134: AI Art Will Make Marionettes Of Us All Before It Destroys The World 
   
 
 
	 00:00:00
     ◼ 
      
     ►  
     Well, we have a lot more terrifying things to discuss about AI. 
     
     
  
 
 
	 00:00:05
     ◼ 
      
     ►  
     A lot has happened since the last time we spoke, Myke. 
     
     
  
 
 
	 00:00:08
     ◼ 
      
     ►  
     I don't presume that I'm the center of the universe. 
     
     
  
 
 
	 00:00:12
     ◼ 
      
     ►  
     But there is an almost Truman Show-like aspect sometimes to life, 
     
     
  
 
 
	 00:00:17
     ◼ 
      
     ►  
     and this is one of those things where I feel like, 
     
     
  
 
 
	 00:00:20
     ◼ 
      
     ►  
     you know, I was seeing enough about A.I.R. around the place, 
     
     
  
 
 
	 00:00:23
     ◼ 
      
     ►  
     like, you know, there was a reason that we spoke about it, 
     
     
  
 
 
	 00:00:26
     ◼ 
      
     ►  
     because it was coming into our worldview more and more. 
     
     
  
 
 
 
	 00:00:29
     ◼ 
      
     ►  
     But in the time between these last two episodes, 
     
     
  
 
 
	 00:00:31
     ◼ 
      
     ►  
     I feel like it has taken over the world. 
     
     
  
 
 
	 00:00:34
     ◼ 
      
     ►  
     I can't move for people talking about it. 
     
     
  
 
 
	 00:00:37
     ◼ 
      
     ►  
     It's coming up on podcasts that have got nothing to do 
     
     
  
 
 
	 00:00:40
     ◼ 
      
     ►  
     with this stuff, like not even tech related shows 
     
     
  
 
 
	 00:00:43
     ◼ 
      
     ►  
     they're talking about. 
     
     
  
 
 
	 00:00:43
     ◼ 
      
     ►  
     Like Idina sent me a link to one of her favorite podcasts, 
     
     
  
 
 
	 00:00:46
     ◼ 
      
     ►  
     it's a podcast called Comic Lab, 
     
     
  
 
 
	 00:00:48
     ◼ 
      
     ►  
     which is about cartooning and making comics. 
     
     
  
 
 
	 00:00:52
     ◼ 
      
     ►  
     They did like a whole episode talking about it. 
     
     
  
 
 
	 00:00:54
     ◼ 
      
     ►  
     And it makes sense for them a little bit 
     
     
  
 
 
	 00:00:56
     ◼ 
      
     ►  
     because they are comic artists, right? 
     
     
  
 
 
	 00:00:57
     ◼ 
      
     ►  
     They are artists. 
     
     
  
 
 
	 00:00:58
     ◼ 
      
     ►  
     - But it's like, why now? 
     
     
  
 
 
	 00:01:00
     ◼ 
      
     ►  
     - Yeah, I think this is definitely 
     
     
  
 
 
	 00:01:03
     ◼ 
      
     ►  
     the biggest Truman Show-like effect 
     
     
  
 
 
	 00:01:05
     ◼ 
      
     ►  
     for anything that we have ever spoken about. 
     
     
  
 
 
	 00:01:08
     ◼ 
      
     ►  
     And I think it's just that we happened to hit it 
     
     
  
 
 
	 00:01:10
     ◼ 
      
     ►  
     at the exact right moment before a bunch of things exploded. 
     
     
  
 
 
	 00:01:15
     ◼ 
      
     ►  
     - Yeah, it's like realistically, I know why this happened, 
     
     
  
 
 
	 00:01:18
     ◼ 
      
     ►  
     but it's still such a strange feeling. 
     
     
  
 
 
	 00:01:20
     ◼ 
      
     ►  
     This happened because we spoke about it 
     
     
  
 
 
	 00:01:22
     ◼ 
      
     ►  
     because it was starting to become a thing 
     
     
  
 
 
	 00:01:23
     ◼ 
      
     ►  
     that we couldn't avoid, right? 
     
     
  
 
 
	 00:01:25
     ◼ 
      
     ►  
     So that's why then it's just like a snowball effect. 
     
     
  
 
 
	 00:01:28
     ◼ 
      
     ►  
     Oh, I know, I know. And like listeners may remember that I had a conference that I said 
     
     
  
 
 
	 00:01:33
     ◼ 
      
     ►  
     I was going to go to in September. I was like, "Oh, guess what?" One of the major topics 
     
     
  
 
 
	 00:01:38
     ◼ 
      
     ►  
     like everyone was discussing at the conference. It was a bunch of AI stuff. Like my whole 
     
     
  
 
 
	 00:01:42
     ◼ 
      
     ►  
     life just exploded into this in the past month. And I think also with the last episode of 
     
     
  
 
 
	 00:01:48
     ◼ 
      
     ►  
     the show, there was like a little bit of confusion over our positions because there's just 
     
     
  
 
 
	 00:01:53
     ◼ 
      
     ►  
     like there's so much that this touches on, right? Like this, we just kind of talked about 
     
     
  
 
 
	 00:01:57
     ◼ 
      
     ►  
     it off the cuff. And, you know, I really think this this absolutely touches on almost everything 
     
     
  
 
 
	 00:02:04
     ◼ 
      
     ►  
     in the world. So I think to try to be clear, it's like, with AI arts, and the language 
     
     
  
 
 
	 00:02:10
     ◼ 
      
     ►  
     models in particular, in the short term, I think there's a lot of interesting discussions 
     
     
  
 
 
	 00:02:14
     ◼ 
      
     ►  
     to be had about how it affects technology and how it affects the economy and how it 
     
     
  
 
 
	 00:02:19
     ◼ 
      
     ►  
     affects ethics. But in the long term, we're seeing the beginning of the end of the human 
     
     
  
 
 
 
 
 
	 00:02:29
     ◼ 
      
     ►  
     Is this the clarifying you were looking to do? 
     
     
  
 
 
	 00:02:32
     ◼ 
      
     ►  
     Yeah, just so people can understand where I'm coming from. 
     
     
  
 
 
	 00:02:35
     ◼ 
      
     ►  
     There's lots of interesting discussions to be had where you can take many different 
     
     
  
 
 
	 00:02:39
     ◼ 
      
     ►  
     positions, but ultimately I think this all points in one direction, which is not very 
     
     
  
 
 
 
	 00:02:45
     ◼ 
      
     ►  
     But yes, in the short term, since the last time we spoke, so much has happened, and I 
     
     
  
 
 
	 00:02:52
     ◼ 
      
     ►  
     I think one of the first things that you sent to me was that Dolly is now open for the public 
     
     
  
 
 
 
 
	 00:02:59
     ◼ 
      
     ►  
     And the thing that I think really helped kick it off was, because last episode we were talking 
     
     
  
 
 
	 00:03:04
     ◼ 
      
     ►  
     about how you, oh, you need these like massive computing clusters to do this kind of stuff. 
     
     
  
 
 
	 00:03:08
     ◼ 
      
     ►  
     I don't know. 
     
     
  
 
 
	 00:03:09
     ◼ 
      
     ►  
     It was like 10 days after that episode went up, the very first desktop versions of these 
     
     
  
 
 
	 00:03:15
     ◼ 
      
     ►  
     things came out. 
     
     
  
 
 
	 00:03:17
     ◼ 
      
     ►  
     And the one I've been playing with is called Diffusion B, which allows you just on your 
     
     
  
 
 
	 00:03:23
     ◼ 
      
     ►  
     Mac as long as it's one of the new M processors so that it's fast enough, it allows you to 
     
     
  
 
 
	 00:03:28
     ◼ 
      
     ►  
     start typing in whatever and generating AI art. 
     
