5/6/2025 - Agentic AI Tools: A Deep Dive and Comparison

Description

Chapters

Transcript

0:01 Come on, you little fat doggy. Hey, you
0:04 little fat doggy. What's going on, you
0:07 little fat doggy? Huh?
0:12 [Music]
0:21 Hey, what song is this?
0:28 [Music]
0:37 [Laughter]
0:40 [Music]
0:52 I met my baby at the F club underneath
0:56 the town cafe on a corner life and four
1:00 street late one
1:02 Wednesday she offered me a red strap I
1:06 said baby not a chance I could go for
1:09 tall cool ginger ale how about we just
1:12 dance we did the boom shaboom the do I
1:15 did it mocha chokata and the
1:16 nitty-gritty we did the coochie cool the
1:19 lazy ballerina baby make me feel like
1:21 I'm a California dreamer
1:26 [Music]
1:32 Well, I asked her for her number. She
1:35 said, "Here's my phone at work." I put
1:37 two and two together. Damn, she must be
1:40 with some jerk. Two summers later on the
1:43 vineyard came another opportunity. She
1:46 came all that way just to see me play,
1:48 but I was with Sweet Lisa Marie. After
1:51 so many months of waiting, well, finally
1:54 come that spring, she was free and clear
1:57 and good to go. And hot damn, we did our
1:59 thing. We did the boom shaboom the
2:01 rumble in the jungle. The who knows who
2:04 in the hast stack shuffle did the I
2:07 don't know the words and the something
2:09 something girl from me.
2:18 Yeah.
2:21 Forgot that song. That's a good old
2:24 Martin 6 to one called Boom
2:29 Shaboom. Good evening, good people. Hope
2:32 you're doing fine on this evening. It's
2:35 late. It's late for a very important
2:38 date.
2:43 [Music]
3:09 Oh, you and I here all along.
3:16 and
3:19 all Sunday morning at
3:24 home. Sky's blue and the coffee is
3:27 strong. It's true.
3:33 Then I open my eyes to a dream realized
3:37 in front of
3:41 me and I haven't got a clue what in the
3:44 world is happening to
3:48 me. Think I think I'm
3:52 happy like first day summer vacation.
3:56 Happy got to get a little rest and
3:58 relaxation.
4:01 Happy like a choir on Sunday morning
4:04 singing true.
4:06 [Music]
4:38 [Music]
4:43 Hello. Hey. Hey. What are you doing? Why
4:47 are you rapping? What are you rapping
4:49 for?
4:50 [Music]
4:52 We're sing We're singing We're singing
4:53 just folk folk music. Folk
4:57 tunes. We're not getting all
4:59 aggressive. We're getting all
5:02 aggressive. Sometimes these dogs, they
5:05 just don't know how to stay in a genre,
5:08 you
5:10 [Music]
5:18 know?
5:22 Feel like I'm watching three different
5:24 views of
5:25 Kyle, but they're all different points
5:28 of
5:28 [Music]
5:30 lag. Oh man. Side hustle meet me in the
5:33 house. What's happening? What's
5:35 happening? What's going down?
5:39 Yo, fantastic cams. Yeah, we do that.
5:43 We'll do
5:44 that. That does look a little slower
5:47 tonight, doesn't it? Something's not
5:50 right. Streamyard. Oh, Tik Tok.
5:57 [Music]
6:07 [Music]
6:20 Cam Katkin, I feel so lost with agents.
6:22 Say more. Like you're trying to you're
6:25 trying to do them and they're just not
6:27 behaving or you're feeling like they're
6:29 just off doing [ __ ] and you don't have
6:32 control. Like what is it about agents
6:34 that you're not grooving on? I mean, if
6:37 you're frustrated with them, I would say
6:39 we don't really have agents right now.
6:41 We've got semi semi-agentic things.
6:44 We've got manis, which is agenty. Or are
6:48 you trying to build them?
6:53 [Music]
7:12 Um, I don't think we've gotten we
7:16 haven't had the chat GPT moment yet for
7:20 agents. I don't know who it's going to
7:22 come from. I mean, Manis is a little
7:26 bit. Um, I would say I would say uh Gen
7:31 Spark is probably
7:33 slicker slicker still than Manis.
7:39 But Manis feels a bit more robust.
7:42 Anyone tried the
7:44 Lortix Suna open-source genurpose agent?
7:48 I have
7:49 [Music]
7:51 not. I have no idea how to use Manis. Is
7:55 that something you could talk about or
7:57 Gen Spark? Yeah.
8:00 Um, yeah. Let's go. We'll we'll go do
8:02 something. I mean,
8:06 [Music]
8:08 to a great
8:11 [Music]
8:15 degree. You can you can almost think of
8:17 it like this. You can almost think of it
8:19 like
8:24 this.
8:27 Okay, we're going to the boxes. We're
8:30 going to the boxes. We're going to the
8:32 boxes, people.
8:41 Just hang on. Hang on. Just calm down.
8:44 Everybody calm down. We're going to get
8:47 there. But we've got the toolbox, too.
8:50 Ah,
8:53 okay. I Ah, okay. I got it.
8:57 Okay. I got it.
9:04 Okay. Okay. It's good. It's good. Good.
9:08 Good question from the crowd. Good. Good
9:11 question. Cam Ken from Cleveland. Okay.
9:16 So, here's what we're going to
9:21 do. Here's your large language model,
9:24 right? That there's your
9:27 GPT. And
9:30 then here's your chat GPT. All right.
9:33 And that's going to be this is going to
9:34 be your basic 40 model. That might look
9:38 like magenta to you, but that's just 40.
9:41 So here's all your knowledge. We took
9:43 all of the knowledge of humanity. We put
9:45 it in this box. Okay? And then we wrote
9:48 a little
9:49 chatbot that you chat into. You got the
9:51 chat hole,
9:53 right? And the basic model 40. This is
9:56 40. All right.
10:00 Perfect.
10:02 Then along comes the reasoning models
10:06 03 04 mini 01 all those
10:11 right where the the toilet paper comes
10:14 in. So it fits perfect. It's a perfect
10:17 metaphor. Okay.
10:22 So, okay. So, before the toilet paper,
10:25 you ask chat JPT for
10:28 something and it goes into the knowledge
10:31 base into the into all of humanity,
10:34 humanity's knowledge and it gets you an
10:36 answer and it gives it to you. And what
10:38 happens with the reasoning models is you
10:41 ask it a question and it goes into all
10:43 the knowledge of humanity and then it
10:45 just keeps asking itself questions like
10:48 like it's just like searching the
10:51 internet going back into the humanity
10:54 just just pulling out lots and lots of
10:57 questions and and then it gives you an
10:59 answer. Okay.
11:02 So, this is
11:07 03. It's solid. I'm liking this. This is
11:10 working. I'm so dead. No, but it gets
11:13 better.
11:15 Okay, this is a box of tools.
11:21 Okay.
11:24 So, so the new the new 03, the the the
11:28 the big MacDaddy 03 that they just
11:30 launched, it can actually write its own
11:34 tools. So, let's say let's say it goes
11:37 into all all of the knowledge of
11:39 humanity and it pulls out some data and
11:41 it it creates a little plan for itself,
11:43 but then it says, "Hey, I actually need
11:45 to analyze some of this data and come up
11:48 with a trend." But, but I can't do that
11:50 by just, you know, talking to it. I
11:52 actually need a tool that's going to do
11:54 trend analysis. It can go into its
11:56 little box of tricks and write Python,
11:58 write a tool that will do that analysis.
12:01 And so, it sort of combines all these
12:03 things together. tools, these things.
12:06 Okay, but 03 is really just using
12:09 Python. It's kind of a single tool.
