6/12/2025 - AI Learning Lab: Runway ML, Midjourney, and Flow

Description

Chapters

Transcript

0:02 [Music]
0:13 [Applause]
0:16 No.
0:25 Woohoo.
0:27 [Music]
0:30 [Applause]
0:31 [Music]
0:50 [Music]
1:30 [Applause]
1:35 [Music]
1:37 Woke Grace is desperately heading his
1:40 old his old place. Dream to discover a
1:44 new space. Buried himself alive
1:49 inside his basement. tongue on the side
1:52 of his face. Working away on
1:55 displacement.
1:57 What it would take to survive.
2:02 Cuz when you're done with this world,
2:07 you know, the next is up to you.
2:12 And for once in his life,
2:16 it was quiet
2:17 [Music]
2:19 as he learned how to turn with the tide.
2:24 Yeah.
2:28 [Music]
2:30 And then
2:38 [Music]
2:44 One thing, this is a piece of advice I
2:46 can give to all of you. Uh, don't start
2:48 singing a song if you don't know how it
2:49 goes.
2:54 It's a It's a It's a What do they call
2:56 it? It's a comedy. Comedy. It's comedy
2:59 right there. That was a joke. That's
3:00 what they call a joke. It's a It's got a
3:03 setup which is a really bad stretch of
3:05 singing there with no words and the
3:07 wrong chords and then uh and then some a
3:10 a delivery of a welltimed mediocrely
3:13 timed
3:15 it words coming out of the mouth in a in
3:18 a direction toward comedy. You see what
3:21 I'm saying?
3:23 How's the hair?
3:25 It Listen, it's not important how we
3:27 feel, people. It's important how we
3:28 look. I think everybody knows that.
3:31 Everybody knows that.
3:53 That sounds like something.
4:15 Champy got a little scrap today with
4:17 another dog. The other dog did better
4:20 than Champy. Champ Champ did not did did
4:23 not farewell. We had our gate open. We
4:26 were just doing garbage or something
4:27 like that and someone walked by with
4:28 their dog and Champ ran out.
4:31 Little son of a [ __ ] He's a bad dog.
4:33 Were you a bad dog today, Champy?
4:36 [Music]
5:06 It's not simple to say
5:10 that most days.
5:13 [Music]
5:40 in a westerly
5:42 direction.
5:46 This car is my train.
5:51 I've been driving. I've been wondering
5:56 what it is I'm running from again.
5:58 [Music]
6:02 Feel like a ka man.
6:06 Falling on the 29.
6:10 And up ahead on that horizon
6:15 is a California line.
6:20 [Applause]
6:25 Another head of trucks carrying a wide
6:27 load.
6:29 Preap house cut in half.
6:33 Cute little front door into two windows.
6:36 My love
6:37 ain't sure whether to cry should laugh.
6:42 You see, I broke a home up myself once
6:46 as I stumbled to that dog.
6:50 I read a note by the dawn's light
6:55 said, "Don't you come around here
6:59 anymore."
7:01 Yeah. Well, I've had enough
7:06 on the street.
7:10 Never was good with the C.
7:14 That's what I've been told.
7:17 [Music]
7:25 [Laughter]
7:30 How is everybody? What's going down?
7:32 What's shaking?
7:34 All right. Valerie Cox in the house.
7:36 Sharon Crawford. Howdy. Howdy. Howdy.
7:37 The hair is more comedy than the
7:39 singing.
7:43 It's not that bad to Well, it's a little
7:45 Yeah, it looks like a bad It looks like
7:48 a It looks like a ventriloquist's puppet
7:51 hair.
7:56 Um, hope everyone's well.
7:59 Um,
8:01 hey Steo, what's happening? What's
8:03 shaking?
8:05 Um,
8:09 we're producerless tonight. All this
8:11 week, Brandon's out. He's having I don't
8:13 what they call Wait, what do they call
8:16 it? Um,
8:18 family time. Yeah, he's having family
8:22 time. I
8:24 That's I had to chat GPT it. I didn't
8:26 know what that meant.
8:28 Uh, it looks like you got in a fight
8:31 with Champion.
8:36 I'll tell you what, that was a that was
8:38 a rough dog had Champion by the neck.
8:42 Bit his ear,
8:44 you know.
8:48 [Music]
8:52 Statement. Heavy rain statement has been
8:54 issued for Denver.
8:56 Looks like it's going to rain.
8:58 [Music]
9:11 Sitting in this lonely town.
9:13 [Applause]
9:14 Wonder what things are going to change.
9:19 Dream my life away.
9:22 Seems these dreams have turned to a
9:24 bunch glass.
9:28 Get my nerve up, but my past is pulling
9:32 me down.
9:34 Yeah.
9:36 Wondering how long
9:38 this black sheep going to stick around.
9:43 [Music]
9:46 Somebody told me once before, he said,
9:49 "You can never go home again.
9:52 Won't you leave?
9:55 Tell me things to steer me away. Yeah.
9:59 From the truth of who I am. What? What's
10:01 going on, champ? Why'd you stop singing?
10:03 You thinking about that other dog, huh?
10:06 You thinking about that dog with teeth
10:09 around your neck? You get a little
10:11 distracted there, buddy?
10:14 He's got a little He's got a little
10:15 champ. He's got a little champ PTSD.
10:19 He was mid song and he just went
10:23 [Laughter]
10:28 Oh man.
10:31 Oh, good lord.
10:34 So, looks like some stuff is happening.
10:36 There's There's some stuff going on out
10:37 there. I got I think I got something fun
10:39 we could play with tonight. Well, I know
10:41 I know I've got something we can play
10:42 with tonight. I don't know if it'll be
10:44 fun. I It should be fun.
10:46 Joyy's here. She might know. Have you
10:48 played with the new Runway ML chat
10:50 interface, Joy?
10:54 [Music]
11:01 No. Okay, we can play with that tonight.
11:06 [Music]
11:08 I never met a fight I couldn't outrun. I
11:10 just I just avert my o eyes to avoid
11:13 confrontation. I'll be over here in the
11:15 corner. You're good. You just y'all take
11:17 care of it over there.
11:23 Not Championy, man. He went right in.
11:25 He's like, "Let's [ __ ] go.
11:29 Joy Pury, you'll let me teach you."
11:31 That's that's a rough road to hoe.
11:34 [Music]
11:39 Well, if you're new here, everyone, my
11:41 name is Kyle Shannon. This is the AI
11:42 Learning Lab. I'm the professor.
11:45 I'm not I I'm not qualified. I don't
11:48 really know [ __ ] So, we're gonna play
11:49 with a thing tonight that I don't I've
11:51 never used once. Well, maybe you should
11:54 have prepared your lesson plan. It could
11:56 that that would have been an option,
11:59 right?
12:02 If I weren't busy with 38 other projects
12:04 because AI makes it possible to amplify
12:07 yourself.
12:10 [Music]
12:14 AI is like, "What if ADHD people had
12:16 executive function?"
12:23 That's actually a pretty good definition
12:25 of it.
12:30 You know, you just have a brain fart and
12:32 rather than it being like a normal ADHD
12:35 brain fart where it just sort of, you
12:36 know, drifts off in the breeze, now you
12:38 got an AI to amplify it, put it in the
12:40 world. So, so we're littering the world
12:43 with all these half-thoughtout ideas
12:46 that we can make well thought out in 15
12:48 minutes.
12:51 [Music]
12:53 I started 86,000 prospects today and
12:55 finished two. Oh, projects today and
12:57 finished two. Well, the fact that you
12:59 finished them, it's that's pretty good.
13:01 That's solid. That's solid. All right,
13:03 let me flip my Let me flip my camera and
13:06 let me get the black bar up for the good
13:08 people on the Tik Tok.
13:10 There you go. Good people. Dr. Jay's in
13:14 the house. Mr. It in the house.
13:17 He's not He's not humble. He's He's
13:19 irregular. Exactly. Exactly. People,
13:25 [Music]
13:29 Kylo's very humble,
13:31 [Music]
13:38 you know. I'm that weird. I'm that
13:40 weird. Um,
13:43 [Laughter]
13:50 I'm that weird combination of delusional
13:53 confidence and low self-esteem
13:55 [Laughter]
14:02 where I can do remarkable things, but
14:04 the minute minute anyone notices it, I'm
14:06 like, I I had nothing to do with that.
14:13 Oh man, [ __ ] being neurosicy is the
14:16 best, ain't it?
14:18 Oh man,
14:20 I thought I thought I finally figured
14:22 out how to get connections to work in
14:24 settings, but no. Okay, so Jeff, tell me
14:27 tell me when those [ __ ] work and I'll
14:29 play with them because it just it just
14:31 seemed annoying. So it sounds like it's
14:35 still annoying.
14:37 Um, okay.
14:39 Same Z's.
14:42 Yeah. Delusional self-confidence and
14:44 crippling low self-esteem.
14:48 It's the best combo.
14:52 There's a false humbleness and a belie
14:54 and and a deep and a projected belief
14:56 that you believe in everything.
15:00 [Music]
15:08 You know what's funny? Um, well, I don't
15:10 know if it's funny, but it's striking me
15:12 as funny right now. Um,
15:17 one of the skills you gain as an old
15:19 only child is to just go play with
15:21 yourself, right?
15:23 And
15:25 and so you kind of have to do
15:28 everything, right? You have to like make
15:31 the fort and then climb in the fort and
15:33 then be the bad guy. I mean, you could
15:36 go get friends, but who does that?
15:40 And so, for all of my life, like we we
15:42 talked about this la last night. I
15:44 forget who it was. Someone said, you
15:46 know, jack of all trades, master of
15:48 none. And I think with AI,
15:53 we're we're quickly moving into the jack
15:55 of all trades, master of all realm. So
15:57 if if you're one of those people that
16:00 was a was a
16:03 a self-starting, selfactuating,
16:07 you know,
16:09 multi-passionate.
16:10 We're coming into our time, people.
16:21 [Music]
16:30 [Applause]
16:31 [Music]
16:34 [Applause]
16:43 [Music]
16:52 Didn't need to make it.
16:56 Um, okay.
17:01 Now to implant in it in the brain meets.
17:04 Um,
17:06 how are we the same person?
17:11 Oh man, I got that sort of world
17:14 worldview as the introvert big brother
17:16 with a highly social sister that that
17:19 that wouldn't be seen near near me.