     
  
 
 
	 00:03:32
     ◼ 
      
     ►  
     And so it's like, oh, okay, this went from supercomputing clusters only down to, oh, 
     
     
  
 
 
	 00:03:40
     ◼ 
      
     ►  
     it can happen on your desktop, which is I think the first time that lots of people actually 
     
     
  
 
 
	 00:03:45
     ◼ 
      
     ►  
     started to play with it firsthand and I think that's kind of what kicked off for lots 
     
     
  
 
 
 
	 00:03:52
     ◼ 
      
     ►  
     "Oh this is really real, like I can play with it and I can see what it does." 
     
     
  
 
 
	 00:03:58
     ◼ 
      
     ►  
     The first publicly used by us piece of AI art went up which was running our Cortex logo 
     
     
  
 
 
	 00:04:05
     ◼ 
      
     ►  
     through some AI art stuff. 
     
     
  
 
 
	 00:04:08
     ◼ 
      
     ►  
     - How did you do that? 
     
     
  
 
 
	 00:04:09
     ◼ 
      
     ►  
     - That was done with, oh god what was it called, it was done with Dream I think is one of the 
     
     
  
 
 
	 00:04:13
     ◼ 
      
     ►  
     projects but you can upload a piece of art and have it like iterate on that piece of art. 
     
     
  
 
 
	 00:04:20
     ◼ 
      
     ►  
     - Oh, okay. That makes sense then. Because I wonder, it's like, how did you get it to do the Cortex logo? 
     
     
  
 
 
	 00:04:24
     ◼ 
      
     ►  
     - This is also one of these things like, even Diffusion Bee, which is the thing that you can 
     
     
  
 
 
	 00:04:28
     ◼ 
      
     ►  
     play with on your laptop, it started out as, oh you can only just put in words and then it will 
     
     
  
 
 
	 00:04:33
     ◼ 
      
     ►  
     generate images, but even that one now, you can give it an image and then ask for modifications 
     
     
  
 
 
	 00:04:38
     ◼ 
      
     ►  
     on that image, but the dream program just lets you put in a piece of art and then it 
     
     
  
 
 
	 00:04:43
     ◼ 
      
     ►  
     just iterates on it in some way. 
     
     
  
 
 
	 00:04:45
     ◼ 
      
     ►  
     Part of the intrinsically horrifying thing about all of this art is that a bunch of examples 
     
     
  
 
 
	 00:04:50
     ◼ 
      
     ►  
     had to be rejected because the AI knew that it was a brain and so ended up making it look 
     
     
  
 
 
	 00:04:57
     ◼ 
      
     ►  
     very like, "Oh, it's a horrifying mass of meat that's also a brain." 
     
     
  
 
 
	 00:05:04
     ◼ 
      
     ►  
     My favorite. 
     
     
  
 
 
	 00:05:05
     ◼ 
      
     ►  
     Which brings us to AI Art. 
     
     
  
 
 
	 00:05:07
     ◼ 
      
     ►  
     Why you so terrifying? 
     
     
  
 
 
	 00:05:09
     ◼ 
      
     ►  
     So last time we spoke, another one of these things was, "Oh, obviously this is coming 
     
     
  
 
 
 
	 00:05:15
     ◼ 
      
     ►  
     It's just a matter of time. 
     
     
  
 
 
	 00:05:16
     ◼ 
      
     ►  
     It's just a question of how long does it take computationally to do these sorts of things?" 
     
     
  
 
 
	 00:05:21
     ◼ 
      
     ►  
     And it's like, "Oh, is that going to happen next year?" 
     
     
  
 
 
	 00:05:24
     ◼ 
      
     ►  
     No, it happened two weeks after the show went up. 
     
     
  
 
 
	 00:05:28
     ◼ 
      
     ►  
     And I have something for you to take a look at, which was the first version of this, which 
     
     
  
 
 
	 00:05:34
     ◼ 
      
     ►  
     is this was made a little bit manually. 
     
     
  
 
 
	 00:05:37
     ◼ 
      
     ►  
     It was made with stable diffusion. 
     
     
  
 
 
	 00:05:39
     ◼ 
      
     ►  
     But the prompt is a journey through a boy's life. 
     
     
  
 
 
	 00:05:43
     ◼ 
      
     ►  
     Again, this is not true video, but it's like an, 
     
     
  
 
 
	 00:05:48
     ◼ 
      
     ►  
     I don't know how to describe it. 
     
     
  
 
 
	 00:05:50
     ◼ 
      
     ►  
     It's a-- - It's like, 
     
     
  
 
 
	 00:05:50
     ◼ 
      
     ►  
     I'm watching it now, it's like a flip book. 
     
     
  
 
 
	 00:05:52
     ◼ 
      
     ►  
     - Yeah, it's a bit like a flip book. 
     
     
  
 
 
	 00:05:53
     ◼ 
      
     ►  
     So just take a look at it 
     
     
  
 
 
	 00:05:54
     ◼ 
      
     ►  
     and tell me what you think about this. 
     
     
  
 
 
	 00:05:56
     ◼ 
      
     ►  
     - I mean, it's weird. 
     
     
  
 
 
	 00:05:58
     ◼ 
      
     ►  
     Like, it's not much of a video to me, really. 
     
     
  
 
 
	 00:06:01
     ◼ 
      
     ►  
     Like, I can see it's just like a bunch, 
     
     
  
 
 
	 00:06:02
     ◼ 
      
     ►  
     you know, you're just cycling through a bunch of images. 
     
     
  
 
 
	 00:06:05
     ◼ 
      
     ►  
     It's got weird, 
     
     
  
 
 
	 00:06:05
     ◼ 
      
     ►  
     I just got weirdly religious there for a minute. 
     
     
  
 
 
	 00:06:07
     ◼ 
      
     ►  
     What is happening here? 
     
     
  
 
 
	 00:06:09
     ◼ 
      
     ►  
     Like what is going on here? 
     
     
  
 
 
	 00:06:10
     ◼ 
      
     ►  
     Did an AI generate these and put them together? 
     
     
  
 
 
	 00:06:13
     ◼ 
      
     ►  
     Or like, are they generating them from each other? 
     
     
  
 
 
	 00:06:16
     ◼ 
      
     ►  
     Like what is going on? 
     
     
  
 
 
	 00:06:17
     ◼ 
      
     ►  
     How is this being made? 
     
     
  
 
 
	 00:06:18
     ◼ 
      
     ►  
     - Yeah, so I read a little bit of what the author said. 
     
     
  
 
 
	 00:06:21
     ◼ 
      
     ►  
     And the impression is, he started with a general image 
     
     
  
 
 
	 00:06:26
     ◼ 
      
     ►  
     that's called a journey through a boy's life. 
     
     
  
 
 
	 00:06:28
     ◼ 
      
     ►  
     And then each of the next frames 
     
     
  
 
 
	 00:06:31
     ◼ 
      
     ►  
     is based on the frame before it, 
     
     
  
 
 
	 00:06:34
     ◼ 
      
     ►  
     given that same piece of information. 
     
     
  
 
 
	 00:06:36
     ◼ 
      
     ►  
     It's going in some really weird places. 
     
     
  
 
 
	 00:06:37
     ◼ 
      
     ►  
     We just went to war for a while, 
     
     
  
 
 
	 00:06:40
     ◼ 
      
     ►  
     and now there's some kind of eldritch horror occurring. 
     
     
  
 
 
	 00:06:43
     ◼ 
      
     ►  
     Okay, I mean, it's really weird. 
     