12:11 Okay, now so that's that's that's gotten
12:15 us to where we are today with chat
12:18 GPT. Manis and GenSpark and and things
12:22 like
12:23 it, they're this, but they have like 80
12:26 different tools, right? I think Gen
12:29 Spark has 30 or 40. Manis has like 80.
12:33 So, Manis can do things like instead of
12:35 going and searching the web, it can
12:37 actually spin up 10 different web
12:38 browsers and have them all doing things
12:41 independently. So, so Manis and GenSpark
12:44 have a really good coordinator role
12:47 that's coordinating all these different
12:49 tools that it's using and all the
12:51 different stuff out there. So, you give
12:53 it a goal, it comes up with a plan, and
12:55 then it figures out, okay, of all the
12:57 tools I have access to, what am I about
12:58 to go turn loose? Um, so hope that
13:02 helps. Um
13:06 um if you've heard of MCP, which is
13:09 model context protocol from from
13:13 Anthropic, what that is is that allows
13:16 developers to take applications like say
13:18 Blender or uh Excel or [ __ ] whatever you
13:24 know some your Macintosh operating
13:27 system and they can create sort of a a
13:31 series of instructions that allow a
13:34 large language model to talk to that
13:37 application. So there's been some some
13:40 examples on
13:41 [Music]
13:43 um on Twitter where you know they just
13:47 tell Blender to like make me a a 3D
13:50 model of a little town with lots of
13:51 buildings in it and you just watch
13:53 Blender like making [ __ ] right? Because
13:57 it's got model context protocol. There's
13:59 going to be thousands of those things.
14:02 And and what Anthropic just announced
14:04 last week is that you can there's now
14:08 like effectively a search engine for all
14:10 the MCP
14:12 servers. So what we're going to have
14:15 like in in very short order. Again, I
14:17 don't think agents are there right now,
14:18 but we'll we'll go play with Manis and
14:20 GenSpark and I'll show you how they
14:22 work. But where we're headed is you're
14:25 going to be able to put a request into
14:27 the chat
14:29 hole and then the thing will start
14:32 reasoning with itself and then it'll
14:34 basically go out and look and say what
14:36 are all the different possible
14:37 applications I could use to help me in
14:40 pursuit of this
14:42 goal. And so it might just go out and do
14:45 a bunch of [ __ ] right? Like maybe you
14:48 want to start an import export business
14:50 and it's finding, you know, all the laws
14:53 in the different countries to get the
14:55 thing from where you want it to where
14:57 you are and it's it's figuring out all
14:59 the tariffs and it's building
15:01 calculators and it's using websites that
15:03 specialize that. Like that's that's the
15:06 kind of thing where it's headed. So, so
15:09 just just
15:10 think your simple request is getting
15:14 increasingly more complicated on for
15:16 what the systems do with it. Now, what
15:19 do you do with that power? I don't know.
15:22 Um, last night I shared an article that
15:25 I wrote. If you go to my ex channel um
15:28 Kyle Shannon and click on articles,
15:32 there's a there's an article in there
15:34 about how to use 03 as a non- STEM
15:38 person if if if you don't have physics
15:41 and math problems to solve how do how do
15:44 you use a reasoning model? So I wrote an
15:46 article on that. So that that might be
15:48 something worth looking at.
15:50 Um we'll come back to this thing. Let's
15:53 Let's just go
16:01 to Let's go to Let's go to Gen Spark
16:05 because I think Gen Gen Spark is like a
16:08 friendlier, simpler
16:11 Manis. All right. So,
16:15 GenSpark.ai. And I if you're if you're
16:18 trying to get to Manis, it's
16:21 mus dot Wait for it. I
16:26 am
16:29 manis.mim and genenspark.ai.
16:32 Okay, so gen Gen Spark has this thing
16:36 called the super
16:37 agent. Um, it can also generate video.
16:40 It can generate images. Um, it can do
16:43 slides. Um, why don't we do slides?
16:47 Because slides are kind of interesting.
16:48 Why don't we do
16:52 um find
16:55 all of the tools
17:02 that are
17:06 like
17:09 manisim and
17:12 genspark and then I'll I'll let it know
17:14 it's it's
17:16 it's that's it.
17:18 Jen Spark, that's
17:23 you.
17:30 Oops. Find all the tools that are like
17:33 Manis and Gen Spark
17:36 that are
17:39 agentic and
17:43 powerful and can use tools
17:49 and create a
17:53 presentation that
17:56 describes what they
18:01 are, the pros and cons of the
18:08 main tools.
18:15 and at
18:17 least three slides of use
18:24 cases
18:27 beyond
18:29 research and
18:32 math. All
18:34 right. So you can give these things
18:37 relatively complicated
18:39 uh instructions and then they just go
18:41 and douche [ __ ] So So we're in slides
18:47 um and I'll just hit return. So we've
18:49 got our
18:51 little area over here where this is
18:54 where our chat is and then over here on
18:56 the right we'll get the So so it says
18:58 using tool search
19:02 manisim Agentic AI
19:05 tools using tools search using tool read
19:09 so it's got a reader tool and and you
19:12 can view any one of
19:14 these right now it's searching again
19:16 claude agentic AI tools
19:19 And then here's a bunch of [ __ ] it's
19:20 finding,
19:22 right? Pretty
19:25 slick. Search
19:31 thinking. Okay, now I have enough
19:33 information to create a comprehensive
19:35 presentation using tool. I don't know
19:38 what tool it is. Oh,
19:40 presentation. New eight
19:42 slides. So, it's got a presentation
19:45 tool, right?
19:48 And then if you notice over here
19:51 thinking I'm creating a title
19:53 slide I'll use tail tailwind CSS for
19:57 styling now it's writing the
20:01 code right so
20:05 off off go do
20:09 research read some
20:11 [ __ ] comprehend some
20:14 [ __ ] figure out of all the [ __ ] you just
20:17 learned, what eight slides you need to
20:19 write. Go think about slide one now.
20:23 Code slide one. And then you'll see here
20:25 in a second, we'll see slide one. And
20:27 then it's going to do this for all all
20:29 eight slides.
20:31 So, and like, you know, that's pretty
20:34 slick. Leading agentic tools, Manis,
20:37 Gen, Spark, Claude, and Perplexity. All
20:39 right. I, you know, I I wish it would
20:42 have given me one I hadn't heard of, but
20:44 whatever. I've heard of all those and I
20:47 I would say, you know,
20:48 perplexity. Yeah, Perplexity is in the
20:51 neighborhood of one of these. Clawed
20:53 with MCP certainly is, but like it gave
20:56 us a nice little design there, right? It
21:00 knew which word to aentic
21:03 AI to
21:07 highlight. Perplexity has agentic tools.
21:10 Well, kind of. If you've used um in
21:16 perplexity, have have you generated a
21:19 page in perplexity? While this is doing
21:20 this, well, let me jump over to
21:22 perplexity and show you something that's
21:24 in the neighborhood of
21:26 agentic. And
21:29 again, everyone right now, everyone,
21:33 [ __ ] everyone is calling everything
21:36 an
21:37 agent, right? Because Sam Alman a year
21:40 ago said, you know, agents are coming
21:43 when whatever the [ __ ] it was. So now
21:45 everyone's calling every automation that
21:47 they create an
21:49 agent. You know, think think of an agent
21:52 is an agent is doing something on your
21:54 behalf without you needing to guide
21:57 it. Um so what perplexity does that's in
22:02 the neighborhood, it's not great, but
22:03 it's in the neighborhood is
22:08 Where is it? Uh, oh [ __ ] they've moved
22:13 things. Um, okay. Let me just go I'll go
22:16 I'll go research something. Uh, find me
22:21 the
22:22 top agentic
22:25 tools as of today.