17:20 Yeah, exactly. That's awesome. Um,
17:25 if you missed last night,
17:28 well worth the rewatch. Really?
17:32 Okay, good.
17:34 See, we have a pleased customer. Sharon
17:36 Crawford here. That's That's what they
17:38 call a testimonial right there. If you
17:40 missed last night, it's well worth the
17:42 watch, the rewatch.
17:47 And then you you would ask me what what
17:49 was on the the I the the curriculum
17:52 yesterday. I I don't remember
17:58 [Laughter]
18:02 the these shows start I just go into a
18:06 tunnel and I don't know what I say. I
18:09 never go back and look at them. When I
18:11 do I'm like oh that was
18:17 That was unfortunate.
18:22 Occasionally I'll crack myself up.
18:24 Occasionally there's some comedy in
18:26 here. Um, all right. Let's see what
18:30 else. What I I saved a uh something in
18:32 my
18:36 what's it called? My my mind.
18:40 So, what did we do last night?
18:42 Someone did a really cool particle
18:44 simulator thing in Gemini 2.5.
18:48 Oh, we did some we did some midjourney
18:50 rating last night. We went through the
18:53 Sam Alman essay. Ah, that was it. I
18:55 don't I don't really remember what I I
18:57 said all that much about it, but it was
18:59 a really interesting essay. Um, yeah,
19:01 that's that's probably worth worth going
19:03 back because there was some pathy
19:04 commentary in there. Source camp finally
19:07 here just right on time. Uh, that should
19:11 be unmerged. If wait on merch, if you
19:13 missed last night, AI learning web, if
19:16 you missed last night, it's all right.
19:19 If you missed last night, dot dot dot,
19:22 you'll be fine.
19:24 Carry on.
19:29 All right.
19:30 Have you tried Miniax? I tried Miniax
19:32 ages ago. Um, I can go look at it. I
19:37 just went and looked at Cling.
19:38 Everyone's blabbing about Cling 2.1. I
19:41 got to tell you though, since doing VO3,
19:44 where it does the acting and it does the
19:46 voices and everything like that, I feel
19:47 like all the other video tools right now
19:49 are just like, eh.
19:52 Like, give me something that just vomits
19:54 out the full thing so I don't have to
19:56 work. I just want to get my ideas out of
19:58 my head and have them be perfect. We're
20:00 not there yet.
20:02 Although apparently there's there's some
20:05 self-recursive
20:07 project within OpenAI called Alice that
20:10 is
20:12 getting better on its own. I don't know.
20:14 Something something's going on inside
20:15 OpenAI, but it it's not good yet. Um
20:19 Kyle should rewatch it. Yeah, I should
20:22 rewatch it to figure out what the [ __ ] I
20:23 said. Um I was there and also don't
20:27 remember. Good archetypal architect.
20:30 That's good. So, if you were here and
20:32 spent two hours with me last night and
20:34 you also don't remember what I said,
20:36 that's good because what that means is
20:40 my education is just happening
20:42 subliminally.
20:44 Like, you know, you were probably out
20:46 there microwaving
20:48 a hot pocket, you know, making, you
20:50 know, images of cat paws in mud or
20:54 something like that, right? But
20:57 everything I said was going in there was
20:59 infiltrating
21:00 Solid. All right.
21:03 If it's your first time here, Exactly.
21:07 Oh, yeah. Oh, if it is your first time
21:09 here,
21:11 I may occasionally finish a sentence. I
21:14 may occasionally do something that I
21:16 promised that I would do. Um, and I may
21:19 accidentally
21:21 occasionally teach something, but really
21:25 not intentional. It's just kind of, you
21:27 know, collateral damage from talking as
21:29 if I know what I'm talking about. You
21:32 know, the the unreasonable confidence
21:36 combined with the crippling low
21:38 self-esteem.
21:43 Oh, good lord. Good lord. Good lord. All
21:45 right, so let's go. Let's go play with
21:48 some stuff. I think we should go let's
21:50 go rate some more videos. rank some
21:52 videos
21:55 because why not? Let's let's be good
21:57 Samaritans. You should you should join
21:59 along in this by the way. So, we're
22:02 going to go to midjourney
22:05 and we're at midjourney/rank-vide.
22:10 midjourney.comrank-vide.
22:16 And then
22:18 if you if you hit one on your keypad,
22:23 that votes for the left one. If you hit
22:24 two, that votes for the right one. If
22:26 you hit three, it skips the the that
22:29 pair.
22:31 All right, let's see. Her legs are
22:35 backwards and weird in both cases cuz
22:38 the image was [ __ ] up. Let's give that
22:39 a two. Let's give that a No, the drops
22:45 are backwards. Let's give that a one.
22:48 Nope. Give that a two.
22:53 Three. Neither of those were good. Oh,
22:56 sorry. Sorry, Tik Tok. All right.
23:00 Why must we do it?
23:02 Oh, you don't have to do it. It's just
23:05 that in doing this ranking, we're
23:08 training like like I feel like people on
23:10 this channel have have good points of
23:13 view, good creative points of view. Um,
23:17 and so by doing this ranking, we're
23:20 actually training the midjourney video
23:22 model. So it will suck less if we if we
23:25 do our do do our rankings.
23:29 We'll go with two on that one. Go with
23:31 one on that one.
23:33 We'll go with one on that one. Two on
23:37 that one.
23:44 One.
23:46 Three.
23:48 Three. Three. Is I'm skipping them
23:51 because I think they both suck. And it
23:54 like I don't think it
23:57 I don't think they care. I think it's
23:59 just
24:01 um
24:03 you're not gonna you're not gonna
24:07 you're not going to reinforce either of
24:09 those videos.
24:13 One. Oops.
24:16 Two.
24:22 Two.
24:24 Two.
24:29 Two.
24:30 One. Two.
24:34 [Music]
24:38 Three. Three.
24:45 Yeah. This is going to It's It's
24:48 fascinating that when a tool comes out
24:50 like V3
24:53 and then all of a sudden you're back to
24:54 silent video.
24:56 It's like the silent movies. It's It's
24:58 just not the same. One.
25:01 One.
25:04 One.
25:12 3.
25:14 3.
25:16 2.
25:18 1.
25:22 3.
25:24 Three. One.
25:28 Two.
25:31 Uh,
25:34 two, two.
25:40 One,
25:42 three,
25:44 two,
25:47 three. Something about bugs. These video
25:51 models suck at. One.
25:54 One.
25:56 Two.
26:00 All right, I'm bored.
26:04 Where's Brandon this week? You know,
26:07 Teton Todd, he um
26:11 there's this thing. It's called a
26:13 vacation, I think,
26:16 like he went with his wife without the
26:18 kids. Like kids stayed at the mom's and
26:20 then the him and the wife were going off
26:22 and having a fine time.
26:26 something like that. I don't get it. I
26:28 don't understand it. Um,
26:32 all right. Isn't this how Facebook
26:34 started? No, that was swipe left, swipe
26:36 right. Yeah, it's pretty much the same
26:37 thing. Um, so how are people making all
26:40 the big blog
26:43 Bigfoot blog videos? V3 or Hedra? That's
26:46 probably V3.
26:49 So, if you want to see V3, let's go to
26:52 you go to go to flow
26:55 dot. There's a bunch of places you can
26:57 do it, but the the official Google one
26:59 is flow.google.
27:05 Create with flow.
27:08 Vicky Baptiste, I don't understand
27:09 vacation either. I I know it's so bad. I
27:13 haven't taken one in so long that it's
27:15 it's getting kind of criminal at this
27:16 point. So, I should probably get my [ __ ]
27:19 together. Um, let's see. Zero AI
27:23 credits. Let me
27:26 Sign out.
27:28 Sign in with Google. Sign in
27:41 LaunchBlow.
27:44 All right, here we are.
27:48 Okay, new project. So, we're going to
27:50 put in here. What did you say? Bigfoot.
27:52 Let's go look at what are the Bigfoot
27:53 videos. I don't know what you're talking
27:54 about. Let's see.
27:57 Bigfoot
27:59 V3. The reason to use V3 is it does it
28:04 does talkie talk [ __ ]
28:11 Okay. Bigfoot vlogging.
28:22 Okay.
28:23 All right. So, that's it. So, we're
28:25 gonna
28:27 we're gonna jump in here to flow.
28:30 Well, let's go to chat jpack. Ty,
28:34 the goal for me for 2025, there was a
28:38 big internet outage today, by the way.
28:41 The goal for me for 2025 is not to use
28:43 my brain, as we know. Yeah, we're pretty
28:46 clear on that. You big dumb dumb. Shut
28:48 up.
28:50 Um,
28:52 when you sit down in front of the prompt
28:54 hole and you want to do something
28:55 creative, like write some a Bigfoot
28:59 vlogging prompt, you could use your
29:03 brain to like come up with like, okay, a
29:06 Bigfoot in the woods of Oregon with the
29:10 and then, but you don't have to.
29:14 Projects can now do more. You can now
29:16 run deep research and use voice mode
29:19 within your projects. That's super cool,
29:21 actually. We're gonna go play with that,
29:23 too. All right. So, I'm gonna say,
29:25 "Write me a video and acting
29:32 prompt
29:35 for a single clip of a Bigfoot
29:41 vlogging.
29:45 Make it. Make it. Damn it. Make Make
29:52 Hang on. It
29:55 hilarious. See, I could write comedy or
29:58 I could just have chat BT write comedy.
30:04 Tone, outrage, self-important, weirdly
30:06 articulate. Wait, I said Oh, wait.
30:10 Sorry.
30:12 Uh, make it
30:16 in 8 second clip. No edits.
30:27 Bigfoot is holding a shaky phone selfie
30:29 style. He's mid rant, sweaty, slightly
30:32 out of breath with a squirrel on his
30:34 shoulder which he ignores completely.
30:36 You get one continuous take. No cuts, no
30:39 effects, just raw Bigfoot energy. Look
30:41 dead into the camera. I eat organic. I
30:44 compost my poop and y'all and y'all
30:47 still think I'm a myth. That's pretty
30:49 good. That's actually pretty good.
30:54 Play it. Play it like you're one
30:56 Starbucks away from a a full breakdown.
30:59 Bonus if a fl bug flies in your face.
31:01 All right, I'm going to take all these
31:03 including the notes.
31:07 Okay, Bigfoot's holding a phone selfie
31:09 style. Okay, so now we go to flow.
31:12 We're going text to video. Now, this is
31:14 important. They always default
31:18 V3. They always default to V2
31:25 because they're losers.