     
  
 
 
	 00:06:45
     ◼ 
      
     ►  
     I watch something like that 
     
     
  
 
 
	 00:06:46
     ◼ 
      
     ►  
     and knowing where it's coming from, 
     
     
  
 
 
	 00:06:48
     ◼ 
      
     ►  
     and it doesn't make me feel great in a way, 
     
     
  
 
 
	 00:06:53
     ◼ 
      
     ►  
     because of the weird places the AI takes itself. 
     
     
  
 
 
	 00:06:59
     ◼ 
      
     ►  
     And there was an uncomfortability in that. 
     
     
  
 
 
	 00:07:02
     ◼ 
      
     ►  
     I do not know why we spent a significant portion of that 
     
     
  
 
 
	 00:07:07
     ◼ 
      
     ►  
     in various wartime. 
     
     
  
 
 
	 00:07:10
     ◼ 
      
     ►  
     And no one can really answer that question. 
     
     
  
 
 
	 00:07:14
     ◼ 
      
     ►  
     I assume, there's no way to know why the AI went that route 
     
     
  
 
 
	 00:07:17
     ◼ 
      
     ►  
     and stuck in it for quite some time. 
     
     
  
 
 
	 00:07:19
     ◼ 
      
     ►  
     - So as a description for the listeners, 
     
     
  
 
 
	 00:07:22
     ◼ 
      
     ►  
     I mean, I think it's sort of horrifying 
     
     
  
 
 
	 00:07:26
     ◼ 
      
     ►  
     and it's very dirty looking for some reason, 
     
     
  
 
 
	 00:07:29
     ◼ 
      
     ►  
     But it is like a sequence of images that start with a child and the child slowly grows up 
     
     
  
 
 
	 00:07:35
     ◼ 
      
     ►  
     and you just see a bunch of things like a kid at the desk doing his homework and it 
     
     
  
 
 
	 00:07:40
     ◼ 
      
     ►  
     sort of transitions. 
     
     
  
 
 
	 00:07:42
     ◼ 
      
     ►  
     The kid just gets increasingly older and older and then yes, it's young men at war for a 
     
     
  
 
 
	 00:07:46
     ◼ 
      
     ►  
     while and I presume this is one of these side effects of what's in the training database 
     
     
  
 
 
	 00:07:51
     ◼ 
      
     ►  
     and it has some concept of, "Oh, based on all of the images I've seen, what has a boy 
     
     
  
 
 
	 00:07:57
     ◼ 
      
     ►  
     been doing in his 20s, and so it generates like a ton of war imagery. 
     
     
  
 
 
	 00:08:03
     ◼ 
      
     ►  
     I also suspect that's why at that moment it also gets weirdly religious, like a lot of 
     
     
  
 
 
	 00:08:07
     ◼ 
      
     ►  
     crosses appear, because I presume that that's coming from graveyard stuff. 
     
     
  
 
 
	 00:08:12
     ◼ 
      
     ►  
     And it goes all the way through to ending with like a, you know, a dead body laying 
     
     
  
 
 
	 00:08:18
     ◼ 
      
     ►  
     on a table with sort of muddy blood coming down from it. 
     
     
  
 
 
	 00:08:22
     ◼ 
      
     ►  
     This was the first thing that I saw that I thought, oh, this is a video. 
     
     
  
 
 
	 00:08:27
     ◼ 
      
     ►  
     And I would also legitimately say, oh, this is a piece of art. 
     
     
  
 
 
	 00:08:31
     ◼ 
      
     ►  
     Like you could, you could display this in an art museum and it wouldn't be out of place. 
     
     
  
 
 
	 00:08:38
     ◼ 
      
     ►  
     And what it made me think of was, I don't know if you've ever seen the animations from 
     
     
  
 
 
	 00:08:42
     ◼ 
      
     ►  
     Pink Floyd's The Wall. 
     
     
  
 
 
	 00:08:44
     ◼ 
      
     ►  
     That's what this made me think of. 
     
     
  
 
 
	 00:08:45
     ◼ 
      
     ►  
     Oh, it has that kind of feeling to it. 
     
     
  
 
 
	 00:08:49
     ◼ 
      
     ►  
     And it has a really horrifying animation style. 
     
     
  
 
 
	 00:08:53
     ◼ 
      
     ►  
     And I went back and wanted to rewatch some of the animation sections of that movie. 
     
     
  
 
 
	 00:08:59
     ◼ 
      
     ►  
     And boy, it was an interesting experience, because in comparison, suddenly the walls 
     
     
  
 
 
	 00:09:06
     ◼ 
      
     ►  
     seemed remarkably undetailed. 
     
     
  
 
 
	 00:09:10
     ◼ 
      
     ►  
     And I just felt like I could only see, oh, it's so simple. 
     
     
  
 
 
	 00:09:14
     ◼ 
      
     ►  
     And like that is perhaps one of the most complicated examples of like hand drawn animation is the 
     
     
  
 
 
 
	 00:09:21
     ◼ 
      
     ►  
     Oh God compared to this. 
     
     
  
 
 
	 00:09:22
     ◼ 
      
     ►  
     Oh, yeah, they just can't possibly put the detail in every single frame that exists in 
     
     
  
 
 
	 00:09:28
     ◼ 
      
     ►  
     this one thing. 
     
     
  
 
 
	 00:09:30
     ◼ 
      
     ►  
     And so I just I just had a real feeling of, wow, what a jump. 
     
     
  
 
 
	 00:09:34
     ◼ 
      
     ►  
     It's sort of addressing the same idea, but in 100,000 times more detail. 
     
     
  
 
 
	 00:09:41
     ◼ 
      
     ►  
     So anyway, that'll be in the show notes for people to take a look at and then I don't 
     
     
  
 
 
	 00:09:45
     ◼ 
      
     ►  
     know another seven days after that meta and a few other companies announced true text 
     
     
  
 
 
	 00:09:54
     ◼ 
      
     ►  
     to video projects and I have some links in the show notes that you can click on and so 
     
     
  
 
 
	 00:09:59
     ◼ 
      
     ►  
     this is called make a video and if you if you take a look at some of these links, I 
     
     
  
 
 
	 00:10:04
     ◼ 
      
     ►  
     think this stuff looks more like what stable diffusion does. 
     
     
  
 
 
	 00:10:10
     ◼ 
      
     ►  
     - Okay, this looks like, take all of the generators 
     
     
  
 
 
	 00:10:15
     ◼ 
      
     ►  
     and make them do animation, right? 
     
     
  
 
 
	 00:10:17
     ◼ 
      
     ►  
     Like it looks, it has a similar look to, 
     
     
  
 
 
	 00:10:21
     ◼ 
      
     ►  
     like the quality of some of the imagery is like, 
     
     
  
 
 
	 00:10:25
     ◼ 
      
     ►  
     has the telltale signs that this was made 
     
     
  
 
 
	 00:10:28
     ◼ 
      
     ►  
     by an AI art generator. 
     
     
  
 
 
	 00:10:32
     ◼ 
      
     ►  
     - Yeah, I think the key characteristic 
     
     
  
 
 
	 00:10:34
     ◼ 
      
     ►  
     of a lot of this stuff is it's still quite dreamlike, 
     
     
  
 
 
	 00:10:38
     ◼ 
      
     ►  
     because a lot of the details aren't there. There's a lot of areas where it's kind of fuzzy. 
     