22:33 [Music]
22:37 So, it's searching. So, like one of the
22:39 things that Perplexity does is it
22:41 searches a bunch of things. It gives you
22:43 the
22:44 sources. Um, so this is kind of doing
22:48 what that does, but you know, it's it's
22:50 not
22:52 really. So, what did I come up with?
22:55 Relevance AI co-pilot studio Watson X
22:59 orchestrator. Um, let's see. These all
23:02 suck. So, let's go. No things that are
23:07 more like
23:10 um
23:11 Claude MCP
23:17 um
23:23 Manis.Mim and
23:26 Genspark dot what was it? AI
23:35 Add to space. They've gotten rid of
23:37 something
23:41 here. Discover home
23:44 plus
23:47 spaces. Did they get rid of pages on
23:54 perplexity? Export answer. Share.
23:59 Rewrite Pro Search
24:06 Copy. Am I losing my mind? This is a
24:10 master
24:12 class with a laugh. Thanks,
24:17 Avi. Oh, my boxes. That was a master
24:20 class. Um, I thought Mr. Smith was an
24:23 agent. Um, which model did you use to
24:27 help script this new analogy? Hey, that
24:29 was just that was just off. I'm like
24:31 Harry Mack with the uh with the
24:37 analogies.
24:39 Um, where the [ __ ] are
24:42 pages? Oh,
24:46 library threads
24:49 page. Jesus, they buried the [ __ ] out of
24:52 it.
25:01 Okay, up at the
25:06 top in Perplexity. Let me make this a
25:09 little
25:10 bigger. In Perplexity, up at the top,
25:12 there's this thing called
25:14 library. And if you click on library,
25:17 it'll take you to
25:19 threads, which I guess are searches I've
25:21 done.
25:23 And then there's a plus
25:25 button which if you click it, you can
25:28 create a thread or a
25:30 page. I don't know why they buried this
25:32 so much. Maybe it's costing them
25:37 money. So you choose the audience. Do
25:40 you want beginners or experts? Let's go
25:42 with beginners. And we'll say create a
25:45 page
25:47 about Why can I not type in here?
25:51 Oh, what's your page about?
25:53 Um,
25:59 Agentic Tools. I I'm so pissed at these
26:02 companies. They [ __ ] change the
26:04 interface every [ __ ] week. Like, how
26:08 are we supposed to use your dumb [ __ ]
26:11 tool? Hey, Gentech tools like
26:16 Manis.im I am and Jen Spark
26:21 um Jen Spark
26:24 um find me others and
26:28 explain what the hell I am supposed to
26:34 do with
26:36 them. Okay. Boom.
26:42 So, where this thing gets kind of
26:45 agentic is it's off doing searches right
26:48 now. And what it's got is it's got kind
26:51 of a layout tool built into it. So
26:54 rather than just doing a search with
26:57 sources, what it's actually doing now is
26:59 finding images, it's doing sort of
27:02 magazine layout. So what it's creating
27:04 is like a little mini magazine
27:06 article about AI agents uses and
27:10 alternatives
27:12 right and if you scroll down so similar
27:16 AI's similar AI's dynamic learning
27:20 capabilities Rasa for contextual
27:22 conversation design think of
27:27 Rasa wait where did Rasa go as a Lego
27:31 set for building smart chat bots unlike
27:34 simple chat bots that follows. Okay, so
27:36 it's finding us all this [ __ ] Cool. Are
27:39 we done
27:40 here? There's
27:45 genark my
27:49 experience. So I think with any of
27:53 these copy it, link it.
27:57 They've completely changed
28:05 this. Continue editing.
28:10 Okay. Ah. Oh, you Jesus Christ. Now you
28:14 have to say continue editing. Used to
28:16 just be able to edit this stuff. Okay.
28:18 So So see it picked this image for Gen
28:21 Spark and it's the the media here is not
28:25 Gen Spark.
28:27 But let's so we can change out the
28:29 image. Super agent
28:32 showdown. Best free
28:35 alternatives. Now there's that's a cool
28:37 thing. Gen Spark. And then here I can
28:40 add media for similar. So if I say add
28:42 media, that'll go find
28:44 media. And so it threw in that chart. I
28:47 don't know what that chart is, but
28:49 whatever. We'll throw in that chart.
28:55 And then let's see. This has got um
28:58 bullet points, but let's say we want
29:00 that to be a table. So I think if I go
29:03 edit here. Yeah, I can just turn that
29:05 into a
29:07 table and then hit that. And now it's
29:10 going to rewrite whatever I just did as
29:13 a table. Yeah, here we go. Boom. Now you
29:15 got a table there.
29:21 So, how this is not agentic is like I'm
29:25 having to do a lot of this, right? Um,
29:30 but the fact that it went out and it
29:32 found three or four things and it wrote
29:34 articles and it knew how to do it and it
29:35 did did all that without me, that makes
29:37 it it's got elements of of Agentic,
29:41 right? Think of Agentic as it's just off
29:43 doing its thing. Haven't made a page in
29:45 a while, but I still get them dropped to
29:47 me. Yeah, never had a subscription, so I
29:50 never used pages. I don't think you need
29:52 a subscription for it. I don't think I
29:54 have a subscription anymore. I had the
29:56 pro account for a while. All right,
29:58 let's go look at what GenSpark did. Oh,
29:59 it's still [ __ ] working on our
30:04 presentation. Let me see. View and
30:10 export. Play slides.
30:17 Nice. All
30:19 right. What are agentic AI tools? Can
30:22 you see this? Oh, no, you can't because
30:24 I went full screen and chat GPT decided
30:28 to or no, Chrome. Oh, no. OSX decided to
30:33 reorient the windows in some random
30:35 [ __ ] way.
30:37 Um, traditional AI versus agentic AI.
30:41 Traditional AI assistants respond
30:43 directly to prompts, are limited to
30:46 pre-trained knowledge, require explicit
30:48 instructions, have no ability to use
30:50 external tools. Agentic AI tools,
30:53 autonomous decisionmaking, that's
30:56 probably the biggest one. The the toilet
30:58 paper role, that's it. Sort of planning
31:01 and thinking about and and changing its
31:03 plan based on what it learns. External
31:05 tool usage, multi-step planning,
31:08 self-improvement over time.
31:12 Yep. Good. I like
31:17 it. Leading tools. Manis. Okay. So, here
31:21 we go. Autonomous execution, multimodal
31:23 tool integration, adaptive learning.
31:26 Manis, state-of-the-art performance in
31:28 Gaia benchmark, greater than 60%.
31:31 Genspark, agentic agent with dedicated
31:34 AI agents for search, deep research, and
31:37 various automated tasks.
31:39 Multi- aent teams nice deep research
31:43 content generation 80 plus tool
31:45 integrations in
31:47 GenSpark Claude AI assistant from
31:50 Anthropic with agentic capabilities for
31:52 planning. I I I if they're not including
31:57 MCP stuff here I that's that's a stretch
32:01 and perplexity is a stretch frankly both
32:02 of their their other two they're
32:04 stretches for
32:06 me. Here's pros and cons for
32:09 both use cases. Business process
32:14 automation content creation and media
32:19 production personal digital
32:22 assistant. So there you have it. That's
32:25 cool. All right.
32:28 I
32:32 I How do I get out of here? Did it Okay.
32:37 Is it done? Yeah, it's done. So anyway,
32:42 oh god damn that thing.
32:45 Um, so yeah, here's all the slides it
32:48 made and it did all that, you know,
32:52 these are not it's not using what's
32:54 amazing about this. It's not using a
32:57 slide program, right? Like there's slide
33:01 programs out there like I forget what
33:02 they're called. Whatever doesn't matter.