31:29 So, you got to go to V3. We'll do V3
31:32 free. V3
31:35 fast because it's a fifth of the tokens.
31:39 One fifth. 20 tokens instead of 100.
31:41 pretty good. Um, and then I'm going to
31:43 paste in that prompt and then we're just
31:45 gonna see what [ __ ] happens. And then
31:48 we're going to have our own Bigfoot vlog
31:50 video,
31:51 which by the way,
31:55 get used to
31:58 what we're starting to experience in
32:00 2025 where someone puts something out
32:04 and if it's good and original, it will
32:07 be knocked off immediately. and entire
32:10 um Tik Tok channels will be started
32:12 using that gen genre within a day,
32:16 probably within an hour. Who has the
32:19 most PTO on the books?
32:23 Well, I So, I'm 13 years into Story
32:25 Vine. We've got we've got unlimited
32:27 vacation policy. I think I've probably
32:30 taken
32:33 maybe four weeks
32:36 in 13 years.
32:40 That's bad.
32:45 [Music]
32:53 [Music]
32:55 Bigfoot is holding the phone selfie
32:57 style. Okay, it's doing its thing. Let's
32:59 flip over to
33:02 Runway. Okay, so Runway, if you don't
33:05 know, RunwayML is another one of the
33:07 video tools,
33:09 but they've now got a chat interface, so
33:11 you can just start chatting and making
33:13 making images.
33:16 We have a use it or lose it policy. Oh,
33:18 that sucks. Yeah, that's that's drag.
33:23 It should be use it or get paid for it.
33:28 Um, so how are people making all the
33:30 Bigfoot? Oh, that's what we're doing
33:31 right now. Okay. Um, so let's see. Um,
33:34 let's let's paste. Well, let's say, hey,
33:38 I want to make
33:42 one of those
33:45 Bigfoot
33:48 vlogging parody
33:51 videos.
33:55 Make me an image to start.
34:01 Let's just see how it does.
34:05 I'll create an image of Bigfoot vlogging
34:07 so that you can use that as your
34:09 starting point for your parody video.
34:12 This is cool. Generating 10%.
34:19 Let's go back to Flow. See if we have a
34:20 video yet. We do. All right.
34:25 Let's see if our little prompt worked
34:27 here, people. I got to change up my
34:29 tabs.
34:33 Oh, this is really good. This is really
34:35 good. It's really good.
34:40 This is going to be good. This is going
34:42 to be good, people. Um, no, that one.
34:45 Okay, maybe I should just download it.
34:47 Why can I not?
34:51 Oh, there we full screen. Okay.
34:55 And we'll go back here and I'll turn
34:57 this up. All right.
35:00 And we'll do the black bar. Look at it.
35:01 Look at me self-producing here. Okay.
35:06 I eat organic. I compost my poop. And
35:09 y'all still think I'm a myth.
35:12 [Laughter]
35:14 I eat organic. I compost my poop. And
35:17 y'all still think I'm a myth? Ah,
35:21 a squirrel. I eat organic. I compost my
35:25 poop. And y'all still think I'm a myth.
35:27 Ah, it's so good.
35:35 That's going right up on the Twitter.
35:39 It's good, people. It's so good. How do
35:42 you download this [ __ ]
35:45 How do you download it?
35:48 here. No.
35:51 Add to scene. Can I download this?
35:54 Save as
36:00 I eat organic. I compost my poop. And
36:03 y'all still think Oh, download. Jesus.
36:08 Oh, and I can upscale it. All right. So,
36:09 we're going to upscale it. That's That's
36:10 a cool thing, too. In V3, you can
36:13 upscale it to 1080p. It It does it.
36:16 Right click the video. No, that didn't
36:18 work. But you know what did work? Irana
36:21 or who is that? Iran ruff. Um the
36:26 download button actually works
36:33 while I was yelling at the screen. I get
36:35 mine paid out each year. Oh, that's
36:37 cool. Very nice. The legacy hours are
36:40 from before the use it or lose it kicked
36:42 in. That's really funny. Uh, I've been
36:45 playing with it, too. But does anyone
36:48 else smell their wallets burning? Yeah,
36:51 V3 is not cheap. It really is not cheap.
36:55 So, that's the $1 per video. Or is it
36:57 less? Um,
37:00 well, the first three months are 125 a
37:03 month and you get 12,500 credits. So,
37:09 um, hang on.
37:11 So, if you're using if you're using um
37:16 the fast mode, which I just did.
37:19 Uh oh. Oh, my iPad's dead.
37:23 Hang on.
37:26 Calculator.
37:29 So, if you get 12,500
37:33 credits divided by 20,
37:37 you get that can't be right. 625. No, I
37:40 think that's right because if you do
37:43 12,500
37:45 divided by 100,
37:48 that's 125. Yeah, I think they said it
37:50 was like 87 videos
37:54 in a month was the limit. So maybe it's
37:56 like 125. So 125 bucks.
38:02 If you're if you're using the fast mode,
38:06 then you're talking like 20 25 cents a
38:10 video, which is not bad for that for
38:13 that quality is pretty good. It's pretty
38:15 good. Finished upscaling. Dismissed. Did
38:17 it download?
38:19 Yes. All right. So, let's Oh, you should
38:22 be able to see this. So, here it is in
38:25 QuickTime.Uh.
38:28 Uh,
38:31 and this is now upscaled. I eat organic.
38:34 I compost my poop. And y'all still think
38:37 I'm a myth.
38:48 All right, we're going to we're going to
38:49 chop off the end.
38:55 So, we'll just have him jump out of
38:57 screen.
39:09 I think that's good there. We'll say
39:12 split clip. Bang.
39:15 Drop the end off this bad boy. Go done.
39:18 Hit save.
39:20 We'll call this Bigfoot.
39:23 That's how they're making these Bigfoot
39:25 videos.
39:28 It's pretty easy, wasn't it?
39:31 So, what's the lesson we learned here?
39:33 Don't use your brain. Have ChatGpt write
39:36 your creative prompts for you because
39:38 it'll write it better than you were. The
39:39 fact that it gave acting notes at the
39:41 end there and it and it seemed to follow
39:45 them, I wouldn't have written those
39:47 because I'm lazy. I got ADD. I want to
39:49 move on. I want to make a video. Make me
39:50 a video. Make me a video. point me
39:52 video.
39:54 Don't use your brain, people. You're the
39:57 bottleneck. You're the bottleneck.
40:00 Okay?
40:04 Give up the ego. Let it go. You're not
40:07 smarter than the machine. You're not.
40:14 I'm smarter than the machine. I have a
40:16 degree in philosophy. Okay. All right. I
40:19 guarantee you Quinn knows more than you
40:21 do about philosophy. She's read all the
40:23 books and has them memorized.
40:29 I love the Yeti ones. We could we could
40:31 come up if you guys want to come up with
40:32 some other meme like thing like, you
40:35 know, babies doing comedy scenes or [ __ ]
40:37 like that. We can do that. Let's go.
40:40 Let's go. Uh let me change my uh change
40:42 my little sharing settings here so you
40:44 can all watch me
40:47 write it. So, so you can all watch me
40:49 write a Twitter post.
40:53 What did you do last night? I was on
40:56 this channel called AI Learning Lab. Did
40:58 you learn anything? Uh,
41:01 no. I uh I watched a guy write a Twitter
41:05 post.
41:08 How long did you watch it? Uh, two
41:11 hours. Two hours. Yeah. Yeah. What are
41:13 you doing with your life?
41:16 Right.
41:18 I wouldn't there's there's probably like
41:20 support groups for people that show up
41:22 to this thing a lot. It's called the AI
41:24 salon. Go into the irregulars channel.
41:30 Okay, let's see. Um,
41:34 word on the street.
41:40 Watch this. or should I say
41:44 in the woods
41:49 is that
41:51 Bigfoot
41:58 um
42:04 has
42:09 Bigfoot But
42:12 vlogging
42:14 videos
42:19 keep turning up.
42:25 Oh, lookie.
42:28 Here's one. Now,
42:36 by the way, if you're new here,
42:39 what this channel really is is an
42:42 opportunity for me to entertain myself.
42:45 You guys are welcome to hang out, but
42:48 that's that's pretty much what this has
42:50 devolved into. Okay? All right. I don't
42:54 want judgment. I don't want your pity.
42:57 I got things going on up here. It's just
43:01 they're not always that apparent to
43:03 people, you know, considering intellect
43:06 and expertise and qualifications.
43:09 But it's there.
43:11 It's right in this general probably
43:14 toward the back vicinity.
43:21 Do you laugh in your sleep? I don't
43:22 think so. Apparently, I snore like a
43:24 [ __ ] but I've never heard it.
43:26 [Laughter]
43:30 Oh, good lord. Good lord. Good lord.
43:33 Bigfoot. All right. I eat organic. I
43:35 compost my poop. And y'all still think
43:37 I'm a myth?
43:42 [Laughter]
43:46 I think it's really good.
43:49 There's a fly buzzing around him.
43:53 But it's Kyle a parent
44:00 AI anonymous. Yeah, exactly. Hi, my
44:03 name's Kyle. I like AI. Hi, Kyle.
44:08 What remarkable things have you done
44:10 today? So many things.
44:14 All right.
44:15 Ah, look. Twitter. Still still can't
44:18 figure out how to [ __ ] upload
44:20 anything. Can we? No.
44:27 Dear Elon Musk,
44:30 there's this thing called an upload
44:31 service sitting somewhere in Twitter. Uh
44:35 what you now call X. Um it doesn't work.
44:40 Note to Elon, [ __ ] fix it.
44:48 Post. All right. It worked the second
44:50 time. Second time's the charge. Charm,
44:53 charm, charm. Jesus, I sound drunk. Oh
44:58 man, they need extend. They do need
45:01 extend. Well, I think they do have
45:02 extend, but you can't extend voice
45:04 clips.
45:07 Wait, why did it upload two of them?
45:09 Seriously?
45:14 Got to be [ __ ] kidding me. Update.
45:16 All right, there we go. So, do me a
45:18 favor, people.
45:20 Go to my Twitter, Kyle Shannon,
45:24 find a Bigfoot and like like it and
45:27 tweet it and retweet it and then like
45:29 copy the link and put it on LinkedIn and
45:32 say, "Listen, I work with one of the
45:34 most talented,
45:37 genius AI instructors in the history of
45:40 AI.
