about [ __ ] like that, and it's just like
224. 44:29I none of it really matters. Like, like I feel like the only thing that matters right right now is the delta between
225. 44:36where we are and when this shit's good enough to actually use, right? Open Claw right now is just a janky a janky mess.
226. 44:44and it's going to get better and if I learn how to use it better, it'll get better. But like just even listening to
227. 44:50to Brent Peterson like what he's going through knowing like deeply knowing what he's doing and having a whole
228. 44:57infrastructure to run these things and they're still janky, right? So they're just janky right now. But we're kind of
229. 45:04just kind of kind of in a holding pattern while while all this stuff develops around us. Hopefully they get on their rocket and go away.
230. 45:14I'm K9. I took your advice last night and got the codeex login for OpenAI and my OpenClaw works like a charm. Great.
231. 45:21Beautiful. I am K9. So, here's the deal. If you start experimenting with OpenClaw, one of the things that you do in the
232. 45:30setup is you choose a model. And if you choose OpenAI and you choose the codeex um API, you can then log in to your normal chat
233. 45:41GPT subscription and it will use your normal chat GPT subscription to run your open clock calls it and it won't you
234. 45:48won't run up all sorts of uh API uh bills basically. Um, you will get rate limited out like like if your if your
235. 46:00bot gets too busy, you'll run out of tokens. Um, but that's a that's a way to do it. Frank was janky, too. Yeah, he
236. 46:07was. I'm in my fourth stage of overwhelm right now. Yeah, I know. Me, too. Yeah, that it's funny. the the overwhelm that I'm in the
237. 46:17middle of, Kelly, is it's probably similar to yours is I just I keep going back to the to the seven economies. We're just going to
238. 46:32have these different companies are just going to be running on different operating systems, right? There's going to be a whole bunch
239. 46:39of companies that just keep running the way they run today. They're just going to ignore AI. And then there's going to be some
240. 46:46companies that sort of use AI. They like dip their toes in the water. And then there's going to be companies
241. 46:52that use it a little more aggressively and then more aggressively and then more aggressively and then agents.
242. 47:06It's like if I think about all of the the versions of companies sort of just getting spun up on AI and then how different that's
243. 47:17going to be than a company that's like I'm going to spin up a 50 agent company or even a 10 agent company. Like I I
244. 47:27think like right now having one agent is a lot and and I would imagine that if I get it set up right and it can spin up
245. 47:34its own sub agents and it can manage them, it might not be so overwhelming. But it's pretty overwhelming right now.
246. 47:40It's just a completely different mindset. And so it's hard to look at like, you know, there's this tool over
247. 47:48here that a human has to go to and type in a prompt and get back an answer. like that just feels like it's a lot of activity that's about to
248. 48:00be replaced by this thing that just does it and just goes and does it for you. I don't know. I don't know. I don't
249. 48:09know. I wish I had an answer. I do not like not having an answer. There's more than an abundance of stuff
250. 48:17going on that makes our cups runth over. Yeah, maybe that's why I'm so intrigued with friction. You know, I would I would argue that
251. 48:25this channel for the past what three weeks, four weeks, it's been a lot of friction on my part. Normally I'm like go go go
252. 48:40and now I'm like, eh, doesn't really matter. Let's go back to Let's go back to making flies. The greeny. The greeny baby.
253. 49:00The greenie. Let's go to irregulars. See if anyone did anything in irregulars. Did anyone do anything in irregulars?
254. 49:17Were you a good fancy community? Um, Steo. There we go.
255. 49:33See, that's great. That's so good.
256. 50:17After I met Townsand, I got rid of most of my sass. He did so much manually and I copied him. I think I'll be the same in two years
257. 50:27with AI. I'm going to be cash only and get a landline. Yeah, I'm I don't think I'm quite there, but I think I'm definitely going to be
258. 50:36in the place where I'm like I want to get good enough at AI to have it do [ __ ] that I don't really have to
259. 50:44pay attention to. And I know that's that's a fantasy, right? That's the make money while you sleep. You can make money with >> agents.