33:06 This is actually writing the code and
33:08 the CSS to do the design of these
33:11 slides. Can you say
33:14 move? Can you say more about personal
33:16 digital
33:17 assistant?
33:21 Sure. Yes, I can read the slide. Hang
33:27 on. Um, view and export.
33:32 Your Tik Tok cam is off left. How's
33:37 that?
33:39 Better. Hang on. Let me let me go. Hang
33:42 on, Cam. Yes, I will read this slide for
33:46 you. I know. I didn't really read it.
33:48 Okay. Personal digital assistant. How
33:50 agentic tools can enhance daily
33:51 productivity by autonomously? Well, you
33:54 know what I'll do? I'll I'll also export
33:56 Oh my god. I'll also export this um and
34:01 and put it in the uh in the salon, but
34:03 let me let me just read through it here
34:05 and see if it's worth exporting. Um,
34:08 Agentic tools enhance daily productivity
34:10 by autonomously managing personal tasks,
34:13 calendar, and scheduling. Autonomous
34:15 meeting scheduling across time zones.
34:18 It's kind of cool. Dynamic priority
34:21 based
34:22 rescheduling. Intelligent buffer time
34:25 management. I don't know what that is.
34:27 Oh, this is cool. It's also saying which
34:29 tools would do this. Well, Manis and Gen
34:32 Spark would do that one. Research and
34:34 learning, Perplexity and
34:35 Claude. Travel planning, Manis and Gen
34:38 Spark. Financial management, Manis and
34:44 Perplexity. Time saved weekly, 15 plus
34:47 hours. Decision quality plus 42%. Stress
34:51 reduction minus 38%.
34:54 Not if you got to go learn all these
34:56 [ __ ]
34:57 tools. Gamma was the other one. That's
35:02 right.
35:04 Um, but let me let me
35:07 um let me export
35:10 this. We'll export it as a
35:15 [Music]
35:18 PDF. Uhoh. What did I do?
35:34 And a gentic tools comparison 2025
35:38 desktop. All right. So I am going to the
35:41 AI salon. If you don't know what the AI
35:42 salon is, it is swell. We had a lovely
35:45 meeting tonight. We had uh Telicia White
35:47 from lovable.dev dev uh which is one of
35:51 them their vibe coding tools. Uh she
35:54 came in and demoed Lovable for us and
35:56 was just awesome. It was a really good
35:58 meeting tonight. I can't get one to work
36:01 on clearing out my Gmail inbox. Yeah.
36:04 Yeah. Like I said, listen all
36:08 everybody
36:11 don't think you're missing the agent
36:13 bus. You haven't missed the bus. The bus
36:16 hasn't been assembled yet.
36:20 Right now, we've got like an agentic
36:25 unicycle and and it's only got one
36:28 pedal, so it's really [ __ ] hard to
36:33 ride. Um, where do you want me to put
36:36 this? You want me to put this in the
36:37 regular
36:41 channel? Can I upload a PDF to this
36:44 thing? I bet I can't. That's gonna piss
36:47 me off.
36:50 [Music]
36:55 Uh the pro there was a problem. There's
36:59 always a [ __ ]
37:05 problem. Shut up. Just all of you leave
37:08 me alone. Leave me
37:10 alone.
37:13 Export. Export. Quartz filter. Reduce
37:16 file
37:18 size.
37:20 Save.
37:23 Replace. Let's make things smaller. This
37:26 shouldn't be that big because it's all
37:27 [ __ ] code. But let's let's go back
37:31 here. Let's try this
37:33 again.
37:42 Nope. Anybody? Ber, anybody know how I
37:46 can upload a PDF to uh Mighty Networks?
37:52 No, I'm exhausted. I'm exhausted,
37:57 people. Tired
38:02 Tuesday. I just want a real assistant to
38:05 handle all of my
38:07 111,239 unreads. Yeah, exactly. Save
38:11 them as JPEGs. Yeah, I know. I just
38:13 don't want to upload [ __ ] eight
38:15 different JPEGs cuz I'm
38:18 lazy.
38:25 Um, nice artwork, dude. What the [ __ ]
38:28 Are you
38:36 okay? That's Amelio's wife's handiwork
38:39 right there.
38:42 Oh my god. Save it in a Google Drive and
38:45 paste the link. Okay. Thank you, Vicki.
38:48 Once again, serve serving as my as my
38:53 brain. Okay. So, here's what I'll do.
38:56 I'll go to I'll go to here. I'll go to
39:00 here. We'll take a screenshot of that
39:03 bad boy.
39:09 [Music]
39:11 Uh, we'll save that to the
39:15 desktop.
39:17 Boom. Then we'll go to drive and I'll
39:21 put it in just my drive loose because I
39:26 know how to live. I know how to do file
39:28 management. You know how you do file
39:30 management? You save
39:33 everything in the root
39:36 folder and make sure that it's not
39:40 dated or has any kind of name that would
39:43 make
39:44 sense. Then you got to find it with your
39:48 shitty naming
39:51 convention. By the way, I do consulting.
39:54 You can hire me for file management
39:57 strategies. I'm reasonable.
40:00 Um,
40:03 share share with anyone with a link. All
40:06 of you can have
40:08 this. Copy link. And then we're going to
40:11 go back
40:12 to uh, what's it called? AI
40:15 salon. We're going to go get our picture
40:17 that we took a picture
40:19 of. Come on. Where is it? Where's my
40:22 screenshot? Uh, I have a second
40:24 desktop. Damn it. Screenshot.
40:28 I listen all of you people that know how
40:30 to use computers. I don't want to hear
40:32 it. I don't
40:34 care. Is that it? That's
40:37 it.
40:38 Uhuh. Hey, Marge. Hey, Marge. Marge.
40:42 Yeah. Hey, listen. I'm Oh, I know you're
40:46 watching the wheel,
40:48 hun. Yeah, I
40:51 just figured out myself. Marge is not as
40:54 patient as she used to be. I think it's
40:57 her
41:07 bunions. All right. So, this is uh what
41:11 is this? This is the here is
41:15 the genspark.ai
41:18 AI um
41:20 aentic tool compare is son
41:27 presentation I
41:31 made. People are going to read this and
41:33 be like what the hell is he talking
41:35 about? But there you go. We're not going
41:38 to notify all of the world that we put
41:40 up a PDF. All
41:42 right. Got it.
41:46 You're a complicated man,
41:51 Kyle. By the way, lovable oneshot my app
41:54 using 03 instructions based on my
41:56 half-ass prompt. So, archetype, that's
42:00 that's where we're headed. Like, this is
42:02 the thing. Okay, so I'm I'm going to
42:04 we're going to we're going to jump now.
42:05 I So, Cam, I hope that I hope the
42:08 masterful
42:10 um explanation of
42:16 large language models and tools and
42:19 toilet paper. I hope that cleared things
42:22 up.
42:26 Um, but I think what what archetype just
42:29 said right there is is absolutely right.
42:33 And uh we so we we read this on the
42:37 salon tonight. So I know some of you
42:38 have probably heard this, but I think
42:40 this is really important and I'm I'm
42:42 probably going to talk about this for a
42:44 while.
42:45 Um so this is from Mika
42:48 Kaufman. He's the CEO of Fiverr. And if
42:52 you don't know what Fiverr is, it's like
42:53 a freelance marketplace, right?
42:57 Um, I'm pretty sure it was
43:01 Fiverr. Yeah, I'm pretty sure it was
43:04 Fiverr that just added um AI tools for
43:09 freelancers to be able to do [ __ ]
43:10 faster, right? So, they could actually
43:12 charge five bucks and, you know, make
43:14 money. No one charges five bucks on
43:16 Fiverr anymore, but whatever. It's like
43:19 Dollar General stores. Nothing's a
43:20 dollar in a Dollar General store
43:21 anymore. So,
43:23 um, Manis did it, too, but it's uglier.