45:43 And this is the kind of work he's
45:44 producing.
45:47 [Music]
45:48 [Laughter]
46:00 I compose my poop. I love the way he he
46:04 like the acting in this is actually
46:06 quite good. Like he hits the right
46:09 timing.
46:20 Oh,
46:23 it's coming through. But think about
46:25 Let's see. Kyle, go look at irregulars,
46:28 please. Oh, boy.
46:31 All right. Fine, fine, fine. Let's go to
46:34 irregulars, shall we? I love that we
46:36 have Serena's flowers around irregulars
46:38 now. That's going to be a permanent
46:39 thing in the in the channel. No, good.
46:42 Here's a here's this is good Danielle.
46:45 Okay, wait. I got to change my uh Yeah,
46:48 by the way, if you've got access to VO3,
46:51 um make cool videos, share them in
46:53 irregulars. Um we also now have in the
46:57 community corner section. The look what
46:59 I made channel is now open to everyone.
47:01 It was behind the firewall uh be behind
47:03 the mastermind firewall for for a couple
47:06 of weeks there, but we we liberated it.
47:08 It's back out. Um, so, so put it in
47:11 either irregulars or into uh, look what
47:14 I made. All right. And that's at the AI
47:17 salon. And if you don't know what where
47:18 what that is or where that is, hold,
47:20 please. You go to here, the salon.ai,
47:26 click on join our community, and you'll
47:28 be in. All right. Beautiful. Fantastic.
47:31 Bob, tell him what he's won. So, I got
47:33 to stop this. Start this. Start this. Do
47:37 this. Do this.
47:39 Now, make this video big.
47:44 Uh,
47:46 why do I not have audio here?
47:52 Go.
48:09 Wait.
48:11 Oh, come on. You know, I shared this
48:15 screen and then when I click on this, it
48:17 opens a new tab.
48:20 It's not fair, people. You don't
48:22 understand.
48:24 You don't understand how hard it is
48:26 being me sometimes.
48:30 Hang on, everyone. Just Just shut up. I
48:34 know there's an issue. I understand
48:36 there's an issue. Shut up. I know how to
48:38 fix it. I've got technical skills. Marge
48:42 taught me how to do this. Everybody just
48:45 shut up. Stop picking on me. All right.
48:51 Kim the Gorilla here. So, like Becky
48:54 found my secret food stash. Like
48:57 literally all of it. Becky is boarding
49:00 up the entire southwest corner. The only
49:02 place I could sneak snacks in from you
49:04 guys, it's giving national security
49:05 breach over a bag of Takis. Okay, so
49:08 like small update operation bring Kim
49:11 food is now closed.
49:15 Okay, besties. We're moving to breakout
49:17 attempt number two. Do you like have any
49:19 ideas or anything? Drop them below. It's
49:21 time to get out of here.
49:24 That's great. That is awesome. All
49:27 right, beautiful. Fantastic.
49:30 Bob, tell her what she's won. All right,
49:32 let me go back to I got to share a
49:34 different way now so you can see the
49:35 other thing I do.
49:44 [Music]
49:51 All right, fantastic. Let's see. Let's
49:54 carry on. Carrying on.
49:58 Oh, let's go back to X. See if any of
50:00 you gave my thing some love. Three of
50:03 you did. Stunning footage from Dr. J.
50:09 It's a good comment.
50:12 All right. Nobody I got Listen, there's
50:15 53 [ __ ] people in here. There's 48
50:18 people on on the combination of YouTube,
50:20 LinkedIn, and X. Go to my ex channel,
50:23 Kyle Shannon, and like my Bigfoot video.
50:26 I don't ask much in here. I give and I
50:29 give and I give and this is the thanks I
50:31 get. You people are awful. All right,
50:35 let's listen to him one more time. It
50:37 makes it does make me laugh.
50:46 I compost my poop.
50:53 A pink bow. A pink bow for everyone.
50:57 Pink bows everywhere. All right, let's
50:59 go back to runway. See if we got our our
51:04 image there. I don't like that. That
51:06 doesn't feel very real. Well, let's go.
51:08 Let's go talk to it. Looks like we're
51:10 having some issues.
51:13 Reload page.
51:16 Uh,
51:18 okay. Wait, can I go back into my chat
51:21 history?
51:22 Where's my chat history? You dumb dums.
51:25 Did you lose my flipping history? Oh,
51:27 wait. Sasquatch selfie. Okay. Uh, make
51:32 it more like a video like a a VHS
51:40 video camcorder
51:45 and make
51:47 the
51:50 big foot
51:54 more um
51:59 natural. natural and
52:03 realistic.
52:06 Um, it feels
52:09 like an animation frame
52:13 now. All right, let's see what this
52:15 does.
52:17 All right, you want me to stop folding
52:19 the laundry? I have and do that now.
52:21 Forever hooked. Take taking my Twitter
52:24 request personally. I like it. No, you
52:26 can finish. Listen, Forever Hooked.
52:28 Here's the deal.
52:30 Laundry is important. I encourage you
52:33 finish folding the laundry, but as soon
52:36 as you put the last tidy whiteies in the
52:38 in the little trifold, you know,
52:40 envelope kind of thing and then drop it
52:43 in your in your drawer. Drop your
52:45 drawers in your drawer.
52:47 Jump over to jump over to Twitter. Give
52:52 Wait. Okay.
52:55 No, he's not holding a camcorder.
53:00 I want it to look like he's being
53:06 recorded with one.
53:10 He's
53:12 uh
53:14 filming
53:16 uh like uh he's he's
53:19 talking into the camera as if he
53:25 found it in the woods.
53:29 All right, let's see. Let's see what
53:31 this does.
53:33 I don't know the I don't know the image
53:35 model behind this. I guess it's their
53:37 image model.
53:39 Oh, Valerie Cox just went over to
53:41 Twitter. I just got a notification.
53:43 Thank you, Valerie.
53:47 I like yelling at my audience. Kyle, do
53:50 you still have Chat GBT Pro? I don't.
53:53 Um, I've been wondering what you think
53:55 of the recent upgrade. This is gold cam
53:57 ken. Um,
54:01 let's see. Let's see how this did.
54:05 Oh, that's pretty cool. That looks
54:06 pretty good.
54:09 That's more like it. A POV shot from a
54:11 camcorder. But it it doesn't look like
54:14 analog. Like where are the analog scan
54:20 lines?
54:23 This is VHS,
54:29 not digital.
54:34 Come on. Don't be a dum dum.
54:39 One of my favorite prompts. Don't be a
54:42 dumb dumb.
54:44 Make it Blair Witch style. Oh, that's
54:46 pretty good, Brian Whitney. That's
54:47 pretty good. I like that.
54:50 You're absolutely right about the VHS
54:52 look. You know what? All of you chat
54:55 GPTs, all of you chat bots, I don't need
54:59 you to acknowledge that I told you it
55:01 was a piece of [ __ ] Oh, you're
55:03 absolutely right, sir. I could have been
55:06 much better at that. My apologies. Just
55:08 do your work. Just do your work. Okay.
55:12 When we're having a little conversation
55:14 about the color of the Why is the sky
55:16 blue? Then you can be polite. Then we
55:18 can have a little like understand
55:20 context. You're a large language model.
55:22 Figure it out. Still doesn't have scan
55:25 lines. Still doesn't have scan lines.
55:29 Still doesn't have And it looks like It
55:31 looks like it was sculpted out of a
55:32 tire.
55:34 You got to be kidding me. Got to be
55:36 kidding me.
55:38 How do I get out of here? Oh, close
55:40 that. All right.
55:42 It looks like a fake model
55:50 made of tires.
55:53 Make the big foot
55:57 scary real. And there are still no VHS
56:04 CRT tube
56:07 scan lines.
56:11 It should
56:13 look like I am watching
56:17 this on an old TV,
56:22 you know, like where
56:25 VHS tapes played.
56:32 So, so couple of pieces of strong
56:36 educational best practices and advice.
56:39 Don't use your brain and yell at your
56:41 GPT.
56:43 Call it an idiot. That's if if you take
56:46 one thing from me tonight, call your
56:48 chat GPT an idiot.
56:51 It will not make anything better, but it
56:54 will make you feel good. All right.
57:01 When I start swearing at chat GBT, I
57:03 know it's time to take a break. That's
57:05 just when it's getting good. Oh my god,
57:07 look how bad this is.
57:10 Look how bad that It looks like a bad
57:13 video game.
57:21 It looks
57:24 like a bad
57:26 video game.
57:28 Make it like the big
57:32 foot is
57:35 all too real.
57:39 I'm just exhausted.
57:41 Um,
57:44 SI can affect your mind. telling wait
57:46 telling
57:48 what's SI
57:51 oh AI
57:56 SI can affect your mind I'm like what's
57:58 that an acronym for it's not it's not SA
58:00 I know what SA is not that AI AI can
58:04 affect your mind
58:07 in telling you
58:10 yeah I I absolutely think AI has
58:13 affected my mind Um,
58:18 and I think largely in a in a positive
58:21 way.
58:23 You know, I joke I joke about not using
58:26 my brain, but what I what I actually
58:28 mean by that is I'm I'm I'm working hard
58:31 to not use my creative cycles on
58:34 low-level creative stuff. And you could
58:37 argue that writing a prompt is highle
58:39 creative stuff. I don't think it is. I
58:41 think what's the highle stuff is like in
58:44 this case I want a Bigfoot that looks
58:47 real. Like the thing we did in V3 looked
58:49 real. The thing that we're doing here
58:51 looks like [ __ ] So this I'm just I'm
58:52 not happy with this image model.
58:56 The only thing that's really important
58:58 right here is that
59:03 I'm get I I that I'm putting my creative
59:06 energy in. Is this thing right?
59:09 Um,
59:11 and you know, in fact,
59:14 let me let's go back to chat GPT for a
59:16 second. We're going to copy that
59:21 and go back to runway. I'm going to say
59:23 let's start over. Make a still
59:29 from
59:32 this description.
59:36 So this is the same description that we
59:38 just used V3 for
59:41 that we did in chat GPT.
59:48 Brandon's in the house. What's
59:49 happening, Brandon? We were just trying
59:52 to figure out what that word is for when
59:54 you go away with your family and spend
59:57 personal family time together. Couldn't
1:00:01 remember what it was called, but good on
1:00:03 you for doing that thing. I think it
1:00:05 begins with a V.