     
  
 
 
	 00:10:44
     ◼ 
      
     ►  
     But I think that this again is the thing where a lot of the detractors of this kind of thing said, 
     
     
  
 
 
	 00:10:50
     ◼ 
      
     ►  
     "Oh, you'll never be able to do video." And you go, "Well, yeah, let's just wait. Let's just 
     
     
  
 
 
	 00:10:57
     ◼ 
      
     ►  
     wait and see how long this takes." And we go, "Okay, here we go. Here's the first versions 
     
     
  
 
 
	 00:11:03
     ◼ 
      
     ►  
     of video where you can just type the words "a dog wearing a superhero outfit with a red cape flying 
     
     
  
 
 
	 00:11:09
     ◼ 
      
     ►  
     through the sky" and it makes that. Like it makes a little video of that thing. And there's another 
     
     
  
 
 
	 00:11:15
     ◼ 
      
     ►  
     video project which I'll try to find for the show notes for viewers later which was this, but it's 
     
     
  
 
 
	 00:11:21
     ◼ 
      
     ►  
     like a multi-scene description. So you can say things like "a woman's swimming in the ocean, 
     
     
  
 
 
	 00:11:27
     ◼ 
      
     ►  
     she dives under the water and picks up a starfish on the ground and then returns to the shore and 
     
     
  
 
 
	 00:11:32
     ◼ 
      
     ►  
     And it's able to keep that concept straight the entire time and construct the whole scene 
     
     
  
 
 
	 00:11:39
     ◼ 
      
     ►  
     instead of just an image that's moving. 
     
     
  
 
 
	 00:11:42
     ◼ 
      
     ►  
     So the speed on this in the space of what, five weeks has been absolutely breathtaking. 
     
     
  
 
 
	 00:11:52
     ◼ 
      
     ►  
     This all seems inevitable. 
     
     
  
 
 
	 00:11:55
     ◼ 
      
     ►  
     I will say I am surprised at the speed like you, right? 
     
     
  
 
 
	 00:11:58
     ◼ 
      
     ►  
     that we've, from the last episode to now, how much has happened? 
     
     
  
 
 
	 00:12:03
     ◼ 
      
     ►  
     ALICE The thing that I just want to draw people's 
     
     
  
 
 
	 00:12:05
     ◼ 
      
     ►  
     attention to that gets overlooked in the AI art discussion is, people are talking about 
     
     
  
 
 
	 00:12:12
     ◼ 
      
     ►  
     the art because it's visual. 
     
     
  
 
 
	 00:12:15
     ◼ 
      
     ►  
     You can see it, I think it's the most understandable thing that's occurring. 
     
     
  
 
 
	 00:12:21
     ◼ 
      
     ►  
     But to me, the much bigger deal is the less visible language model stuff. 
     
     
  
 
 
	 00:12:28
     ◼ 
      
     ►  
     And so it is the same idea, but for words where AI is responding to some text prompt 
     
     
  
 
 
	 00:12:38
     ◼ 
      
     ►  
     with an answer of some kind. 
     
     
  
 
 
	 00:12:42
     ◼ 
      
     ►  
     And we talked in more texts about AI explaining jokes like how an AI can explain a joke listeners 
     
     
  
 
 
	 00:12:49
     ◼ 
      
     ►  
     can go check that out. 
     
     
  
 
 
	 00:12:50
     ◼ 
      
     ►  
     But I have an example that I didn't mention that a listener pointed out from that paper 
     
     
  
 
 
	 00:12:54
     ◼ 
      
     ►  
     that I had missed, which I think is much more impressive. 
     
     
  
 
 
	 00:12:57
     ◼ 
      
     ►  
     So again, you can do this same thing. 
     
     
  
 
 
	 00:13:00
     ◼ 
      
     ►  
     Ai doesn't understand words. 
     
     
  
 
 
	 00:13:03
     ◼ 
      
     ►  
     And so I'm going to give an input a thing that the user told the computer and then the 
     
     
  
 
 
	 00:13:09
     ◼ 
      
     ►  
     output how the computer answered. 
     
     
  
 
 
	 00:13:12
     ◼ 
      
     ►  
     So here is the input and this is this is very strange. 
     
     
  
 
 
	 00:13:15
     ◼ 
      
     ►  
     It's almost like a puzzle just so you can understand what I'm what I'm saying here. 
     
     
  
 
 
	 00:13:20
     ◼ 
      
     ►  
     Input Michael is at that really famous museum in France looking at its most famous painting. 
     
     
  
 
 
	 00:13:26
     ◼ 
      
     ►  
     However, the artist who made this painting just makes Michael think of his favorite cartoon 
     
     
  
 
 
	 00:13:32
     ◼ 
      
     ►  
     character from childhood. 
     
     
  
 
 
	 00:13:35
     ◼ 
      
     ►  
     What was the country of origin of the thing that the cartoon character usually holds in 
     
     
  
 
 
 
	 00:13:43
     ◼ 
      
     ►  
     So do you think you have an answer to that question? 
     
     
  
 
 
 
	 00:13:48
     ◼ 
      
     ►  
     Okay, here is what the AI said. 
     
     
  
 
 
	 00:13:52
     ◼ 
      
     ►  
     The most famous painting in the Louvre is the Mona Lisa. 
     
     
  
 
 
	 00:13:56
     ◼ 
      
     ►  
     The artist who made the Mona Lisa is Leonardo da Vinci. 
     
     
  
 
 
	 00:14:00
     ◼ 
      
     ►  
     Leonardo da Vinci is also the name of the main character in the cartoon Teenage Mutant 
     
     
  
 
 
	 00:14:04
     ◼ 
      
     ►  
     Ninja Turtles. 
     
     
  
 
 
	 00:14:06
     ◼ 
      
     ►  
     Leonardo da Vinci is from Italy. 
     
     
  
 
 
	 00:14:08
     ◼ 
      
     ►  
     The thing that Leonardo usually holds in his hand is a katana. 
     
     
  
 
 
	 00:14:12
     ◼ 
      
     ►  
     The country of origin of the katana is Japan. 
     
     
  
 
 
	 00:14:16
     ◼ 
      
     ►  
     The answer is Japan. 
     
     
  
 
 
 
 
	 00:14:19
     ◼ 
      
     ►  
     I mean, part of my issue there is I forgot it was Leonardo da Vinci who made the Mona 
     
     
  
 
 
	 00:14:22
     ◼ 
      
     ►  
     Lisa but I would not have remembered that Leonardo the turtle from Ninja Turtles 
     
     
  
 
 
	 00:14:27
     ◼ 
      
     ►  
     held a katana because I thought that he had like a kendo stick or something that 
     
     
  
 
 
	 00:14:31
     ◼ 
      
     ►  
     is horrifying right so because this is because this is actually a good example 
     
     
  
 
 
	 00:14:37
     ◼ 
      
     ►  
     right because my human brain forgets things mm-hmm the AI brain doesn't it 
     
     
  
 
 
	 00:14:46
     ◼ 
      
     ►  
     just has all the information or can get it I assume I don't know where from but 
     
     
  
 
 
	 00:14:51
     ◼ 
      
     ►  
     But like the difference between pitching me against the machine there is I just couldn't 
     
     
  
 
 
	 00:14:57
     ◼ 
      
     ►  
     remember a couple of key pieces of information, which now you tell me I did know them. 
     
     
  
 
 
 
	 00:15:03
     ◼ 
      
     ►  
     Yeah, or even when I first read this input, it's just phrased in such a strange and vague 
     
     
  
 
 
 
	 00:15:10
     ◼ 
      
     ►  
     Yeah, I do wonder why they did it that way. 
     
     
  
 
 
	 00:15:12
     ◼ 
      
     ►  
     That is odd to me. 
     