43:26 Yeah. No, Lovable is really good
43:28 design-wise. So, anyway, so the CEO of
43:31 Fiverr sent this email to his staff and
43:34 then he posted it to X saying, "Hey,
43:36 this is probably going to come out
43:37 anyway, so you might as well hear it
43:39 from me. I've always believed in radical
43:42 cander and despise those who sugarcoat
43:44 reality to avoid stating the unpleasant
43:47 truth. The very basis for radical cander
43:50 is care. You care enough about your
43:52 friends and colleagues to tell them the
43:54 truth because you want them to be able
43:56 to understand it, grow, and
43:59 succeed. So, here's the unpleasant
44:01 truth. AI is coming for your jobs. Heck,
44:05 it's coming for my job, too. This is a
44:08 wakeup call. It does not matter if
44:11 you're a programmer, designer, project
44:14 product manager, data scientist, lawyer,
44:17 customer support rep, salesperson, or
44:20 finance person. AI is coming for you.
44:23 You must understand that was once
44:25 considered easy tasks will no longer
44:28 exist. What was once considered hard
44:30 tasks will be the new easy, and what was
44:34 once considered impossible tasks will be
44:36 the new hard. If you do not become an
44:40 exceptional talent at what you do, a
44:44 master, you will need to face Wait, you
44:48 you will face the need for a career
44:50 change in a matter of
44:53 months, right? This is not a five-year
44:56 thing. This is like before the end of
44:59 the
45:00 year, you got to get your [ __ ] together
45:02 and [ __ ] level
45:05 up, right? He's saying this to his
45:08 staff.
45:10 Um, if you do not become an ex
45:12 exceptional talent at what you do, a
45:14 master, you will face the need for for
45:16 need for a career change within a matter
45:18 of months. I'm not trying to scare you.
45:20 I'm not talking about your job at
45:22 Fiverr. I'm talking about your ability
45:25 to stay in your profession in the
45:27 industry. He's like, I don't give a [ __ ]
45:29 if you work here or there. If you don't
45:32 do this, you're not going to make it
45:34 anywhere.
45:36 Um, are we all doomed? Not all of us,
45:40 but those who will not wake up and
45:43 understand the new reality fast are
45:46 unfortunately doomed. What can we do?
45:50 First of all, take a moment and let it
45:52 sink in. Drink a glass of water. Scream
45:54 hard in front of the mirror if it helps
45:56 you. Now,
45:58 relax. This is it's a well-written
46:00 letter. Panic hasn't solved problems for
46:03 anyone. Let's talk about what would help
46:04 you become an exceptional talent in your
46:06 field. The fact that you're here
46:09 tonight, it is [ __ ] 9:40 in Denver.
46:12 So, if you're on the East Coast, it's
46:13 [ __ ] 11:40, right? Even on the West
46:16 Coast, it's almost 9:00 at night. There
46:19 are 52 people in the Tik Tok, 48 people
46:23 on YouTube and X and
46:28 LinkedIn. I'm hopeful for you.
46:32 you're here. I'm here. Like, why am I
46:34 here? Cuz this has been coming. But, you
46:39 know, we now got someone articulating it
46:41 quite well. Um, okay. So, he says,
46:44 "Let's talk about what would help you
46:46 become an exceptional talent in your
46:48 field. One, study, research, and master
46:52 the latest AI solutions in your field."
46:55 Right? Listen to people that use boxes
46:58 and toilet paper to explain how AI
47:03 works. Okay, this one is definitely this
47:06 channel. Try multiple solutions and
47:08 figure out what gives you
47:10 superpowers, right? What am I doing in
47:13 here night after night? Just [ __ ]
47:14 around with stuff, trying different
47:16 things. You know, we did it tonight. We
47:18 played with uh Gen Spark and then we
47:20 went to Perplexity and Perplexity's been
47:22 kind of downgraded and is weird now.
47:25 Jen Spark kicked ass. Okay, so we know
47:27 that. Um, archetypal architect said that
47:31 Lovable did better than Manis did
47:32 because it was prettier. Okay, good. So,
47:35 he's understanding, you know, where the
47:37 superpowers are. By superpowers, I mean
47:40 the ability to generate more outcomes
47:43 per unit of time with better quality per
47:47 delivery. Listen to
47:50 this. Superpowers. the ability to
47:54 generate more outcomes per unit of
47:57 time with better
48:01 quality. So if you're used to charging a
48:04 week's worth of your time to
48:07 deliver
48:10 research, you now have to do that in a
48:13 day or two days and it has to be better
48:18 than what you did when you took a week.
48:20 Right? That's super power there. There
48:23 used to be this old adage in
48:24 advertising, good, fast, and cheap. You
48:26 can have any two of the
48:28 three, right? You can have it good and
48:30 fast, but it ain't going to be cheap.
48:32 You can have it fast and cheap, but it
48:34 ain't going to be good, right?
48:37 You not
48:38 anymore. Not with
48:40 AI. Good, fast, and cheap is now the
48:44 expectation or will
48:46 be.
48:49 Um, programmers code customer support
48:53 tickets. You know, he he's given
48:54 examples here. Then he says, and and
48:57 again, that's what this this room is for
48:59 is for number two. Find the most
49:01 knowledgeable people on our team who can
49:04 help you become more familiar with the
49:06 latest and greatest in AI. This isn't
49:08 just about me teaching what I know. This
49:11 is about all of you in here as you're
49:13 commenting to one another, as I'm
49:15 talking about stuff and you're going,
49:16 "Oh, I did that over here. I did that
49:18 over here."
49:20 The power of that knowledge sharing is
49:23 [ __ ] insane. Basically, what he's
49:25 saying
49:26 is learn everything you can on your own
49:29 and then get your ass in community.
49:31 Connect with other people that are
49:32 curious about this [ __ ] and hungry about
49:34 this [ __ ] Time is the most valuable
49:38 asset we have. If you're working like
49:40 it's 2024, you're doing it wrong.
49:44 you are expected and needed to do more
49:47 faster and more efficiently now. Um,
49:51 become a prompt engineer. Google is
49:53 dead. LLM and Gen AI are the new
49:55 basics, right? It's not a big deal that
49:58 you can use chat GPT. That's now
50:00 expected. Um, and if you're not using
50:03 them as experts, your value will
50:05 decrease before you know what hit you.
50:07 Five, get involved in making the
50:09 organization more efficient using AI
50:11 tools and technologies. It does not make
50:13 sense to hire more people before we
50:16 learn how to do more with what we have.
50:18 It does not make sense to hire more
50:21 people until we understand how to do
50:24 more with what we
50:26 have. He didn't quite say there's a
50:28 hiring freeze, but it doesn't sound like
50:29 he's going to be
50:30 hiring. Understand the company strategy
50:33 well and contribute to helping it
50:35 achieve goals. Don't wait to be invited
50:37 to a meeting where we ask you to
50:39 participate uh each participant for
50:42 ideas. There will be no such meeting.
50:44 Instead, pitch your ideas
50:47 proactively. Stop waiting for the world
50:50 or the place of work to hand you
50:51 opportunities and to learn and grow.
50:54 Create those opportunities yourself. I
50:56 vow to help anyone who wants to help
50:58 themselves. Right? So, it's a bit it's a
51:01 bit like, you know, go suck it up,
51:03 buttercup. But suck it up, buttercup,
51:06 right? That's where we are. If you don't
51:07 like what I wrote, if you think I'm full
51:10 of [ __ ] or just an [ __ ] who's trying
51:12 to scare you, be my guest and disregard
51:14 this message, I love all of you and wish
51:16 you nothing but good things, but I
51:18 honestly don't think that a promising
51:21 professional future awaits you if you
51:24 disregard reality.