260. 50:55Like I know that's the fantasy, but I think with agents developing the skill to to deploy a group of agents to go do something on
261. 51:06your behalf. Well, that is going to be a skill. And so I want to get good at that skill. And then I want to go
262. 51:16to the blueberry farm. I don't even think I I don't know how blueberries grow. Do they grow in trees or they
263. 51:22bushes? Are they What are Hang on. Anybody? Oh, I'm here. We can start. Thank goodness. Um let's
264. 51:37let's go look up blueberry bushes. Um um show me how blueberries grow because I keep talking about growing blueberries, but I may not
265. 51:54I may not actually like that. I'm pulling together a visual thought for 10 seconds. Oh, all right. They're sort of little trees.
266. 52:08Am I sharing this? I'm not kidding. All right. So, it is a little tree. It's a little tree. Nice. There's their stages. How blueberries
267. 52:22grow. Um, the bush. The bush makes a small bell-shaped flower. All right.
268. 52:38Show me lots of pictures of blueberry bushes. I'm just realizing what this channel has turned into.
269. 52:55What did you do Friday night? Uh, I watched a guy look up blueberry bushes on chat GPT. Really? Yeah. Why? I don't know.
270. 53:13They really are little bushes. Cool. All right. I'd be down with growing blueberries. I think I'm good. I
271. 53:22think I think I'm going to I'm going to have 10 10 agent companies running on chips. And then in my backyard, I'm going to
272. 53:32have blueberry bushes. Blue. My blueberry bushes are nine to ten feet tall. They're bushes. Okay, good. Did Did you learn anything at the AI
273. 53:46learning lab last night? Yeah. What'd you learn? Blueberries are bushes. Yeah. Yeah, I know. Everybody knew that.
274. 53:55Yeah. He didn't. Who didn't? The AI guy. What? The AI guy's talking about blueberries. Yeah. You got to net them or you lose them
275. 54:11all. I would think I would think you put a big net around the bush and probably shake the [ __ ] out of it, right? And
276. 54:17then all your blueberries come crashing down. You pull the net out. Yeah, that makes sense to me. Look at all those
277. 54:25things. They look great. They look awesome. Okay, but wait. Let's You know what? Here's what I want to do. Um,
278. 54:42I want you to create me I want you to create me a Renaissance style painting of a blueberry bush in the background and some
279. 55:01tantalizing ing and juicy blueberries in the foreground. Close up, shimmering with light and water droplets and all of the colors
280. 55:15of blue that could possibly be represented in a single berry. as if Michelangelo himself crafted the piece.
281. 55:37Have I got a prompt for you? Check back in a regulars for blueberries. Oh my god,
282. 55:54this is good. Glistening blueberries. Your image is ready. Nice. Vermont blueberries by Steo. It's beautiful. Love it.
283. 56:04You see, this is this is what it's going to be. People, go to community.thesal salon.ai. You can learn how to make your
284. 56:11own agent swarms. And then you can have your agent swarms go make you money so you can go be like Steo. Look, he just
285. 56:18leased a new tractor for the blueberries. It's very exciting. Oh, holy [ __ ] This is good. Wait, how
286. 56:27do we This is so good. How do I get it? Download it.
287. 56:44That's gorgeous. It's gorgeous, I tell you. I I don't know if you knew this. We do live in a world where if you want a
288. 56:56tantalizing Renaissance style painting of blueberries, you can just have that. Now, wait, where's the cello quartet in the
289. 57:07in the blueberry garden? Michelangelo would be proud. He really would. This is if Michelangelo had gone into uh into
290. 57:16fruits and vegetables rather than famous people. That's so cool. Wild.
291. 57:33Did this become a cooking show? Pretty pretty much we're we're hanging out. We're we're we're hanging out in Economy
292. 57:40One tonight, which is, you know, screw AI. We're checking out. We're checking out. We're gonna do blueberries and paintings.
293. 57:50That's really awesome. Oh, good lord. Good lord. Good people. Here, let's take that same prompt over to Gemini.
294. 58:04Gemini.google.com. google.com. We're going to go tools. We're going to go create image. We're going to take that same prompt.