1:00:09 Don't you remember, Kyle? That's what
1:00:11 real life looked like. The VH cameras
1:00:13 were perfect.
1:00:19 Oh, that's interesting. It made this as
1:00:20 a vertical image.
1:00:31 Yeah, it's just not it's just not very
1:00:33 realistic.
1:00:39 It's got that AI sheen to it. Like
1:00:41 everything looks like it's been sculpted
1:00:43 out of clay,
1:00:45 you know? It just looks like bad makeup.
1:00:51 As opposed to the VO thing which looks
1:00:53 really good. Like look at that.
1:00:57 [Laughter]
1:01:08 I eat organic. I compost my poop. And
1:01:11 y'all still think I'm a myth?
1:01:15 [Laughter]
1:01:24 Oh man. All right. So I am not groovven
1:01:28 on Runway ML's
1:01:32 chat chat mode. Okay. Let me see. Um now
1:01:37 make it a video.
1:01:41 Let's let's see how this does. We'll see
1:01:43 it because Runway is ostensibly a video
1:01:47 company, not an image company.
1:01:52 I think it would be helpful if it would
1:01:54 show you the prompt it ended up using.
1:01:57 Yeah, I know. I agree with that, Kevin.
1:01:59 I I feel like these tools go out of
1:02:01 their way to to obuscate what they're
1:02:04 doing. Didn't, you know, don't want to
1:02:06 give away the golden secrets, but like,
1:02:08 you know what? Just give us the option
1:02:11 to flip that [ __ ] on so we can learn,
1:02:14 you know? I agree with you.
1:02:19 H.
1:02:22 Does he have a squirrel on his shoulder?
1:02:24 He sure does.
1:02:27 I eat organic. I compost my poop. And
1:02:30 y'all still think I'm a myth?
1:02:33 [Laughter]
1:02:39 I eat organic. I compost my poop. And
1:02:42 y'all still think I'm a myth?
1:02:46 It's the best. It really is good.
1:02:53 Oh, man. Man alive. All right, there's
1:02:56 Sam Alman's blog. We're going to close
1:02:58 that tab. I'm gonna while we're sitting
1:03:00 here, I'm going to close some tabs.
1:03:03 Oh, this was cool. This was the Gemini
1:03:04 thing I did.
1:03:07 Brandon's gone. We did some Gemini.
1:03:11 Oh, that's is that Manis or that? Oh,
1:03:13 that's Gamma. This is Gamma. This is me
1:03:16 trying to make a presentation in gamma.
1:03:17 That didn't go well.
1:03:20 Let's see. What's that? That's gamma. Is
1:03:23 this
1:03:24 What's this? This is Jen Spark's version
1:03:27 of it.
1:03:34 That was the Manis one I opened in
1:03:36 Google Slides. Here's Gemini.
1:03:40 So, this thing,
1:03:44 somebody made a prompt.
1:03:47 Make me a beautiful particle-based
1:03:50 endless 3D symmetrical fractal art piece
1:03:53 inspired by math. Let me I'm going to
1:03:55 pop this into the YouTube comments. So,
1:03:58 where you want to go for this is go to
1:04:02 um gemini.google.com google.com
1:04:06 and then uh just just paste in that
1:04:09 prompt and it'll start writing you code.
1:04:12 So, I'm going to say um there's
1:04:16 no color.
1:04:20 Um
1:04:22 the particles are too large
1:04:27 and it doesn't feel magical.
1:04:32 Let's see if Let's see if this recodes
1:04:34 it to be more magical. A little magical.
1:04:38 Little magical.
1:04:40 I had to get a written note
1:04:43 to even be in here in listenonly mode.
1:04:46 Brandon. Brandon. Brandon on thin ice
1:04:49 with the misses on his vacation. I just
1:04:53 got to check in on the learning lab. See
1:04:54 if everyone's okay.
1:04:59 That's awesome. Okay. Is it What's this
1:05:00 doing? Is this is this writing?
1:05:06 Did it write it? Preview.
1:05:12 Let's see.
1:05:15 Um,
1:05:18 previous version. Reload. Refresh.
1:05:23 Oh, that's cool.
1:05:37 It sort of looks like a a a realistic
1:05:40 COVID
1:05:42 C co simulator.
1:05:45 Let's go dock on a protein and make
1:05:47 these people sick.
1:05:52 Let's see. Um,
1:05:55 how would you suggest
1:05:59 turning this into a game
1:06:05 that is fun and instantly
1:06:10 understandable?
1:06:14 I don't read manuals.
1:06:20 Uh, uh,
1:06:26 What was wait was the W was the original
1:06:29 Bigfoot prompt put somewhere? It was
1:06:31 not, but I can go put it somewhere. I'll
1:06:32 go put it in irregulars.
1:06:37 So, I'll give you the whole
1:06:41 I'll give you the original prompt that I
1:06:43 wrote the prompt with. Write me a Write
1:06:46 me a video and acting prompt for a sense
1:06:50 clip
1:06:51 of Bigfoot vlogging. Make it hilarious.
1:06:56 Well, let me copy the main prompt. Then
1:06:59 I'll go to AI Learning Lab. Then I'm
1:07:02 going to put in the prompt.
1:07:11 Uh, and then I'll go back to chat GPT.
1:07:13 and I'll get my original prompt, which
1:07:15 doesn't even make sense, but that's okay
1:07:17 cuz chatbt understands stuff. Even if
1:07:20 you're an idiot,
1:07:23 I'll go here is the
1:07:27 big foot prompt
1:07:30 and then the prompt
1:07:33 that chat GPT made.
1:07:38 You can make money with chat.
1:07:44 All right, there's that. And then we're
1:07:46 going to put we're going to upload the
1:07:49 video. And this is going to fail because
1:07:54 apparently the people that
1:07:58 figured out how to
1:08:00 upload files to X also wrote the upload
1:08:03 algorithm for Mighty Networks because it
1:08:06 sucks. Do not notify all members of my
1:08:10 Bigfoot video.
1:08:15 All right, we'll see if that uploads.
1:08:17 Good night, guys. Side hustle Mimi. Got
1:08:19 to get my feral boys to bed. All right,
1:08:21 later. Side hustle mimi. So, listen.
1:08:23 Tomorrow, tomorrow's Friday. I do AI
1:08:27 office hours on LinkedIn at 11:00 a.m.
1:08:30 So, 11 a.m. Mountain time. So, if you go
1:08:34 to my LinkedIn profile, Kyle Shannon, um
1:08:37 you can go to any of my previous events.
1:08:39 They're almost all um uh AI office
1:08:42 hours, although I think these live ones
1:08:45 are mixed in there now. So, it might be
1:08:46 just find one of the office hours
1:08:48 things. The the URL for the Meetup link
1:08:51 is always the same. So, just go grab
1:08:53 that and then I'll do a post in the
1:08:54 morning. Meta AI does video now. Manis
1:08:57 does playbooks. They feel like custom
1:09:00 GPTs. And then and then yeah um
1:09:04 projects within chat GPT just got a big
1:09:06 upgrade today as well. So what happened?
1:09:10 Did this did this upload? Oh no, it's
1:09:12 still uploading.
1:09:15 I saw it flash like there was a preview
1:09:17 there for a second.
1:09:20 Come on. You can do it. You can do it,
1:09:23 big boy.
1:09:31 [Music]
1:09:40 While we wait,
1:09:42 I eat organic. I compost my poop. And
1:09:46 y'all still think I'm a myth?
1:09:48 [Laughter]
1:09:56 I don't know why I find it funny. It
1:09:59 It's got good timing. It's actually
1:10:01 really good timing.
1:10:03 All right, that does look like it hung.
1:10:11 Seriously, let me reload this and see if
1:10:13 it just trashes my post.
1:10:18 Yep, it did. God damn it.
1:10:26 Post for Bigfoot video.
1:10:32 Make it hilarious. Period.
1:10:35 Here's
1:10:39 what I
1:10:41 pasted into
1:10:45 flow.g Google
1:10:47 and then VO3
1:10:52 and then let me go get the prompt again
1:10:54 because everything sucks and nothing
1:10:57 works. And I'm the one that told you to
1:10:59 embrace the jank. But that doesn't mean
1:11:00 I have to embrace the jank because
1:11:02 sometimes it's so janky that it just
1:11:05 makes you [ __ ] mad.
1:11:08 And then come back here and then go back
1:11:10 to the AI learning lab and then post
1:11:12 that in there. That's going to be super
1:11:14 cool. You're going to be like, "Ah,
1:11:15 that's so cool. Kyle did that." Hey,
1:11:17 could I see the video? I'm trying to
1:11:19 upload the video, people. You think it's
1:11:21 easy.
1:11:25 All right.
1:11:27 25%. Oh, something went wrong. Try
1:11:29 again. There we go. That's That's the
1:11:33 That's the X algorithms video upload
1:11:36 thing. Oh, 100% ready to post.
1:11:41 All right,
1:11:43 it looks like it actually uploaded.
1:11:46 Don't notify anyone.
1:11:52 Oh, see, now it's actually uploading. Or
1:11:54 maybe that's the processing. So, tonight
1:11:57 it's give the jank aggressive bear hug.
1:12:00 Yeah. Yeah. I I think I I think I was
1:12:03 inspired by Champ having his another
1:12:06 dog's fangs around his neck. That That's
1:12:08 kind of how I feel with the jank
1:12:10 tonight. I'm like
1:12:13 [Laughter]
1:12:17 and here's Harry.
1:12:22 Joy pretty. It's really good, isn't it?
1:12:25 Joy, I assume you're playing with V3,
1:12:27 right? Your video is done processing.
1:12:29 Yay. Watch now.
1:12:33 But it's not there. I don't see it
1:12:35 anywhere.
1:12:37 Oh,
1:12:44 all right. It works.
1:12:47 All right. Beautiful. Done. It's in
1:12:51 there. At some point, it'll show up for
1:12:52 everyone else.
1:12:55 All right.
1:12:57 [Music]
1:12:59 Let's go look at chat GPT's new project
1:13:01 thing. So if I go to
1:13:05 Sydney,
1:13:08 so I can now So a couple of new things.
1:13:14 So here's all the stuff that's in
1:13:16 Sydney. These are all my Sydney chats.
1:13:20 There's my there's my custom
1:13:24 instructions.
1:13:25 I uploaded one file which is the script
1:13:27 itself.
1:13:30 You can now run deep research in a
1:13:33 project.