51:26 So, um, and then he wraps it
51:29 up. Brandon made a good point there.
51:32 There were two two other CEO letters
51:34 that preceded this one. I have a feeling
51:37 we're about to see a rash of these. This
51:40 was the one that was the most directly
51:42 worded where it was just like [ __ ]
51:45 you're going to have to [ __ ] deal.
51:47 So, anyway, Google is not dead while
51:50 they are rising from the dead. Well, I
51:52 would agree with that. I I I think I
51:54 think Google's AI is good,
51:58 but I saw a thing today. It was some SEO
52:02 dude and he said he said SEO is dying,
52:06 right? He said they they were having
52:09 traffic go from like 10,000 somethings a
52:11 day to 3,000 like like you know twothird
52:15 drop off. Um so becau
52:19 because people are not using Google as
52:22 much. They're using chat GPT. So that
52:24 search traffic is is diminishing
52:26 now. Might there be
52:30 LLM
52:31 optimization?
52:33 Yeah. So, so yeah, there there How do
52:36 you get How do you show up in a chat GPT
52:38 search when someone's asking about Ecoin
52:41 toothbrush or toothpaste? I don't
52:45 know. Someone's gonna have to figure
52:47 that [ __ ] out, too.
52:49 So any anyone that says Google is dead
52:52 is being a bit hyperbolic, but SEO's
52:55 toast. I could see Google's agent space
52:58 making some waves. They showed they're
53:00 very serious with 2.5. Yeah, it looks
53:02 like the the a new version of 2.5 just
53:05 came out that's supposed to be really
53:06 good at
53:07 coding. Google's problem is not their
53:10 technology. Google's problem is they
53:13 fired their [ __ ] marketing layer.
53:16 They fired all the managers. They fired
53:19 all the all the non-engineers. They're
53:21 like, "This is just about
53:24 technology." And so now they've got a a
53:29 horrific product strategy. Well, they
53:31 don't have one. They They're literally
53:33 just launching features into one of
53:36 about a dozen
53:37 URLs. No one can find them. Like, the
53:40 only ones worse than Google right now
53:41 are Microsoft. And the only ones worse
53:44 than Microsoft are Apple because they
53:46 haven't done [ __ ] anything.
53:53 Oh my god, it's crazy. It's
53:57 crazy.
54:00 Um, absolutely
54:02 bonkers. Agent space is going to be huge
54:05 and agent to agent is going to be huge.
54:07 Yeah, don't don't worry that you're not
54:10 caught up on agents. Probably I'm going
54:13 to guess
54:17 sometime
54:18 after GPT5. So, I'm going to guess GPT5
54:22 comes out in June or maybe
54:25 July. And then I'm thinking probably
54:27 August, September, October is when we
54:31 start
54:32 seeing
54:34 real autonomous agent systems show
54:38 up, right? Because MCP came out what,
54:42 three months ago, two months ago.
54:46 So developers are developing all their
54:48 servers and the infrastructure is now
54:50 catching up and then you know OpenAI is
54:52 going to have their version of it.
54:53 They're probably going to support MCP.
54:56 They're going to have their own agentic
54:57 framework. Like all of this stuff is
55:02 coming. Here's what to pay attention to
55:05 because this doesn't really exist right
55:07 now.
55:10 start paying attention for companies
55:12 like OpenAI or Anthropic or or any of
55:16 these talking about the um the
55:20 coordination
55:21 layer, right? Where they're going to
55:23 have specialized LLMs that their only
55:26 job is to take your they're going to
55:29 they're going to act like project
55:30 managers, right? There's going to be a
55:32 project manager AI that you talk to and
55:35 it's going to go manage all these
55:38 different activities.
55:40 Um, I think we'll see some of the
55:42 frontier companies starting to talk
55:43 about that layer and GPT5 could could
55:47 have that conversation in it, but that
55:48 that's what to pay attention to. When
55:50 that starts showing up, then then agents
55:52 are probably on their way. News brief.
55:54 Google miner finds an app nobody knew
55:57 existed. Exactly. Story at
56:03 10. Yeah, exactly.
56:06 Why is Apple so unambitious with AI? I
56:10 don't [ __ ] know. Well, I I do know. I
56:13 do
56:14 know. Steve
56:18 Jobs, you know, on his way out, he had
56:21 he had two
56:23 options
56:26 and he chose Tim Cook. He chose a CFO.
56:32 The reason you choose a CFO is because
56:35 you don't want your company to go out of
56:39 business. The reason you would choose
56:41 another innovator is you want the
56:43 company to keep
56:44 innovating. He chose the C CFO, right?
56:48 He chose Tim
56:49 Cook. So naturally, as the innovation
56:56 pipeline that was in Steve Jobs head
56:58 dries
57:00 up, if the priority of the company is
57:04 sell more [ __ ] and keep the
57:07 margins, you know, there's not a
57:10 commitment to innovation. I don't
57:11 [ __ ] know. I don't get it. I don't
57:13 get it. It's It's like one of the
57:16 saddest things in in technology history.
57:19 out. It's possible that Apple, like
57:22 Apple's never been a leader. They they
57:24 always sort of sit back, watch everyone
57:26 sort of [ __ ] around, and then they're
57:28 like, "Oh, you want to see how to really
57:30 do it? Here you go." They've had that
57:33 opportunity now for a year and a half,
57:37 and it's just their their AI [ __ ] sucks.
57:40 It's it's like unusable. It's it's
57:42 depressing. It's depressingly bad. So, I
57:46 don't know. I don't know what's going on
57:47 over there.
57:50 Does anyone else fall for the April
57:54 Fool's saying Apple acquired sesame? No,
57:58 I didn't see
57:59 that. Apple's historically been the
58:02 perfector. Yeah, but but like
58:05 normally normally you've got some
58:08 sense that they're perfecting something,
58:12 right? That they've got some capability.
58:15 They're like, "Yeah, we're not going to
58:16 talk about that." But you know,
58:18 something's cooking now. I just get the
58:20 sense they're just in I don't even think
58:22 they're in [ __ ] panic mode. It's
58:25 bizarre. It's bizarre. I think they
58:28 fired someone from something AI and they
58:30 they they collapsed the the goo the
58:33 goggle team, the the [ __ ] VR goggle
58:38 team, and put them on I I don't know
58:40 what's going on. I don't know. I don't
58:42 know. I don't know. can't talk about it.
58:44 Makes me sad. Like I'm a I'm a diehard
58:48 Apple fanboy and I just, you know, about
58:52 four or five years ago, their product
58:54 quality just [ __ ] went off a cliff.
58:57 Maybe even longer than that. Seven,
58:58 eight years
58:59 ago, and now they're like, they're
59:01 missing the AI boat. They've got more
59:05 [ __ ]
59:06 cash. They could be buying up all these
59:09 companies that are doing good work.
59:11 They're not [ __ ] doing a thing.
59:12 They're not doing a thing. I feel like
59:15 Apple's given up. I I feel like that's
59:17 that's the thing that makes it
59:18 depressing. It's like where's the
59:20 [ __ ] panic? Where's the
59:23 urgency? They're like, "Oh, you can make
59:26 emojis of yourself now in in at the
59:29 Eiffel
59:30 Tower.
59:32 Really? That's what we're going
59:36 with? [ __ ] fancy mimojis.
59:40 [Laughter]
59:50 Maybe Ethel knows that the aliens are
59:52 going to land soon, so there's no point.