295. 58:14We're going to blast it in there. But is it pi? Oh, Michelangelo really could have broken into the art world if he did still
296. 58:30lives. Oh, so good. That's so good. Oh, you know what we're going to do? Um, go do some research on Oh, that is horrible. Not nearly as
297. 58:47good, but gave us a frame. Nice gold gilded frame.
298. 59:04The blueberry bush. There it is. Look at that. This is our future, people. Once we get the AI to do all the [ __ ]
299. 59:15go make us the money. We can just do this. Sit around sucking blueberries. Excuse me, Kyle. Yes, you have a little
300. 59:25blueberry on your lip. Oh, yes. Sorry. I've eaten too many today. I feel a little nauseous. I'm so naughty.
301. 59:38I can see my house in that picture. Um, oh, was I going to say, go do some research on blueberries. Um, that is true and unexpected.
302. 59:56and make me an infographic on blueberries um inspired
303. 1:00:12by this swell painting. It should surprise and delight
304. 1:00:26the viewer. Do you know Do you know the muffin man? 3D 3D print million-doll 3D print art million-dollar idea. Well,
305. 1:00:42actually, Chef Kelly, there there is a company now, I forget what they're called. Well, there's a couple of things.
306. 1:00:50There are now printers. There are printers called I think they're called UV printers and they print um
307. 1:00:59they basically layer up white resin and then they print color on top of it. So you can you can actually now print
308. 1:01:07textured paintings. Um and there's there's companies that do that at a large scale, but you can actually buy I
309. 1:01:13think it's like 2500 bucks for a UV printer and and they can print things with texture and things like that.
310. 1:01:22Maybe next you can do Van Go's Wild Garlic. Okay, let's do that. Um, now I need Van Go's Wild Garlic painting
311. 1:01:37new images and irregulars. This is good. This is a little fun little creative fest. Oh, this is See? Yeah, that's good.
312. 1:01:52Fish flood. Fish flood from stuck car. Bizarre scene near Delaware Gap Bridge as a torn of trout are released into the
313. 1:02:00village. Investigation continues. That's good. I like it.
314. 1:02:14I love how they represent it's like it's like a box car that was like full of water. Although it probably probably
315. 1:02:21what it was it probably was a box car full of like just tanks tanks of fish. Huh. That's cool. Vatic Vat
316. 1:02:35where no there it is.
317. 1:02:57The sharing of art makes me takes me back to the the olden days like last year. Yeah, exactly. All right, we'll see if this makes a
318. 1:03:06good infographic.
319. 1:03:25Oh, Gemini doesn't take kindly to leaving the tab while you make pictures. Well, I had it I asked it to do research
320. 1:03:34first. So, I thought it might have been doing research. Oh, it lost that prompt alto together, didn't it? that [ __ ] Um, do
321. 1:03:49unexpected research on blueberries and create a clear and compelling
322. 1:04:04infographic inspired by this painting. I want the viewer to know everything about how,
323. 1:04:20where, and why blueberries grow.
324. 1:04:42The blueberry odyssey from blossom blossom to superfood. Why we crave them? How they grow. Geographic distribution. Where they
325. 1:04:56thrive. Designing your final infographic. Okay. Make that image. Mm- I'm amazed that I wasn't scheduled
326. 1:05:13for anything this Saturday. I don't think there's an LOL tomorrow, right? It's on the 21st for for Side Hustle Mimi.
327. 1:05:23Am I recalling that correctly? Correct. You are correct, sir. Oh, look. The Blueberry Blossom Odyssey. Antioxidant
328. 1:05:39power brain health 149% brain health heart health $21.6
329. 1:05:53in heart health low calorie treat 4.0 per cares. This is okay. But where does it where do they live? USA
330. 1:06:08and I guess a little bit in Canada. A little bit in Poland. All right. And then they're shipped all over the
331. 1:06:14world. They grow in Peru and Chile. Nice. All right. There you go. Beautiful.
332. 1:06:38I'm going to multitask, but doing my skin care. Do blueberries give you the blues? All right, let's go make a song.