1:13:36 You can create images, but you can also
1:13:38 put it in conversation mode.
1:13:41 Hey Quinn, let's talk about Sydney. Um,
1:13:44 what's how does the how does the show
1:13:46 open?
1:13:47 Oh, wait, hang on.
1:13:52 See, just because Brandon's in here now,
1:13:55 I'm all like hyper paranoid about are
1:13:56 you sharing the right tab? Did you do
1:13:58 your black bar?
1:14:00 [Laughter]
1:14:04 You know what I mean?
1:14:06 And I hope the trip's going well,
1:14:08 Brandon. All right, let's let's go back
1:14:10 in here. Hey, Quinn. What's happening,
1:14:13 man?
1:14:15 Hey there. I'm doing great. Just hanging
1:14:17 out and ready to chat. How's everything
1:14:19 going on your end? Anything exciting
1:14:21 happening? No, it's good. Um, tell me
1:14:23 how the show opens. The the Sydney
1:14:29 All right. So, the musical opens with a
1:14:33 kind of bustling newsroom scene. There's
1:14:35 a lot of energy and we meet Kellen, the
1:14:38 tech reporter, who's about to be
1:14:40 assigned to investigate this new AI
1:14:42 chatbot called Sydney. So, it's got that
1:14:45 kind of high energy techy vibe right
1:14:48 from the get-go. But, do don't we meet
1:14:50 some other people before we meet Kellen?
1:14:54 Oh, absolutely. We do get to meet some
1:14:57 of Kellen's colleagues in the newsroom.
1:14:59 First, there's a bit Go read the script.
1:15:02 Go read the script and tell me how the
1:15:03 show opens.
1:15:10 In the opening scene, Kellen and her
1:15:12 colleagues in the newsroom engage in a
1:15:14 lively chorus, setting the energetic
1:15:16 tone of the show. This ensemble number
1:15:18 introduces the
1:15:21 Quinn, you're just making this up.
1:15:26 Got it. I'll make sure to stick to what
1:15:28 we have in the script. Um, let me know
1:15:30 if there's anything specific you'd like
1:15:32 me to reference or if you We should be
1:15:35 in a project right now and you should
1:15:37 have an uploaded script as part of your
1:15:40 um part of your history and you should
1:15:42 be able to access the memory of all the
1:15:45 different versions of the script that
1:15:47 are in there that we talked about using
1:15:50 chat GBT.
1:15:52 are am I talking to a project right now
1:15:54 or I am I talking to generic Quinn?
1:15:58 You're definitely talking to me within
1:16:00 the project context and I do have access
1:16:02 to all the history we've built up. Um
1:16:05 let me take a moment to check that
1:16:07 script version and I'll give you the
1:16:09 exact details of how the show opens.
1:16:15 Joy pretty talked to to Moshi today.
1:16:24 Oh, well, I hope Moshi's doing great.
1:16:26 And don't worry, I've got you covered.
1:16:28 Um, I'll make sure to go through the
1:16:29 script details carefully so we can get
1:16:31 that opening scene spot on. Did you Did
1:16:34 you read it?
1:16:40 That says she's doing search. That That
1:16:43 shouldn't work.
1:16:45 I appreciate the heads up. I'll make
1:16:48 sure to focus on what we have in memory
1:16:50 and what we've discussed before. If
1:16:52 there's anything specific you want to
1:16:54 dive into, just let me know. Yeah, I I
1:16:56 want to dive into the opening of act
1:16:59 one. How does it open?
1:17:04 So, in the opening of act one, we start
1:17:06 with the newsroom buzzing with activity.
1:17:10 No, we start with microte. We don't
1:17:12 start in the newsroom. And Kellen's not
1:17:14 a woman. It's a man. All right. This
1:17:16 doesn't work.
1:17:22 That was a That was a huge waste of
1:17:23 time. So, wait. But wait, am I in the
1:17:29 I'm not in the project anymore.
1:17:33 So, now I'm in the project or am I?
1:17:36 What's up, chat?
1:17:57 How does act
1:18:00 one
1:18:04 open?
1:18:14 Okay, baby. Do your stuff. Too good to
1:18:16 be true. Jason,
1:18:21 that's how it opens. That's correct.
1:18:29 So, how does how does act one open now?
1:18:34 All right. So, act one now opens with
1:18:36 Kellen, the tech reporter, actually on a
1:18:39 video call with his editor. They're
1:18:41 chatting about his next assignment, and
1:18:43 he's not particularly thrilled about it.
1:18:45 It sets a more laid-back, modern vibe
1:18:48 right from the start. Then we transition
1:18:50 into the world of microte. No, that it's
1:18:54 possible that I don't have
1:19:01 it says voice chat is within this chat,
1:19:03 right? So, it's got the context of it
1:19:06 just wrote
1:19:08 the correct way act one opens and then I
1:19:11 pushed her little [ __ ] talk button
1:19:13 and and uh she got it wrong again. So, I
1:19:18 don't [ __ ] know. All right. Well,
1:19:20 whatever.
1:19:22 You can do more [ __ ] in those things.
1:19:25 Let me show you another thing that they
1:19:26 added today.
1:19:43 creepy people.
1:19:45 Feature prompt book outline.
1:19:48 All right. Somewhere in here.
1:19:51 Oh, maybe not. Let's see. Do I have a
1:19:54 canvas?
1:19:57 Yes, here's a canvas. So, if I pop open
1:19:59 a canvas. So, a canvas, if you don't
1:20:01 know, is an editable document within
1:20:04 chat GPT.
1:20:07 If you're like working on something and
1:20:09 having it write it write you an outline
1:20:10 for a book, for example,
1:20:13 it'll do that. They just added you can
1:20:16 now download
1:20:18 a PDF of this, a word document of this,
1:20:22 or markdown of this. So, I'm going to
1:20:24 download a word doc.
1:20:26 We're going to go look at it in the
1:20:28 finder.
1:20:30 And we're going to go bang. Oh, wait.
1:20:33 Are you not seeing that? You're not
1:20:35 because I'm sharing the wrong flipping
1:20:39 tab.
1:20:46 All right. So, there's our there's our
1:20:48 Word document with formatting. Nice. I
1:20:52 mean, it's not great, but it's fine. Let
1:20:54 me do a PDF. Let's grab the PDF and see
1:20:56 what that looks like.
1:21:00 That actually looks much better.
1:21:06 All right.
1:21:09 There you go, people.
1:21:11 You're not in the projects. You're in
1:21:13 Queens. That was pretty good. That's
1:21:15 solid. That was nearly like comedy. So,
1:21:18 was she making it up about the script?
1:21:21 Well, she knew that the main character's
1:21:24 name was Kellen. She knew that Kellen
1:21:26 worked at a newspaper, so that part was
1:21:28 correct. The part she made up was he's
1:21:31 not video conferencing with his editor.
1:21:35 The play opens with the lead tech guy
1:21:37 from Microte discovering that he made,
1:21:42 you know, what is effectively Chad GPT
1:21:44 work.
1:21:47 Um, and there's a whole big song called
1:21:49 Too Good to Be True.
1:21:52 And then after too good to be true is
1:21:55 when we meet Kellen. Not even that
1:21:58 after. So we we we get too good to be
1:22:00 true which is this big opening number
1:22:02 here. Here I'll we'll do the music.
1:22:05 We'll do the music part of this.
1:22:10 [Music]
1:22:16 So it opens with too good to be true.
1:22:21 [Music]
1:22:28 Too often we come close. So close to a
1:22:31 breakthrough. Standing on hope and it
1:22:35 all falls apart. Just follow the dream.
1:22:39 Wherever it takes you and pray you find
1:22:42 the pulse of a brand new heart.
1:22:46 If the promise of a new dawn break,
1:22:51 a leap of faith we're all taking.
1:22:55 Is it too good to be true? We're asking
1:23:00 in its glow we find ourselves basking.
1:23:04 Too good to be true. Or it seems.
1:23:13 All right. So, we get that something's
1:23:15 going on and then Kellen goes over to
1:23:18 Varrage's office. Varrage is the CEO of
1:23:21 Microte
1:23:22 and he says, "We need six months to te
1:23:26 test it." And Vrage says to him, "You
1:23:28 got seven days." And then we go into
1:23:31 [Music]
1:23:39 seven days ago. The clock is ticking.
1:23:41 Ticking.
1:23:42 pressure building up. Keep it spike.
1:23:49 What's the word? So that's that's um
1:23:53 Jason, the lead tech guy, and his tech
1:23:55 team trying to get this testing done in
1:23:58 seven days. And then after that is when
1:24:01 Kellen meets with his editor. And then
1:24:03 there's a a Kellen version of this song.
1:24:11 Running out of time. The clock is
1:24:13 ticking.
1:24:16 The story's clicking. So anyway, so it's
1:24:19 not until the third song that we meet
1:24:21 Kellen in the editor. It's not how the
1:24:23 show opens. It's not even [ __ ] close.
1:24:26 So Quinn can bite my butt tonight.
1:24:30 Ray Edwards, what's up, Kyle? What's
1:24:32 happening? Ray, what's shaking? What's
1:24:34 going down?
1:24:36 Oh, what would be handy? would be an
1:24:38 agent that can handle tabs and cues
1:24:41 across the live stream. It actually
1:24:43 would. It'd be really good. All right,
1:24:44 let me let me get this out of here.
1:24:48 Stop screen. Start sharing. Start
1:24:50 sharing. Do this. Do this. Do this. Um,
1:24:54 I thought there might have been some
1:24:56 video I wanted to watch today. Oh, you
1:24:59 know what would be a good video for us
1:25:00 to all watch together. Actually, now
1:25:02 that I think about it,
1:25:05 David Shapiro did an 8-minute version of
1:25:09 um post labor economics.
1:25:13 Post labor economics.
1:25:19 Good day to you, sir. Good day,
1:25:24 [Music]
1:25:28 Bob. So, we're going to do some actual
1:25:30 education now. Okay, kids. It's video
1:25:33 time. So, a couple of things here. We're
1:25:35 going to dim the lights and I know you
1:25:37 might get a little sleepy. We just had
1:25:39 lunch
1:25:40 and I know you had your PB&Js. Boy,
1:25:42 those are tasty. I love it when they
1:25:44 have the strawberry. You know, the grape
1:25:46 the grape jelly I'm not such a fan of,
1:25:47 but that strawberry jam. M tasty. Uh,
1:25:50 listen, you might get a little sleepy. I
1:25:54 just I don't want you to to fall asleep.