59:54 Yeah, I like that would be that. You
59:56 know what, Shane? At least that would be
59:58 a [ __ ]
1:00:00 explanation. I I I honest to God, I
1:00:02 don't get it. I don't get
1:00:05 it. They should bin Vision Pro and
1:00:07 compete with Neurolink. They should do
1:00:09 [ __ ] something. do their car, compete
1:00:12 with Tesla, do
1:00:15 something. I I I
1:00:18 just it just makes me sad. And it's
1:00:21 Listen, I I have experienced like a
1:00:24 teeny
1:00:25 tiny micro version of this. When I when
1:00:28 I started agency.com in the 90s, you
1:00:32 know, we started as a small scrappy
1:00:34 digital agency and then we grew really
1:00:36 fast and we grew through acquisition.
1:00:40 And I ended up leaving the company
1:00:42 because it had
1:00:44 become the company we used to make fun
1:00:47 of, right? The company I founded ended
1:00:51 up being one of those companies that
1:00:53 couldn't make decisions. It was always
1:00:55 political. People were [ __ ] doing
1:01:01 shenanigans. Who does who says that? M.
1:01:05 Baabab says, "Hi, Dad."
1:01:08 Hello. I don't know what that
1:01:12 means. Um, Vision Pro had such promise
1:01:15 and it just disappeared. I listen, I I
1:01:18 don't think they're going to kill Vision
1:01:19 Pro. I I I think there's got to be a
1:01:22 version two and then probably a version
1:01:24 three. Version three will be in a form
1:01:26 factor that most people can use and can
1:01:29 afford. I mean, this is how they've
1:01:32 always done it. So, I I don't think
1:01:33 Vision Pro goes away. Um,
1:01:37 but it wasn't the runaway hit they
1:01:40 thought it would be. Like, it's got some
1:01:41 seriously impressive technology in it.
1:01:44 But what the [ __ ] with the AI?
1:01:47 Anyway, all right, people. Um, my voice
1:01:51 is getting crispy. It's been a long one.
1:01:53 I'm curious to see what they say about
1:01:55 AI at Worldwide Developer Conference.
1:01:57 I'm assuming they realize how far behind
1:01:59 they
1:02:00 are. Why are you lying? L I G H I N
1:02:05 G. I don't know what that
1:02:07 means. What am I lying
1:02:11 about? Lighting. What am I
1:02:14 lighting? Um, Apple deals with teachers.
1:02:17 Teachers hate AI. Maybe that's why they
1:02:19 don't push
1:02:20 it. Teachers hating
1:02:23 AI. Okay, let's let's run this logic
1:02:27 train, shall we?
1:02:37 teachers. Their job is to impart
1:02:40 knowledge and teach kids how to
1:02:44 learn. And along
1:02:47 comes the most
1:02:51 profound knowledge access tool in the
1:02:55 history of humanity.
1:03:00 And schools are are saying kids are
1:03:04 going to use it to cheat. Let's not let
1:03:07 them use it. [ __ ] really really
1:03:12 really that that like the only
1:03:16 thing less excusable than Apple not
1:03:19 innovating on AI is the
1:03:23 education
1:03:24 sector [ __ ] missing this boat. It is
1:03:28 shocking to me. It's [ __ ]
1:03:31 shocking. I mean, education's been
1:03:33 broken for [ __ ] decades because we're
1:03:36 still
1:03:37 teaching based on the the the
1:03:40 [ __ ] curriculum from the industrial
1:03:45 revolution, right? We haven't even
1:03:47 updated to like the internet age, much
1:03:50 less. We now have knowledge on demand at
1:03:54 at our fingertips.
1:03:57 students are going to enter a world that
1:03:59 does not resemble the world we have
1:04:01 right now. It's It's not even going to
1:04:03 resemble
1:04:04 it. And we're ignoring AI. It's it's
1:04:07 it's it's [ __ ] shocking. It's it's
1:04:10 it's shockingly
1:04:12 destructive. Apparently Claude had a
1:04:14 mass massive context window in
1:04:17 preferences. Oh, has a massive context
1:04:19 window in preferences. Let's go look at
1:04:22 that. I don't know what massive is.
1:04:24 They've always had like 200,000 tokens.
1:04:26 Is it bigger than
1:04:28 that?
1:04:31 Claude.ai.
1:04:36 [Music]
1:04:44 Oops.
1:04:46 Preferences
1:04:49 settings. Let's see.
1:04:52 Artifacts on feature preview
1:04:57 on appearance wouldn't be there.
1:05:01 Account
1:05:12 cla
1:05:14 settings. Uhuh. I don't see it.
1:05:18 They're going to find out real quick
1:05:20 that the kids can dance circles around
1:05:22 them at home. Yeah. Well, but the thing
1:05:25 that's [ __ ] up is you've got like what
1:05:27 what's here here's the result here.
1:05:29 Here's what I'm going
1:05:31 to what I'm seeing
1:05:34 is you've got Gen Xers kicking ass at
1:05:38 AI. You've got millennials like, "Yeah,
1:05:40 whatever, Dad." Millennials are just
1:05:43 kind of like pissed off that they're not
1:05:47 working right then you've got Gen Z who
1:05:54 are have been told that AI is bad and so
1:05:57 they're afraid of it and then I think
1:06:00 whatever whatever Gen next or whatever
1:06:02 the [ __ ] next one is below Z I think
1:06:05 they're going to be fine because I think
1:06:07 I think kids in grade school right now
1:06:09 like the education system's going to
1:06:11 [ __ ] work something out before they
1:06:13 get to high school I think they're going
1:06:14 to be fine but I think this Gen Z where
1:06:17 they're like afraid of it, you know,
1:06:19 combined with like, you know, who their
1:06:21 who they look up to, the millennials and
1:06:24 who their parents are. I don't know,
1:06:26 man. I It's just It just doesn't seem
1:06:28 good. It just seems like there's a big
1:06:29 [ __ ] black hole around those those
1:06:32 two categories of
1:06:35 people. Alpha then beta. Yeah, beta is
1:06:38 going to be absolutely fine, right? Beta
1:06:40 beta are going to um they're going to
1:06:43 grow up native with
1:06:44 AI. Like like Brandon's kid, he's four
1:06:48 and he says, "Daddy, I want to listen to
1:06:50 the song I wrote that he wrote in
1:06:54 Sunno." And Adnan's daughter who's seven
1:06:57 that co-wrote a book with her
1:06:59 dad, you know, based on the the adult
1:07:02 version of the book that her dad wrote
1:07:03 with Brian Moahan. Those kids are going
1:07:06 to be fine. The cusp kids that are
1:07:09 probably junior high right now,
1:07:11 yeah, they'll be okay, but touchy. The
1:07:15 high school kids, people going into
1:07:17 college right now, I think they're
1:07:18 [ __ ] I think they're [ __ ] because I
1:07:21 I also like if you're graduating college
1:07:25 in the next four
1:07:27 years, where's the entry level? Like,
1:07:30 where are the entry level
1:07:31 jobs? What are entry-level jobs?
1:07:35 [ __ ] low-level tactical execution work,
1:07:40 right? Oh, he's new. Just give him uh
1:07:43 tell him to do those TPS reports, right?
1:07:46 Tell him to write the copy and paste
1:07:48 this [ __ ] into the into the deck for the
1:07:52 CEO. Those jobs are
1:07:55 gone. Where do those kids
1:07:59 go? Especially if they've been told,
1:08:01 "Don't [ __ ] use AI. It's evil. It's
1:08:04 the world's greatest plagiarism m
1:08:06 machine. You're cheaters, you [ __ ]
1:08:09 losers. Learn our book
1:08:13 stuff. I don't know. They got to get
1:08:17 self motivated. They They should read
1:08:18 the Fiverr CEO's
1:08:21 letter. Anyway, beta are going to be
1:08:23 cyborgs. I agree with that. Kim Katkin,
1:08:25 I think there will be so many new fields
1:08:27 of research. Zool linguistics learning
1:08:30 interspecies. Yeah. Listen again. Clean
1:08:34 the windows. Exactly. Gen X will be
1:08:37 totally rad. Exactly. [ __ ] them, man.