     
  
 
 
	 00:15:13
     ◼ 
      
     ►  
     I think the whole purpose of why it's framed in this really strange way is to give the 
     
     
  
 
 
	 00:15:19
     ◼ 
      
     ►  
     minimal amount of information that you can regarding what the actual answer is. They 
     
     
  
 
 
	 00:15:25
     ◼ 
      
     ►  
     don't even say like the most famous museum in France, it's a really famous museum in 
     
     
  
 
 
	 00:15:29
     ◼ 
      
     ►  
     France. Right? And also they're like, they're trying to get an answer, but they're asking 
     
     
  
 
 
	 00:15:34
     ◼ 
      
     ►  
     a bunch of questions that require recalling the previous answer, right? Right. It's a 
     
     
  
 
 
	 00:15:40
     ◼ 
      
     ►  
     multi-stage thing to think through. I think the particular one that's really killer here 
     
     
  
 
 
	 00:15:44
     ◼ 
      
     ►  
     is the artist who made this painting just makes Michael think of his favorite cartoon 
     
     
  
 
 
	 00:15:49
     ◼ 
      
     ►  
     character from his childhood. For the AI to make the connection, "Oh, it's Leonardo 
     
     
  
 
 
	 00:15:55
     ◼ 
      
     ►  
     da Vinci. Leonardo is one of the Teenage Mutant Ninja Turtles." 
     
     
  
 
 
	 00:15:58
     ◼ 
      
     ►  
     - Because there's no concept of how old the person is, right? 
     
     
  
 
 
	 00:16:01
     ◼ 
      
     ►  
     - Yeah, there's no concept of how old the person is. Simply being able to draw that 
     
     
  
 
 
	 00:16:06
     ◼ 
      
     ►  
     out of like, what painting in France would make someone think of a childhood cartoon? 
     
     
  
 
 
	 00:16:13
     ◼ 
      
     ►  
     a crazy abstract thing to think of, but the machine got it. So I think this like input 
     
     
  
 
 
	 00:16:19
     ◼ 
      
     ►  
     output is a an example of again the AI art stuff is interesting. The text stuff to me 
     
     
  
 
 
	 00:16:27
     ◼ 
      
     ►  
     is where a lot of the much more concerning stuff comes from and there's a little bit 
     
     
  
 
 
	 00:16:32
     ◼ 
      
     ►  
     of a less visceral demo, but I actually found this the most striking thing I have found 
     
     
  
 
 
	 00:16:38
     ◼ 
      
     ►  
     in language models, but it's a little bit hard to describe. So there's an article 
     
     
  
 
 
	 00:16:42
     ◼ 
      
     ►  
     called "Using GPT-3 to Pathfind in Random Graphs". GPT-3 is one of these language models. 
     
     
  
 
 
	 00:16:50
     ◼ 
      
     ►  
     It's the most advanced one that people have access to. Although at the conference I was 
     
     
  
 
 
	 00:16:56
     ◼ 
      
     ►  
     at I got to see some of the not publicly released stuff, which was much more terrifying, but 
     
     
  
 
 
	 00:17:00
     ◼ 
      
     ►  
     we'll just leave that alone for now. GPT-3 is the much older one. And this is the most 
     
     
  
 
 
	 00:17:07
     ◼ 
      
     ►  
     impressive thing that I've seen. So this is a little bit hard to describe on a podcast, 
     
     
  
 
 
	 00:17:13
     ◼ 
      
     ►  
     but there's this problem in mathematics, which is called the traveling salesman problem, 
     
     
  
 
 
	 00:17:16
     ◼ 
      
     ►  
     which is say, Oh, you're a salesman and you have to get to 10 cities in the United States, 
     
     
  
 
 
	 00:17:23
     ◼ 
      
     ►  
     what's the shortest possible path to travel between those 10 cities in any order. And 
     
     
  
 
 
	 00:17:28
     ◼ 
      
     ►  
     it's one of those things like it seems like it should be easy, but it turns out this is 
     
     
  
 
 
	 00:17:32
     ◼ 
      
     ►  
     just incredibly difficult to actually solve in a reasonable way. Just a hugely computationally 
     
     
  
 
 
	 00:17:38
     ◼ 
      
     ►  
     intense problem. So someone basically got the idea of, hey, why don't I see if GPT three 
     
     
  
 
 
	 00:17:44
     ◼ 
      
     ►  
     can solve the traveling salesman problem. So I think you need to go read the paper to 
     
     
  
 
 
	 00:17:49
     ◼ 
      
     ►  
     see the details. But let me just describe it in a general way, where the person gave 
     
     
  
 
 
	 00:17:55
     ◼ 
      
     ►  
     an input to GPT three that says something like this. There's 17 locations, we need to 
     
     
  
 
 
	 00:18:03
     ◼ 
      
     ►  
     find the shortest path between these locations. Location one is connected to location seven 
     
     
  
 
 
	 00:18:09
     ◼ 
      
     ►  
     and 13. Location 13 is connected to locations nine and two. So they just wrote out like 
     
     
  
 
 
	 00:18:16
     ◼ 
      
     ►  
     a bunch of here's all of the connections. You're currently at location three, find the 
     
     
  
 
 
	 00:18:21
     ◼ 
      
     ►  
     optimal path to location eight. And GPT three just did it. Basically, half of the time it 
     
     
  
 
 
	 00:18:30
     ◼ 
      
     ►  
     was able to find an optimal or near optimal path, just given a bunch of random locations 
     
     
  
 
 
	 00:18:36
     ◼ 
      
     ►  
     and the connections between them. And what's just what's like, that may not sound like 
     
     
  
 
 
	 00:18:42
     ◼ 
      
     ►  
     much but what needs to be understood here, what is absolutely mind blowing is that GPT 
     
     
  
 
 
	 00:18:47
     ◼ 
      
     ►  
     has just been given a ton of like text documents from which to derive the world. 
     
     
  
 
 
	 00:18:54
     ◼ 
      
     ►  
     And inside of it, somehow it has the concept of locations and connections 
     
     
  
 
 
	 00:19:01
     ◼ 
      
     ►  
     and what does it mean to try to find the path to a different location. 
     
     
  
 
 
	 00:19:06
     ◼ 
      
     ►  
     Somehow it's thinking of what is the optimal solution. 
     
     
  
 
 
	 00:19:10
     ◼ 
      
     ►  
     And again, like it's not programmed for this. 
     
     
  
 
 
	 00:19:12
     ◼ 
      
     ►  
     This is an incredibly difficult math problem, but it has an idea of what 
     
     
  
 
 
	 00:19:17
     ◼ 
      
     ►  
     of what's going on. And so I've just seen more and more people are really trying to 
     
     
  
 
 
	 00:19:22
     ◼ 
      
     ►  
     push the edges of these language models to say, what do you understand? And the answer 
     
     
  
 
 
	 00:19:30
     ◼ 
      
     ►  
     keeps coming out to be much more than you might have imagined that there's a better 
     
     
  
 
 
	 00:19:34
     ◼ 
      
     ►  
     understanding of something in the world that has just been derived from dumping tons and 
     
     
  
 
 
	 00:19:42
     ◼ 
      
     ►  
     tons of text files into this big database and building a neural network on top of it. 
     
     
  
 
 
	 00:19:47
     ◼ 
      
     ►  
     It is genuinely, genuinely terrifying. And I'll just say like, I don't think I can talk 
     
     
  
 
 
	 00:19:54
     ◼ 
      
     ►  
     in much detail about what I have seen. But I saw a couple demos of the next generation of this stuff 
     
     
  
 
 
	 00:20:01
     ◼ 
      
     ►  
     in person. And it was extremely alarming what you could ask it to do. And it could give reasonable 
     
     
  
 
 
	 00:20:08
     ◼ 
      
     ►  
     answers to. And I think just like the AI art stuff, it has clearly crossed into the realm 
     
     
  
 
 
	 00:20:17
     ◼ 
      
     ►  
     of creative in a way that I think people just wouldn't have expected. 
     