1:25:56 I This is important. We're going to
1:25:58 watch a video together. will dry, but
1:26:01 there's pictures in it. So, I want you
1:26:02 to pay attention because none of you are
1:26:05 gonna have jobs in two years. Okay? All
1:26:07 right. Fantastic.
1:26:09 Um,
1:26:11 I wish that were a joke.
1:26:15 [Laughter]
1:26:19 It's not going to be two years. It's
1:26:21 going to be three.
1:26:27 Jason the tech guy. I get it. I see what
1:26:30 you did there. Um, yes, th this should
1:26:33 be interesting to discuss. Well,
1:26:34 technically she was right. You didn't
1:26:36 restrict her to only the first act.
1:26:38 Yeah, that's true. I said the I said,
1:26:39 "How does this first act open?" Well,
1:26:41 that's true. She But she didn't get it
1:26:43 right. He He's not having a video call
1:26:45 with his editor. It's an in-person
1:26:47 meeting, so Quinn just has her head up
1:26:50 her butt. Okay, I got to change my tabs.
1:26:56 Dank gank dank
1:27:01 [Music]
1:27:03 gank.
1:27:05 We'll go full screen with old Mr.
1:27:07 Shapiro here. If you don't follow David
1:27:09 Shapiro, I I would strongly encourage
1:27:11 it. This is a guy that's written five
1:27:13 books on cognitive architectures and AI
1:27:16 and machine learning and had a bunch of
1:27:19 open- source projects. And then when
1:27:21 when the reasoning models came out from
1:27:23 open AI, he quit AI
1:27:26 because he's like I can't there's
1:27:28 nothing more for me to contribute that
1:27:30 like basically they've cracked it. And
1:27:32 then he came back and what he's come
1:27:33 back to talk about is okay how are we
1:27:36 going to deal with you know we now know
1:27:39 that AI is going to get really good.
1:27:40 We're going to have AGI at some point.
1:27:42 We're going to have artificial super
1:27:43 intelligence and the machines will be
1:27:46 smarter than all of us.
1:27:48 He's putting his time and energy into
1:27:51 understanding what we do next. So, this
1:27:53 is there's a four-part series. I think
1:27:55 each video is like
1:27:59 20, 30, 40 minutes. They're pretty long.
1:28:02 So, he took all four of those videos and
1:28:04 condensed them or basically did this new
1:28:06 sort of five minute version of of of
1:28:09 post labor economics. Okay, so that's
1:28:11 what we're going to be watching. So,
1:28:12 take notes.
1:28:16 Let's cover post labor economics in five
1:28:19 minutes or so. Let's dive right in. So,
1:28:21 first and foremost, post labor economics
1:28:24 is what we sometimes call the great
1:28:26 decoupling. Now, let me just read it to
1:28:28 you real quick. Post labor economics
1:28:29 acknowledges the irreversible decoupling
1:28:31 of GDP growth from wage employment and
1:28:34 builds institutions that convert the
1:28:36 resulting productivity surplus into
1:28:38 broad property-based income streams
1:28:40 thereby freeing people from involuntary
1:28:42 tedious labor while safeguarding shared
1:28:44 prosperity. Now that is a lot of words
1:28:47 to basically say automation is going to
1:28:49 take all of our jobs. Moving on the
1:28:52 primary mechanism that we're looking at
1:28:55 here is what's called labor
1:28:56 substitution. And basically labor
1:28:57 substitution means that works goes from
1:29:00 humans to machines when the machines are
1:29:03 better, faster, cheaper, and safer. This
1:29:05 has been historically true for all of
1:29:07 human history and continues to be true
1:29:09 and is only accelerating with artificial
1:29:11 intelligence and robotics, which are
1:29:13 just the next wave of automation.
1:29:16 Automation is nothing new. It's been
1:29:17 around for literally centuries. It's
1:29:19 only become more and more sophisticated.
1:29:22 And by the way, as automation has become
1:29:24 more sophisticated, more labor
1:29:26 substitution has occurred.
1:29:28 Now, this leads to what we call the
1:29:29 economic agency paradox, which is best
1:29:32 summarized in this meme that I found on
1:29:34 Reddit. Uh, so step one, we replace 90%
1:29:37 of our workforce with AI. Our operating
1:29:39 costs are now the lowest they've ever
1:29:40 been, and then every other company does
1:29:42 the same thing, and no, no one is buying
1:29:44 our products since they're all jobless.
1:29:46 So that's kind of the automation that
1:29:48 sorry the economic agency paradox in a
1:29:51 nutshell. Next is aggregate demand or
1:29:53 household income. So if everyone loses
1:29:56 their job, you have to look at where
1:29:58 does income come from. First there's
1:30:00 wages, then there's property and then
1:30:02 there's transfers. Right now 60 to 80%
1:30:05 of income nationally on average comes
1:30:07 from wages, but that's declining slowly.
1:30:10 The rest comes from property which is
1:30:12 stocks, bonds, rental properties, those
1:30:14 sorts of things, real estate and then
1:30:16 the and then uh also trans. Um Rick Rick
1:30:20 Olsen, what is this? This is uh the
1:30:22 latest video from David Shapiro on pace
1:30:25 post labor economics. This is his little
1:30:27 mini eight minute version of a four-part
1:30:32 uh much longer series that really digs
1:30:34 into this. This is the the sort of eight
1:30:36 minute recap transfer. So this is the
1:30:39 the ratio nationally in America
1:30:41 basically post labor economics. This is
1:30:43 his theory about how we move to a
1:30:46 society where if AI is going to do all
1:30:49 of our jobs then how do we all make
1:30:51 money and that's that's what this is all
1:30:53 about is about 60% 20% and 20%. And
1:30:57 transfers include things like Medicare,
1:30:59 Social Security, SNAP and those uh those
1:31:01 sorts of things. Basically stuff that is
1:31:03 paid for directly from taxes. Uh now if
1:31:07 then we're losing wages then we need to
1:31:10 increase the amount of income we have
1:31:12 coming from property and transfers. Now
1:31:14 if you're entirely dependent upon
1:31:16 transfers that means you're entirely
1:31:17 dependent upon uh the government which
1:31:19 means that you're a welfare state or a
1:31:21 client state which is not good because
1:31:22 then all of your eggs are in one basket
1:31:24 and you don't have any control over your
1:31:26 future. And by the way, if you know the
1:31:29 other party gets elected next time and
1:31:31 they say, "We're going to cut your tra
1:31:33 you're we're going to cut your uh UBI or
1:31:35 whatever," then you're up the creek
1:31:36 without a paddle. So the one of the
1:31:38 keystone principles of post labor
1:31:40 economics is that we need a distributed
1:31:42 property-based future. That means uh
1:31:45 property and dividends. So moving on,
1:31:48 when we talk about a property-based
1:31:49 income stream, we're talking about uh
1:31:51 several different sources. So number
1:31:53 one, we do want some UBI to provide a
1:31:56 floor. Um so that is going to be
1:31:58 government uh government based uh tax
1:32:00 based uh UBI universal basic income uh
1:32:05 you know distributions. Next is going to
1:32:07 be wealth funds. So wealth funds include
1:32:09 sovereign wealth funds at the national
1:32:11 level but also urban wealth funds um and
1:32:14 community investment trusts and those
1:32:16 sorts of things. So these are often run
1:32:18 either by the government or by public
1:32:21 private partnerships. Um think of them
1:32:23 like endowments. So you'll create
1:32:25 endowment funds that basically just by
1:32:27 virtue of being a citizen of a
1:32:29 particular region, you get a check in
1:32:31 the mail every month or every quarter or
1:32:33 every year. The next level above that is
1:32:36 going to be private collective property.
1:32:38 So this is stuff that you own in common
1:32:40 um either through credit unions or Dows
1:32:43 or those sorts of things um which are
1:32:45 also likewise going to be paying rent.
1:32:47 So when you say like well what do you
1:32:48 mean? We could mean data centers, we
1:32:50 could mean robots, we could mean any
1:32:52 kinds of resource in so in in the longer
1:32:54 in the longer videos he talks about
1:32:56 these these things the private
1:32:58 collective property and the wealth
1:32:59 funds.
1:33:01 He said you you'll be able to choose
1:33:04 sort of where you want to live based on
1:33:07 the resources they have. So like you
1:33:09 might you might live somewhere down in
1:33:11 Arizona and they've got lots and lots of
1:33:13 sun and the the county puts in lots and
1:33:17 lots of solar farms, right? And then the
1:33:19 AI needs the energy from the solar farms
1:33:22 and that generates revenue for for the
1:33:25 city for the wealth fund, right? So, or
1:33:28 you might go to, you know, some east
1:33:30 coast city that's on the water and the
1:33:32 big thing there is fishing and tourism
1:33:35 and so that's that's how it funds it. So
1:33:37 basically these wealth funds and private
1:33:39 collective property things are
1:33:41 essentially local um revenue generation
1:33:46 based on the resources of the region
1:33:50 versus solar farms, fusion reactors,
1:33:53 quantum computers, any kind of property
1:33:55 that can be owned and instead of buying
1:33:58 it individually or buying shares, you
1:34:00 put your money and your resources
1:34:01 together and you own it collectively.
1:34:03 Next is private wealth. So private
1:34:05 wealth is basically what you have today,
1:34:08 stocks, bonds, shares, companies, those
1:34:10 sorts of things, real estate, land, none
1:34:13 of that really changes. And then
1:34:14 finally, the last uh source of revenue
1:34:17 is going to be uh residual wages. So
1:34:20 basically, we're kind of right now
1:34:21 anticipating that about 20% of wages
1:34:24 might might stick around. Um time will
1:34:26 tell. It could be more, could be less,
1:34:28 but right now we're trending towards
1:34:29 that direction. 20%. Did you did you
1:34:34 take that in? 20% of wages will re will
1:34:37 remain.
1:34:39 Next, there are four pillars of civic
1:34:42 society. So when you think about, you
1:34:44 know, what is the social contract? The
1:34:46 social contract is generally between the
1:34:48 governed and the governors. So the the
1:34:51 people and the state. However, uh the
1:34:53 social equilibrium today is wait, you
1:34:57 like this guy's AI content, but he's way
1:34:59 overboard on on UBI. He's the only one
1:35:02 that is talking about something other
1:35:04 than UBI.
1:35:06 UBI is just one of his components here.