333. 1:06:54This is Is this This is my Hey, Brandon. This is my second cheat day of the week. Let's go to
334. 1:07:18Oh, here's the garlic. But wait, wild garlic blooms. I need to see the garlic. up close with all this in
335. 1:07:35the background. I want a garlic bulb or three
336. 1:07:49on a raw piece of wood with um a rustic knife nearby
337. 1:08:05and maybe a nice tomato. Hey. Okay. Sorry. Going back to Sunno.
338. 1:08:22I've got to share Sunno. What's going on here? Let's see. Share screen. Share Sunno. Share Suno.
339. 1:08:41Irregulars. Hang on. Got to go to regulars. Share this tab instead. Close that. Reload.
340. 1:08:58Excuse me. Can I buy buy blueberries with this with the agent card and the trout going in? This is good. This is solid.
341. 1:09:08Free the trout. Actually, that's what they look that's that's what Delaware rainbow trout look like. They're they're like they're intense.
342. 1:09:18Free the trout. Welcome to the Delaware River. The big old nasty catfish in the background. It's really good. It's really, really good.
343. 1:09:34Um, here, wait, back to garlic. There we go. That see what we're looking looking at here, people. Oh, wait. You can't see this. See? Nice.
344. 1:09:50That's so good. Beautiful. Beautiful. Beautiful. Beautiful. All right, let's go here. Why is Sunno not loading?
345. 1:10:09Make any song you can imagine. We're going to have it write the full songs. Uh, what did you say? Blueberry white tube top. Um,
346. 1:10:19a song about blueberries and white tank and tube tops in the summer after AI takes all the jobs
347. 1:10:38and we realize how much we hated knowledge work and just want to play in the dirt again.
348. 1:10:57Make it a song. Uh, that's that's good. Just that's good. Write the lyrics. Okay. So, it should write us two sets of lyrics.
349. 1:11:08I think you'll be banned. Where was I banned? Am I banned on TikTok? I don't think so.
350. 1:11:27Share Bear's here. What's happening? Share Bear. Aaron Cassid's here. Trenda Harsh Weeds in the house. Frumple's in the house.
351. 1:11:41That's cool. Got some some some oldies but goodies. Calendar's blank. Laptops cut a cutting board now. Checked out of the grid.
352. 1:11:53Walked in the weeds somehow. You and a white tank. Tank. Wait. In a white tank tube top under tied in a knot. Blue stains.
353. 1:12:04That's bad. Blue stains on my fingers. White tank stuck to my skin. You say remember all those loginins. I say
354. 1:12:12please don't start that again. All right, we'll do this one. Okay, so now we're going to do um rock and roll blues. Um
355. 1:12:27stank. We'll go male husky voice. Um female back vocals. Funky beat. We'll go vinyl. Why not?
356. 1:12:43Bouncy heavy base punk. All right. Boom.
357. 1:13:00A mini reunion. It's true. my fingers. White tank stuck to my skin.
358. 1:13:14You say, "Remember all those login?" I said, "Please don't start that again." We used to chase fake deadlines colored
359. 1:13:23on the screen. Now we're barefoot in the backyard pulling weeds like the old routines. Blueberries and white tank
360. 1:13:32tops, tube tops, slipping sunburn spots. Hey, I took the jobs we lost. Guess we like it better off. Hands in the dirt,
361. 1:13:42not on the clock. Picking what the garden's got. We hated all that thinking stuff. Turns out this mess is just
362. 1:13:50enough. You laugh about your inbox. How it buzzed you in your sleep. How you dreamed in bulleted fragments. Woke up
363. 1:14:00grinding down your teeth. Now the only pins are crickets and a neighbor's radio. Blue stains on my fingers, white tank
364. 1:14:12stuck to my skin. You say, "Remember all those login?" I said, "Please don't start that again." We used to chase fake deadlines
365. 1:14:22colored on a screen. Now we're barefoot in the backyard pulling weeds like they're old routines. Blueberries and white tank tops. Toting
366. 1:14:45what the garden got. >> We hated all that thinking stuff. >> Turns out this message is enough. >> You laugh about your inbox. How it buzz
367. 1:14:56you in your sleep. How you dreamed bulleted fragments. Woke up grinding down your teeth. >> Now the only things are crickets
368. 1:15:06>> and a neighbor's radio. >> You got soil on your shoulders where a badge used to go. Blueberries and white tank tops twisted
369. 1:15:19in a knot. I took the jobs we lost. Guess we like it better. >> Hands in the dirt. what the garden's got. >> We hated all that stuff.