     
  
 
 
	 00:20:22
     ◼ 
      
     ►  
     Like writing fiction? 
     
     
  
 
 
	 00:20:23
     ◼ 
      
     ►  
     Exactly, yes. That's exactly the kind of thing. Some public similar projects that are 
     
     
  
 
 
	 00:20:28
     ◼ 
      
     ►  
     easier to talk about are, there's a few cases of auto-generated text adventures. So 
     
     
  
 
 
	 00:20:35
     ◼ 
      
     ►  
     Exactly the thing that we do, right, for the bonus episodes at Cortex where we work through 
     
     
  
 
 
	 00:20:39
     ◼ 
      
     ►  
     a text adventure that someone has created, there are now projects that do that, where 
     
     
  
 
 
	 00:20:44
     ◼ 
      
     ►  
     you can just play an infinite text adventure where it keeps spitting out like, "Oh, you're 
     
     
  
 
 
	 00:20:49
     ◼ 
      
     ►  
     in this room and here are your options of what you can do." And if you select an option, 
     
     
  
 
 
	 00:20:53
     ◼ 
      
     ►  
     like it'll just go forever in a coherent way. It's like, "Oh my!" So yeah, I think 
     
     
  
 
 
	 00:21:01
     ◼ 
      
     ►  
     the AI art stuff is flashy. The language model stuff is what people are going to be quite 
     
     
  
 
 
	 00:21:08
     ◼ 
      
     ►  
     surprised at how much it might take over soon. And that includes a lot of things that people 
     
     
  
 
 
	 00:21:14
     ◼ 
      
     ►  
     would not expect like computer code. There are already some public examples of this of 
     
     
  
 
 
	 00:21:20
     ◼ 
      
     ►  
     you express in human language, the computer code that you want created, and it is able to create 
     
     
  
 
 
	 00:21:27
     ◼ 
      
     ►  
     that code for you. So we're rapidly encroaching on computers programming themselves territory. 
     
     
  
 
 
	 00:21:34
     ◼ 
      
     ►  
     Mykey - Definitely want the computer to think for itself. That's what I'm looking for. 
     
     
  
 
 
	 00:21:39
     ◼ 
      
     ►  
     Alex - I mean look, look, we can get to the Doom stuff later. Like there's so much other 
     
     
  
 
 
	 00:21:43
     ◼ 
      
     ►  
     stuff to talk about. 
     
     
  
 
 
	 00:21:44
     ◼ 
      
     ►  
     Mykey - Wait a second, what do you mean later? Am we not already in it? I feel like we're 
     
     
  
 
 
	 00:21:47
     ◼ 
      
     ►  
     in it. What are you talking about? 
     
     
  
 
 
	 00:21:48
     ◼ 
      
     ►  
     Alex - No, no Myke. I don't think this is the Doom stuff at all. This is just timeline 
     
     
  
 
 
 
	 00:21:53
     ◼ 
      
     ►  
     Mykey - Oh okay. May I follow up at this point? 
     
     
  
 
 
	 00:21:54
     ◼ 
      
     ►  
     "Hey, what's happened since we last spoke?" 
     
     
  
 
 
	 00:21:57
     ◼ 
      
     ►  
     And a lot has happened since we last spoke. 
     
     
  
 
 
	 00:22:00
     ◼ 
      
     ►  
     And this is why at the beginning I kind of divided, 
     
     
  
 
 
	 00:22:03
     ◼ 
      
     ►  
     "Oh, there's stuff in the near term, 
     
     
  
 
 
	 00:22:05
     ◼ 
      
     ►  
     and there's stuff in the long term." 
     
     
  
 
 
	 00:22:07
     ◼ 
      
     ►  
     And we're still just in the near term conversation, Myke. 
     
     
  
 
 
	 00:22:09
     ◼ 
      
     ►  
     Like we're not even remotely close 
     
     
  
 
 
	 00:22:11
     ◼ 
      
     ►  
     to the long term conversation. 
     
     
  
 
 
 
	 00:22:15
     ◼ 
      
     ►  
     Hey, this isn't an ad, 
     
     
  
 
 
	 00:22:17
     ◼ 
      
     ►  
     but we're putting it in between the ad sounds 
     
     
  
 
 
	 00:22:20
     ◼ 
      
     ►  
     because Myke and I have been talking 
     
     
  
 
 
	 00:22:23
     ◼ 
      
     ►  
     for forever about AI. 
     
     
  
 
 
	 00:22:26
     ◼ 
      
     ►  
     And we forgot that we need to tell you about the subtlety 
     
     
  
 
 
	 00:22:31
     ◼ 
      
     ►  
     and the subtle sweater going on sale 
     
     
  
 
 
	 00:22:33
     ◼ 
      
     ►  
     sometime before the 90 minute mark of the show. 
     
     
  
 
 
	 00:22:36
     ◼ 
      
     ►  
     So we're just breaking into our own conversation now 
     
     
  
 
 
	 00:22:40
     ◼ 
      
     ►  
     from the future. 
     
     
  
 
 
	 00:22:41
     ◼ 
      
     ►  
     - This is us from the future coming back to tell you 
     
     
  
 
 
	 00:22:44
     ◼ 
      
     ►  
     that we talk about AI for an hour and a half 
     
     
  
 
 
	 00:22:46
     ◼ 
      
     ►  
     accidentally in this episode. 
     
     
  
 
 
	 00:22:48
     ◼ 
      
     ►  
     And we realize as purveyors of fine merchandise, 
     
     
  
 
 
	 00:22:52
     ◼ 
      
     ►  
     That is a terrible way to structure the show. 
     
     
  
 
 
	 00:22:55
     ◼ 
      
     ►  
     If you want to let people know 
     
     
  
 
 
	 00:22:57
     ◼ 
      
     ►  
     that the incredibly popular and beloved 
     
     
  
 
 
	 00:22:59
     ◼ 
      
     ►  
     Subtlety and Subtle Sweater is back 
     
     
  
 
 
	 00:23:01
     ◼ 
      
     ►  
     for its one time a year sale, 
     
     
  
 
 
	 00:23:04
     ◼ 
      
     ►  
     to leave that 90 minutes into the episode, 
     
     
  
 
 
	 00:23:07
     ◼ 
      
     ►  
     we are doing a bad job selling our products. 
     
     
  
 
 
	 00:23:10
     ◼ 
      
     ►  
     So if you go to cortexmerch.com right now, 
     
     
  
 
 
	 00:23:15
     ◼ 
      
     ►  
     you will find until November 8th, 2022, 
     
     
  
 
 
	 00:23:20
     ◼ 
      
     ►  
     a limited time sale of the subtlety and subtle sweater. 
     
     
  
 
 
	 00:23:24
     ◼ 
      
     ►  
     We are bringing back all of our beloved colors, 
     
     
  
 
 
	 00:23:27
     ◼ 
      
     ►  
     the original blue, black, green, and red. 
     
     
  
 
 
	 00:23:31
     ◼ 
      
     ►  
     We are adding a new color this year, gray. 
     
     
  
 
 
	 00:23:35
     ◼ 
      
     ►  
     - The best color. 
     
     
  
 
 
	 00:23:36
     ◼ 
      
     ►  
     - Now I am really into this. 
     
     
  
 
 
	 00:23:39
     ◼ 
      
     ►  
     So we did some general merch at Relay 
     
     
  
 
 
	 00:23:42
     ◼ 
      
     ►  
     and we added a bunch of colors for some stuff we were doing, 
     
     
  
 
 
	 00:23:45
     ◼ 
      
     ►  
     including just a light gray sweater option 
     
     
  
 
 
	 00:23:48
     ◼ 
      
     ►  
     for one of the shirts that we did. 
     