1:08:40 Like, listen. We have We have all of our
1:08:43 lives, we just want to [ __ ] [ __ ] up,
1:08:45 right? Like AI is the ultimate [ __ ] [ __ ]
1:08:48 up
1:08:49 tool. It is. All right. Me got book.
1:08:53 Yeah, exactly. Did you see that the
1:08:55 number of translator roles has increased
1:08:57 since chat GPT is released? That's
1:08:59 fascinating.
1:09:03 There are going to be jobs we simply
1:09:05 can't imagine right now. And you know,
1:09:08 and I I can promise you who gets those
1:09:11 jobs are going to be the people that
1:09:13 radically get radical about learning AI.
1:09:17 Get like radically
1:09:19 curious, radically adventurous. Just
1:09:21 [ __ ] go in, start a [ __ ] Tik Tok
1:09:25 channel.
1:09:31 It's all about asking questions. It's
1:09:33 absolutely about asking questions.
1:09:35 Someone asked me the other day, their
1:09:37 kids in college I think might be
1:09:38 studying computer science and they're
1:09:40 like, you know, what else should they
1:09:41 study? I'm
1:09:42 like, instead of studying, they should
1:09:44 be learning AI. And I said, quite
1:09:46 frankly, they should probably get a
1:09:48 minor in
1:09:50 philosophy. And I actually I firmly
1:09:52 believe that philosophy,
1:09:55 psychology,
1:09:57 drama, something about language,
1:10:01 right? Language, sociology,
1:10:04 understanding the human
1:10:06 condition because those people that
1:10:09 understand people are going to be the
1:10:11 ones that kick ass at
1:10:14 AI. Throw in sprinkle in a little
1:10:16 technical skills with that. They're
1:10:19 going to be amazing.
1:10:21 All
1:10:24 right. I'm amazed how heavily
1:10:26 represented Gen X is around this. I It's
1:10:28 Well, Gen X So, a couple of things. Gen
1:10:31 X, we got to experience the worldwide
1:10:34 web. Like, we were there before the
1:10:36 worldwide web and we got to sort of live
1:10:38 through it changing the
1:10:41 world. So, we're we've got a visceral
1:10:46 understanding of what AI is about to do.
1:10:50 like we we have no sense of what it is
1:10:53 because it's just so much bigger than
1:10:55 the worldwide web was and so much
1:10:58 faster. But I think that's why you're
1:11:00 seeing Gen X do well in this because
1:11:01 it's like, "Oh [ __ ] this is a party.
1:11:05 Let me jump on this
1:11:06 bandwagon. Let me tell my boss to go
1:11:09 screw
1:11:10 himself. Go reinvent me a career."
1:11:18 Oh man, my degrees in technical and
1:11:20 creative writing are now serving me very
1:11:22 well. Yes, Joker. I'll get my for
1:11:25 folding chair and popcorn for
1:11:27 that. Oh
1:11:30 man. Round two
1:11:33 fight. We are the Mad Max generation.
1:11:36 All right, my voice is gone. I'm out of
1:11:37 here, people. Um, happy Tuesday night.
1:11:41 Hope you had fun. Um, I'm I'm I'm really
1:11:44 digging my new metaphor, the box, the
1:11:46 toilet paper, and the uh the tools. It's
1:11:50 working for me. Did I mention the Oh,
1:11:54 no.
1:11:56 Um,
1:11:57 so starting June 1st, so we've got the
1:12:01 AI salon. So, if you go to the the AI
1:12:03 salon website, which is the salon.ai,
1:12:06 and click join our community. Um, we've
1:12:09 been around for two and a half years.
1:12:10 We're coming up on two and a half years.
1:12:12 Um, tonight we announced we're launching
1:12:16 what we're calling AI Salon Mastermind,
1:12:18 which is a subscriptionbased area in the
1:12:21 salon for people who really want to go
1:12:23 dig deeper with AI, build tighter
1:12:26 relationships, things like that. Um, and
1:12:29 we're going to have like a ridiculously
1:12:31 cheap price for all of 2025, starting
1:12:33 June 1st, which is 1995 a month. And
1:12:36 then in 2026, it'll probably go up to 50
1:12:39 or 60 or 70 bucks. I don't know what
1:12:41 it'll be. Um, and you you come in as a a
1:12:45 founding mastermind and and you get that
1:12:47 that price for life if you come in in
1:12:49 2025. So, if you go to the AI Salon
1:12:52 website
1:12:55 um under the community corner section,
1:12:59 there's an FAQ about the AI salon
1:13:01 mastermind. You can go learn about what
1:13:02 we're up to there. So, do that. And
1:13:04 then, um, tomorrow, there's two things
1:13:06 to pay attention to.
1:13:08 One is at 4 pm mountain time. Um the AI
1:13:12 readiness project podcast with myself
1:13:15 and Ann Murphy happens. Um we have an
1:13:18 amazing artist on tomorrow who you're
1:13:20 going to [ __ ] love. So come to that.
1:13:23 Um if you go to
1:13:25 airespro.com you can see past episodes
1:13:28 and there's information about where you
1:13:29 can watch it there. And then tomorrow at
1:13:32 5:30 Mountain is at 5:30 Mountain. Is
1:13:35 that correct,
1:13:38 Brandon? Hold, please. Yes. Okay. 530
1:13:42 Mountain is the AI salon life hacks um
1:13:47 club. And if you haven't been to one of
1:13:50 those, they're a blast. It's just like
1:13:52 how to use AI to do cool [ __ ] in your
1:13:54 [Laughter]
1:13:56 life. Hale Lima is going to be fabulous.
1:13:59 Did you see the artwork that she made,
1:14:00 Ann? She made this really cool piece of
1:14:02 artwork she put on LinkedIn that
1:14:04 featured you and I and her with a cool
1:14:06 groovy background. It's really cool. So,
1:14:09 um, Ann Murphy's in the house. Ann
1:14:12 Murphy,
1:14:14 rockstar. Hey, Ann, I have a very
1:14:16 serious question for
1:14:19 you. How many pairs of sunglasses do you
1:14:25 own? It's got to be over a hundred. Am I
1:14:29 wrong?
1:14:31 every every video you're in. I'm like,
1:14:33 "Well, there's a banger set of glasses
1:14:35 and then the next one. And then the next
1:14:37 one." She's
1:14:41 laughing. She's got hats
1:14:44 going. She got the hair going and just
1:14:47 always some hip glasses. I lose them
1:14:49 all. Oh, it's an add
1:14:54 thing. That's the best. Okay, cool. Um,
1:15:00 all right
1:15:01 everybody. Yeah, I'm going to get out of
1:15:03 here. So, tomorrow night's Wednesday. I
1:15:05 don't think I have anything. Let me just
1:15:06 look at my schedule quick. Do I have it
1:15:08 up here?
1:15:11 No. Should be 8 o'clock tomorrow.
1:15:14 Wednesday. Um, yeah, I'm pretty sure
1:15:16 that's right. And,
1:15:19 uh, think that's it. I think that's it.
1:15:23 Hope you had
1:15:24 fun. Wear your party hats. Oh, yeah.
1:15:28 I have a
1:15:30 uh I achieve another decade
1:15:33 [Laughter]
1:15:37 tomorrow. I don't like it.
1:15:42 later.