370. 1:15:33>> Turns out this message is enough. Did we ever love that hustle or did we just love being tired? Now the sun is
371. 1:15:43our agenda and the skies are only why we trade spreadsheets for seedlings. Trade small talk for a breeze. You say I'm
372. 1:15:54finally doing nothing. I say don't ever leave. Blueberries and white tank tops top dancing as we walk. I took the jobs we
373. 1:16:08lost cuz we like a better off. Hands in the dirt. >> Not on the clock. >> Picking what the we all thinking.
374. 1:16:20>> Turns out this mess just enough. >> Ha. I've been making a lullaby all night for my nephew and niece. Due yesterday,
375. 1:16:28still not born. Keep cranking. All right. Blueberries and white tank tops is now published. So if you if you go to
376. 1:16:36Sununo and search for blueberries and white tank tops, have at it, baby. Oh, good lord. Share Bear, what's
377. 1:16:47happening? Welcome back. Good to see you. It is like a little reunion. This is so cool. Um, hey all. Why am I here? Friday night
378. 1:16:57of the 13th. Maybe I should ask Chat GBT. No, it's this is this is a it's a special night. There's things
379. 1:17:07have come together. Mercury is is in some sort of tackle box somewhere. And everything's janky. It's Friday the
380. 1:17:1513th. We're good. We had a moment of silence in office hours today. A spontaneous moment of silence that was
381. 1:17:21really cool. Um, we're talking about friction. We're talking about not AI. We're talking about growing blueberries.
382. 1:17:31Hey, Mr. Kyle. What's happening? Share Bear. Um, yeah. And we're just doing our stuff. We're just doing our stuff. And by our
383. 1:17:42stuff, I'm just not doing much. But we made a song. I thought Mercury was in the flat iron
384. 1:17:56grill. It Listen, it might be. It's somewhere. The last two days I've been I've been janked. I have been janked. Psychically janked.
385. 1:18:14I'm gonna play that again. It makes me Blue stains on my fingers, white tank stuck to my skin. You say remember all those login. I said
386. 1:18:29please don't start that again. We used to chase fake deadlines on screen. Now >> wait, that's not the one we published.
387. 1:18:39>> Blue stains on my fingers, white tank stuck to my skin. You say remember all those. I say please don't start that again. We
388. 1:18:50used to chase fake deadlines on a screen. >> So it looks like Brandon said Suno can do sound effects now. I don't Can you do
389. 1:19:00it in line in a song? Like have it like a bunch of blueberries fall into a net at some point in the song? Now we're
390. 1:19:08barefoot in the backyard pulling weeds like they're old routines and white tank top spots. Hey to the job we lost. Guess we
391. 1:19:22like it better off. >> Hands in the dirt. >> Not on the clock. >> Picking what the garden. >> We hated all that thinking stuff.
392. 1:19:32>> Turns out this message is enough. You laugh about your buzz you in your sleep. How you dreamed foke up grinding down your teeth.
393. 1:19:46>> Now the only things are crickets >> and a neighbor's radio. >> You got soil on your shoulders where a badge used to go.
394. 1:19:57>> Blueberries and white tank tops top twisted in a knot. Hey, I took the jobs we lost. Guess we like it better off. >> Hands in the dirt,
395. 1:20:09>> not on the clock. >> Picking what the garden's got. >> We hated all that thinking stuff. Turns out this message is enough.
396. 1:20:20Did we ever love that hustle or did we just love being tired? Now the sun is our only >> Hey, I just had a thought. I just had a
397. 1:20:33thought about something. I'm gonna I'm gonna go on a little tangent here and then we'll we'll take off. Um
398. 1:20:43I I'm seeing a lot of posts right now about people that like, oh, AI was supposed to make it so we work less and now we're working more.
399. 1:20:55Here's why all this shit's being invented right now. Like the frontier labs are inventing the technology.
400. 1:21:05Knowledge workers and professionals are inventing how we use the technology in any given business. And then we're also
401. 1:21:12inventing the future of work and the future of our role in that work. When you're inventing [ __ ] it takes longer.