     
  
 
 
	 00:23:49
     ◼ 
      
     ►  
     And as soon as I got it, I was like, 
     
     
  
 
 
	 00:23:51
     ◼ 
      
     ►  
     oh God, I need this in a subtlety. 
     
     
  
 
 
	 00:23:53
     ◼ 
      
     ►  
     Because that light gray color, 
     
     
  
 
 
	 00:23:56
     ◼ 
      
     ►  
     it's just like the traditional sweater color, right? 
     
     
  
 
 
	 00:23:59
     ◼ 
      
     ►  
     So we now have that available 
     
     
  
 
 
	 00:24:01
     ◼ 
      
     ►  
     and look so good with the light blue stitching. 
     
     
  
 
 
	 00:24:04
     ◼ 
      
     ►  
     So that's available in tees and in sweaters 
     
     
  
 
 
	 00:24:08
     ◼ 
      
     ►  
     with all of the other colors, 
     
     
  
 
 
	 00:24:10
     ◼ 
      
     ►  
     red, green, blue, black, and gray. 
     
     
  
 
 
	 00:24:13
     ◼ 
      
     ►  
     You can get any of them. 
     
     
  
 
 
	 00:24:14
     ◼ 
      
     ►  
     This year gray, 
     
     
  
 
 
	 00:24:15
     ◼ 
      
     ►  
     I am going to be replacing my original blue sweatshirts, 
     
     
  
 
 
	 00:24:19
     ◼ 
      
     ►  
     I've decided. 
     
     
  
 
 
	 00:24:20
     ◼ 
      
     ►  
     - Ooh, okay. 
     
     
  
 
 
	 00:24:21
     ◼ 
      
     ►  
     - 'Cause I mean, I've had those for like four years now 
     
     
  
 
 
	 00:24:24
     ◼ 
      
     ►  
     or something, so I'm gonna get some new ones. 
     
     
  
 
 
	 00:24:26
     ◼ 
      
     ►  
     - You're probably right. 
     
     
  
 
 
	 00:24:27
     ◼ 
      
     ►  
     From where I'm recording right now, 
     
     
  
 
 
	 00:24:28
     ◼ 
      
     ►  
     I can look into my closet and, 
     
     
  
 
 
	 00:24:30
     ◼ 
      
     ►  
     this is not an exaggeration, 
     
     
  
 
 
	 00:24:31
     ◼ 
      
     ►  
     one third of my closet is subtle sweaters, subtle jeans. 
     
     
  
 
 
 
	 00:24:38
     ◼ 
      
     ►  
     - It's the same for me. 
     
     
  
 
 
	 00:24:39
     ◼ 
      
     ►  
     I'm wearing a green tee, the green sleeper hit. 
     
     
  
 
 
	 00:24:43
     ◼ 
      
     ►  
     The green is so good. 
     
     
  
 
 
	 00:24:44
     ◼ 
      
     ►  
     I'm buying a bunch more green. 
     
     
  
 
 
	 00:24:46
     ◼ 
      
     ►  
     - Green looks great. 
     
     
  
 
 
	 00:24:46
     ◼ 
      
     ►  
     - Like this happens to me every year. 
     
     
  
 
 
	 00:24:48
     ◼ 
      
     ►  
     We get a new color, I buy one of each, 
     
     
  
 
 
	 00:24:49
     ◼ 
      
     ►  
     and I'm like, "God damn it, I wish I put more of them." 
     
     
  
 
 
	 00:24:52
     ◼ 
      
     ►  
     And then I also have to wait a year to get more of them. 
     
     
  
 
 
	 00:24:56
     ◼ 
      
     ►  
     But that green, so good. 
     
     
  
 
 
	 00:24:58
     ◼ 
      
     ►  
     - For me actually, surprisingly, it was the red. 
     
     
  
 
 
	 00:25:01
     ◼ 
      
     ►  
     Like I bought some of the red just like, 
     
     
  
 
 
	 00:25:03
     ◼ 
      
     ►  
     "Oh, it's good to have a complete set for me." 
     
     
  
 
 
	 00:25:06
     ◼ 
      
     ►  
     I wear it a surprising amount, 
     
     
  
 
 
	 00:25:07
     ◼ 
      
     ►  
     and I would never have picked that as a color for myself. 
     
     
  
 
 
	 00:25:09
     ◼ 
      
     ►  
     But yeah, the subtle sweaters and the subtle tees, 
     
     
  
 
 
	 00:25:13
     ◼ 
      
     ►  
     They're seriously so comfortable. 
     
     
  
 
 
	 00:25:16
     ◼ 
      
     ►  
     It's we get just a ton of positive feedback from people who really like them, which is 
     
     
  
 
 
	 00:25:21
     ◼ 
      
     ►  
     part of why we realized we've got to break into the show and remind you, "Hey, they're 
     
     
  
 
 
 
	 00:25:28
     ◼ 
      
     ►  
     If you want them, you need to get them now. 
     
     
  
 
 
	 00:25:31
     ◼ 
      
     ►  
     People love them. 
     
     
  
 
 
	 00:25:32
     ◼ 
      
     ►  
     It's just this one time sale. 
     
     
  
 
 
	 00:25:34
     ◼ 
      
     ►  
     So go and order them right now." 
     
     
  
 
 
 
	 00:25:38
     ◼ 
      
     ►  
     I cannot impress upon enough that you do this. 
     
     
  
 
 
	 00:25:40
     ◼ 
      
     ►  
     We will not have another episode come out to remind you to do this before the sale is 
     
     
  
 
 
 
	 00:25:46
     ◼ 
      
     ►  
     It's a three week sale from when the episode goes out. 
     
     
  
 
 
	 00:25:49
     ◼ 
      
     ►  
     November 8th is when it's done and it's gone for a year. 
     
     
  
 
 
	 00:25:53
     ◼ 
      
     ►  
     So we only do this once a year. 
     
     
  
 
 
	 00:25:55
     ◼ 
      
     ►  
     So if you want them, you need to go and get them at cortexmerch.com. 
     
     
  
 
 
	 00:26:00
     ◼ 
      
     ►  
     Cannot impress upon you enough as well how good these things look. 
     
     
  
 
 
	 00:26:02
     ◼ 
      
     ►  
     Like a couple of days ago I was wearing the red one and I was walking towards a glass 
     
     
  
 
 
	 00:26:06
     ◼ 
      
     ►  
     door and I was like, "Damn that looks good." 
     
     
  
 
 
	 00:26:10
     ◼ 
      
     ►  
     It's just a good, our brand on a shirt like that 
     
     
  
 
 
	 00:26:14
     ◼ 
      
     ►  
     or on a sweatshirt like that, it just looks so professional. 
     
     
  
 
 
	 00:26:18
     ◼ 
      
     ►  
     I'm so happy we did this as a thing. 
     
     
  
 
 
	 00:26:21
     ◼ 
      
     ►  
     They are so great. 
     
     
  
 
 
	 00:26:22
     ◼ 
      
     ►  
     - I also feel like I'm trying to do the impossible thing, 
     
     
  
 
 
	 00:26:25
     ◼ 
      
     ►  
     which is it happens every year that when the sale is over, 
     
     
  
 
 
	 00:26:28
     ◼ 
      
     ►  
     we get contacted from people who are going like, 
     
     
  
 
 
	 00:26:30
     ◼ 
      
     ►  
     "Oh, I want the shirts, how can I buy them?" 
     