1:35:11 Estimating based on what? He's really
1:35:13 trying to be an economist, huh? Yeah. I
1:35:16 don't know, Pate. He's I'm sure it's
1:35:19 based on it's based on um Chad GPT, you
1:35:23 know, doing analysis and deep research.
1:35:26 So,
1:35:28 do you follow him on X? Yeah, I do. Oh,
1:35:32 he really goes deep on UBI. Well, no, he
1:35:34 he has gone deep on UBI, but I I don't
1:35:36 know if you've watched all four of these
1:35:38 um uh the the uh the post labor
1:35:42 economics videos, but there's a there's
1:35:44 a lot more there than UBI now,
1:35:46 collectively. Next is private wealth.
1:35:48 So, private wealth is basically what you
1:35:51 have today, stocks, bonds, shares,
1:35:53 companies, those sorts of things, real
1:35:55 estate, land, none of that really
1:35:57 changes. And then finally the last uh
1:36:00 source of revenue is going to be we're
1:36:02 trending towards that direction. Next
1:36:05 there are four pillars of civic society.
1:36:07 So when you think about you know what is
1:36:09 the social contract the social contract
1:36:11 is generally between the governed and
1:36:14 the governors. So the the people and the
1:36:16 state. However uh the social equilibrium
1:36:19 today is maintained by four primary
1:36:22 stakeholders which is we the people. So
1:36:24 civilians, citizens, the state, which is
1:36:27 the government, which ostensibly is
1:36:28 built for and by us and should serve us.
1:36:31 But more and more states are becoming
1:36:32 less about the people and more about
1:36:34 serving businesses and banks. Now, we're
1:36:37 not going to build a society that gets
1:36:39 rid of businesses and banks anytime
1:36:40 soon. It's possible in the long run, but
1:36:43 let's not get ahead of ourselves. So,
1:36:45 what we really need is to rebalance uh
1:36:48 the p the balance of power that happens
1:36:50 here if we lose wage power and labor
1:36:52 power. So when we talk about economic
1:36:55 agency, there are three primary pillars
1:36:57 of power that we have. Number one above
1:37:01 all is labor rights. Labor basically is
1:37:04 the one thing that we have intrinsic
1:37:05 control over until machines take away
1:37:09 our ability to work and demand uh money
1:37:12 for that labor. because the ability to
1:37:16 withhold labor is is a one of the
1:37:18 fundamental levers of power that we have
1:37:20 which then guarantees property rights
1:37:22 and democratic rights. If we lose labor
1:37:25 rights, which we are losing, um not only
1:37:27 are we not only are labor rights eroding
1:37:30 under the neoliberal regime, they will
1:37:32 erode further um as automation, AI, and
1:37:36 robotics further encroach upon one of
1:37:38 the intrinsic levers of power that we
1:37:40 have as civilians,
1:37:43 which then means that our property
1:37:45 rights and democratic rights will also
1:37:46 erode. This is far and away the larger
1:37:49 problem other than the economy. It's
1:37:51 it's fundamentally about power. If we
1:37:54 lose power, we lose everything. So how
1:37:56 do we fix this problem? What we need is
1:37:59 we need a replacement pillar. So that
1:38:01 becomes algorithmic rights. Um in this
1:38:03 new paradigm, we replace labor rights
1:38:06 with algorithmic rights in uh in this in
1:38:09 this new paradigm which then shores up
1:38:12 property rights and democratic rights.
1:38:14 And by the way, in case it's not clear,
1:38:16 this is his proposal for how we do this,
1:38:18 right? This is his proposal to say, you
1:38:22 know, people aren't going to fix this.
1:38:24 His system is the thing he's proposing
1:38:26 here. And and
1:38:28 to everyone's point that he may not be
1:38:32 right. Absolutely, he may not be right,
1:38:34 but there are very few people I see
1:38:37 digging deep on, hey, this shit's about
1:38:40 to take a bunch of jobs. What are we
1:38:42 going to do about it? So, I I think it's
1:38:44 important just because it's at least a
1:38:46 voice with an idea. So that's why I
1:38:49 think it's important to watch data
1:38:50 sovereignty, algorithmic auditability,
1:38:52 participatory algorithmic governance and
1:38:54 algorithmic dividend and liability. This
1:38:57 is all based upon technologies like
1:38:59 blockchain, decentralized autonomous
1:39:01 organizations, cryptocurrency, central
1:39:03 banking, uh digital currencies, um
1:39:06 digital identity wallets and those sorts
1:39:08 of things. We will need we are already
1:39:10 building this infrastructure but it is
1:39:12 not yet not ready yet. Sorry. Um, but
1:39:16 with that being said, blockchain is
1:39:19 central to this future. There are some
1:39:22 of the technological affordances of
1:39:24 blockchain make it the ideal baseline
1:39:26 technology for this new social contract.
1:39:28 Number one, it's intrinsically
1:39:30 democratic. Number two, it's
1:39:32 intrinsically decentralized. Number
1:39:34 three, it's unstoppable. You can't shut
1:39:36 it down. And number four, it's
1:39:37 permissionless. You don't need the
1:39:38 government's permission to build a
1:39:40 blockchain. So, with all that being
1:39:42 said, thank you for watching. You have
1:39:43 now learned about post labor economics
1:39:45 in about five minutes. Cheers. There you
1:39:48 go. All right. So that's it. So it's
1:39:49 basically, you know, it's it's basically
1:39:53 do all the governance stuff in an
1:39:56 algorithmic way tracked on the
1:39:57 blockchain so that as these funds get
1:40:00 funded um the the the the
1:40:06 algorithms combined with blockchain are
1:40:08 how all the stuff gets divvied up and is
1:40:10 fair and and all that sort of stuff.
1:40:13 um you know, as he said, like while that
1:40:16 infrastructure is being built, like
1:40:18 blockchain and and crypto right now is
1:40:20 still an abject [ __ ] mess. So, um but
1:40:25 anyway, this was the first thing that I
1:40:28 saw that was that was more than just
1:40:30 UBI, right? It it was talking about the
1:40:33 resources of regions and acknowledging
1:40:37 the fact that we're going to have to
1:40:39 [ __ ] figure this out. We're going to
1:40:40 have to figure it out fast. So anyway,
1:40:42 it's something for us to think about,
1:40:45 start talking about. All right. H makes
1:40:49 me tired just thinking about it. I know.
1:40:50 I know. Listen, this is this is all sort
1:40:53 of chat TMZ kind of future gazing kind
1:40:56 of stuff, but
1:41:02 trying to think like when it's when it's
1:41:04 probably like what's my gut of like when
1:41:06 we're going to start to see
1:41:10 major shifts.
1:41:15 The one thing I know is that business
1:41:17 moves slow and government moves slow,
1:41:20 but AI is moving fast.
1:41:26 So, what it probably looks like is this.
1:41:28 What it probably looks like is in the
1:41:30 next 18 months,
1:41:33 the technology part of this here, let me
1:41:35 turn this thing off. I think what this
1:41:38 looks like is this. In the next 18
1:41:39 months or so,
1:41:41 the technology gets so good that it
1:41:43 becomes apparent that, oh yeah, this can
1:41:46 do our jobs or this can do some major
1:41:49 swath of jobs and then it's probably
1:41:54 another 18 months to three years
1:41:58 before businesses really adopt that.
1:42:01 Now, unless it's so good that businesses
1:42:03 are just like, well, [ __ ] it. Like if I
1:42:06 can, you know, take this 100,000 person
1:42:09 company and I can drop that down to
1:42:11 10,000 people and be more efficient and
1:42:15 produce more profit. Well, I'm going to
1:42:16 [ __ ] do it, right? Like it is
1:42:19 possible something gets that good. I
1:42:21 just doubt it, right? I think like the
1:42:24 fact that AI tools are still so bad.
1:42:28 Like the thing we just did with Quinn
1:42:29 where she she didn't know how my musical
1:42:31 [ __ ] opened and I've got the the
1:42:33 actual script is in her data set. It it
1:42:37 just makes [ __ ] up. So like that's going
1:42:40 to have to be fixed. And I don't think
1:42:41 that's like that hasn't really gotten
1:42:43 better in in the two years that that
1:42:46 we've been playing with this stuff. Um,
1:42:49 so
1:42:52 yeah, so we'll probably see something
1:42:54 that's the the foundation of the big
1:42:57 disruption within two years and then
1:43:00 it's probably another two to three after
1:43:02 that where we really start to see
1:43:05 massive impacts on employment. But I
1:43:07 don't [ __ ] know. I have a degree in
1:43:09 acting.
1:43:13 Correction, Kyle. Government moves
1:43:15 slower. Yeah, I know. Exactly. Exactly.
1:43:19 Although I mean listen when COVID
1:43:22 happened, government moved fast. Like
1:43:24 when it when it needs to, when there's a
1:43:28 crisis, it can move fast, right?
1:43:32 Um
1:43:33 if if unemployment jumps from 2% to 4 to
1:43:38 8 to 10 to 20 in a short amount of time,
1:43:42 like over 12 to 18 months, that's that's
1:43:45 catastrophic, right? and government will
1:43:47 will move fast. You know, that's that's
1:43:49 the equivalent of a pandemic. Um, if
1:43:52 it's a really slow decline where it's
1:43:54 just like two, three, four, 4.5, five,
1:43:59 back to four, back to six, seven, eight.
1:44:04 Then then they may not move fast. So, I
1:44:06 don't know. We'll see. We will say. All
1:44:10 right. Um, I'm going to get on out of
1:44:12 here. Hope you have fun tonight, dear
1:44:14 people. Tomorrow, Friday.
1:44:21 I don't Is there an art thing going on
1:44:24 tomorrow? There might be.
1:44:27 I don't think I have to deal with it. I
1:44:29 just might have to go pick up Gabby at
1:44:30 some point cuz I think she's hanging out
1:44:33 with her friend. All right, so should be
1:44:36 regular time tomorrow. Tomorrow's Friday
1:44:37 night, date night. So, come back. Let's
1:44:39 play tomorrow.
1:44:41 um 8:00 PM Mountain time and then 11:00
1:44:45 a.m. Mountain time is LinkedIn office
1:44:47 hours, AI office hours. Okay, so come at
1:44:50 hang out with that crew if if you're
1:44:52 trying to figure this [ __ ] out. AI
1:44:54 office hours is really good. All right,
1:44:56 cool. Peace out everyone. Have a good
1:44:59 night and I will see you tomorrow.