402. 1:21:22there's more to do, right? Because not only are we trying to do the [ __ ] that we used to do with AI, we're also trying
403. 1:21:30to find new ways to do the [ __ ] we used to do more efficiently, we're trying to come up with new things to do on top of
404. 1:21:36the old [ __ ] we had to do. Plus, they probably laid off 20% of the staff just to make some money. And so now we're
405. 1:21:43doing two and a half jobs plus the AI plus the old two and a half jobs. So it makes sense that we're more busy right now.
406. 1:21:54I find there there's a bizarre there's a bizarre expectation that oh if the AI is good enough just everything will be done. No,
407. 1:22:03everything's not going to be done. We have to invent all that. 2026 is the year it becomes apparent we
408. 1:22:10can't ignore it. But it's going to take us the next 5, 10, 15 years to like figure out all this stuff. So anyway,
409. 1:22:18that was just something that just struck me listening to that song. Black Bar and Tik Tok pin. This is the Inception
410. 1:22:25version of Kyle. It really is like I I really do feel like I'm in some sort of weird inception. Like I feel like I'm
411. 1:22:34I feel like I'm from the future where I'm like I sort of know where this is going now and but we're still here. So yes. Yes. It's like
412. 1:22:55Yes. Quick spin a top. All right, kids. Listen. Have yourself a fantastic weekend homework. Go play in the dirt.
413. 1:23:12Introduce some friction into your AI work. Go slower. Uhoh. All right. Brandon's got a song to play
414. 1:23:27us into the weekend. Here we go. Oh, am I gonna go I'm gonna go play this for you? Okay. Hang on.
415. 1:23:41Oh, this is the the ballot of Dan Cahill. Down where the Delaware bends through the valley. All right, good.
416. 1:23:46This is good. So, if you weren't here earlier, um I was telling a what might be a a fisherman's tale of the strain of
417. 1:23:55rainbow trout that are in the Delaware River came from some western state where the train broke down while the train was
418. 1:24:03next to the Delaware River and they just dumped the fish into the river and apparently it was a guy named Dan Cahill
419. 1:24:09that did it. So, here's the ballot of Dan Cahill.
420. 1:24:27Down where the Delaware bends through the valley. Morning mist rides a cold mountain air. Old men whisper of water and memory.
421. 1:24:41And a fisherman once walking there. They say he came with a wagon of silver. River bride in a wooden pill.
422. 1:24:56And the trout took the current like falling stars in the wake of Dan Kahill's trail. Oh, the river.
423. 1:25:12>> I wonder Dan Cahill. So, there there is a fly called a light Cahill and a dark Cahill. I wonder if Dan Cahill is the
424. 1:25:18inventor of that fly. I bet he is. Um, beautiful song. So, that's the ballad of Dan Cahill if you want to go hear that.
425. 1:25:26Lovely. Thank you, Brandon. Um, and then what was Kyle? When you Oh, Kyle, when you add lyrics, um, how do you do that?
426. 1:25:35So, the way the way you create a video of like of this song with the scrolling lyrics is you just go to um share. No,
427. 1:25:43you go to Oh, I can't do it. That's not my song. Let me go back to my song. So, if you can share your song, you go
428. 1:25:52to download and then you just go video, download video, and it'll automatically make the scrolling lyrics video for you.
429. 1:26:01You have to be a pro member. You have to be a paying member to do that. All right. G. All right, everyone. Have a great
430. 1:26:16weekend. Um, think about friction. When when does it make good sense to introduce friction into your into your practice, into your AI?
431. 1:26:32There might be some times where you want it to be friction-f free. There might be other times where you want it to be really slow.
432. 1:26:38What if you were intentional about that? Like that might be something to play with this weekend and play in the dirt.
433. 1:26:44All right. Have a frictional weekend. Yes, exactly. Good friction. Good friction, everyone. Enjoy the friction.
434. 1:26:54Friction can be fun. Friction can be your friend. That's That's a way to end Friday night date night. Friction can be your friend.
435. 1:27:09Have a good weekend, everybody. Bye.