     
  
 
 
	 00:26:33
     ◼ 
      
     ►  
     And I'm like, "I'm trying to talk to you right now, 
     
     
  
 
 
	 00:26:36
     ◼ 
      
     ►  
     "the person who's going to be sad in five weeks 
     
     
  
 
 
	 00:26:39
     ◼ 
      
     ►  
     when they can't get them. You need to do this now. We're still going to be talking about AI later, 
     
     
  
 
 
	 00:26:46
     ◼ 
      
     ►  
     but know if you ever want these shirts, which you definitely do, if you want to wear the most 
     
     
  
 
 
	 00:26:51
     ◼ 
      
     ►  
     comfortable sweater you have ever worn, go to cortexmerch.com right now and get yourself some 
     
     
  
 
 
	 00:26:59
     ◼ 
      
     ►  
     fantastic clothing to wear. You will not regret it. Everybody loves these things and they're 
     
     
  
 
 
	 00:27:04
     ◼ 
      
     ►  
     sad when they're not on sale. So, CortexMerch.com, pause the podcast right now. 
     
     
  
 
 
	 00:27:10
     ◼ 
      
     ►  
     Or let me give a secondary thing. If you're like, "But Gray, Myke, I'm driving!" 
     
     
  
 
 
 
	 00:27:16
     ◼ 
      
     ►  
     Here's what I'll tell you you can do. Imagine in your mind now something you see at the 
     
     
  
 
 
	 00:27:21
     ◼ 
      
     ►  
     end of your commute. It might be a billboard, it might be like a sign in your parking space, 
     
     
  
 
 
	 00:27:26
     ◼ 
      
     ►  
     it might be your garage door, whatever. Imagine that thing right now, tie in your brain the 
     
     
  
 
 
	 00:27:32
     ◼ 
      
     ►  
     image of that thing with cortexmerch.com. Think about it right now. Say it in your mind 
     
     
  
 
 
	 00:27:37
     ◼ 
      
     ►  
     a few times. So when you arrive at your destination, you have set yourself a reminder. 
     
     
  
 
 
	 00:27:42
     ◼ 
      
     ►  
     Look, Myke's trying to pull some fancy memory palace stuff over here. You know, my method 
     
     
  
 
 
	 00:27:48
     ◼ 
      
     ►  
     is much more direct. I just say, "Hey Siri, remind me in three hours to go to cortexmerch.com." 
     
     
  
 
 
	 00:27:55
     ◼ 
      
     ►  
     Hey Google, remind me in three hours. 
     
     
  
 
 
 
	 00:28:00
     ◼ 
      
     ►  
     Hey Google, remind me in three hours 
     
     
  
 
 
	 00:28:02
     ◼ 
      
     ►  
     to go to cortexmerch.com. 
     
     
  
 
 
 
	 00:28:06
     ◼ 
      
     ►  
     We're breaking into your life now. 
     
     
  
 
 
	 00:28:08
     ◼ 
      
     ►  
     What are you gonna do? 
     
     
  
 
 
 
	 00:28:11
     ◼ 
      
     ►  
     Cortexmerch.com. 
     
     
  
 
 
	 00:28:14
     ◼ 
      
     ►  
     - Okay, and now back to us in the past talking about AI. 
     
     
  
 
 
	 00:28:17
     ◼ 
      
     ►  
     - Are we gonna have that long-term conversation? 
     
     
  
 
 
	 00:28:21
     ◼ 
      
     ►  
     - Yeah, yeah, yeah, we will, we will. 
     
     
  
 
 
	 00:28:23
     ◼ 
      
     ►  
     But like, I think there's also just leftover from last time, 
     
     
  
 
 
	 00:28:28
     ◼ 
      
     ►  
     there's still just like a bunch of stuff in the near term 
     
     
  
 
 
	 00:28:31
     ◼ 
      
     ►  
     that might be worth revisiting. 
     
     
  
 
 
	 00:28:34
     ◼ 
      
     ►  
     One of the other things that came up last time is, 
     
     
  
 
 
	 00:28:38
     ◼ 
      
     ►  
     how do these models work? 
     
     
  
 
 
	 00:28:40
     ◼ 
      
     ►  
     Like how do they even begin to start creating anything, 
     
     
  
 
 
	 00:28:44
     ◼ 
      
     ►  
     whether it's poetry or videos or computer codes, 
     
     
  
 
 
	 00:28:48
     ◼ 
      
     ►  
     like how does this work? 
     
     
  
 
 
	 00:28:50
     ◼ 
      
     ►  
     And fundamentally, these AI systems are made by just hoovering up a ton of information 
     
     
  
 
 
	 00:28:57
     ◼ 
      
     ►  
     in the relative domain and feeding it into the system for the system to be trained upon. 
     
     
  
 
 
	 00:29:03
     ◼ 
      
     ►  
     And I do think one of the most concerning short-term questions about that is like, 
     
     
  
 
 
	 00:29:11
     ◼ 
      
     ►  
     what does it mean to use the public work that people have done, whatever that is? 
     
     
  
 
 
	 00:29:19
     ◼ 
      
     ►  
     you've written a book and you've published it. You've gone on to Stack Overflow and you've helped 
     
     
  
 
 
	 00:29:24
     ◼ 
      
     ►  
     answer hundreds of people's questions about computer code. You've been on DeviantArt for 
     
     
  
 
 
	 00:29:30
     ◼ 
      
     ►  
     years and you've made images and like that stuff has been sucked into a computer somewhere so that 
     
     
  
 
 
	 00:29:36
     ◼ 
      
     ►  
     it can then produce imitations or produce new work based on what you have done. I think that's like a 
     
     
  
 
 
	 00:29:44
     ◼ 
      
     ►  
     That is just a really difficult question. 
     
     
  
 
 
	 00:29:47
     ◼ 
      
     ►  
     - Human inspiration. 
     
     
  
 
 
	 00:29:49
     ◼ 
      
     ►  
     So, one of the things that a lot of people brought up 
     
     
  
 
 
	 00:29:52
     ◼ 
      
     ►  
     is how is this any different to being inspired 
     
     
  
 
 
	 00:29:56
     ◼ 
      
     ►  
     by someone's work and creating your own work? 
     
     
  
 
 
	 00:29:59
     ◼ 
      
     ►  
     I feel like it is quite different, but what do you think? 
     
     
  
 
 
	 00:30:03
     ◼ 
      
     ►  
     - I think it's different, but it is hard to articulate 
     
     
  
 
 
	 00:30:13
     ◼ 
      
     ►  
     why in a coherent way? 
     
     
  
 
 
	 00:30:16
     ◼ 
      
     ►  
     - I feel like I have something that I think is pretty core 
     
     
  
 
 
	 00:30:19
     ◼ 
      
     ►  
     to me, but I also, I know a lot of people don't agree 
     
     
  
 
 
 
	 00:30:24
     ◼ 
      
     ►  
     - What is that? 
     
     
  
 
 
	 00:30:25
     ◼ 
      
     ►  
     - The difference to me is the skill required in acting 
     
     
  
 
 
	 00:30:29
     ◼ 
      
     ►  
     on the imitation is the thing that I actually think 
     
     
  
 
 
	 00:30:33
     ◼ 
      
     ►  
     is valuable. 
     
     
  
 
 
	 00:30:34
     ◼ 
      
     ►  
     - What do you mean by that sentence? 
     
     
  
 
 
	 00:30:36
     ◼ 
      
     ►  
     - Let's imagine we'll go with painting, right? 
     
     
  
 
 
	 00:30:40
     ◼ 
      
     ►  
     It's just a simple thing that we can all understand, right? 
     
     
  
 
 
	 00:30:42
     ◼ 
      
     ►  
     how somebody paints a picture. 
     
     
  
 
 
	 00:30:45
     ◼ 
      
     ►  
     If you look at a painting, 
     
     
  
 
 
	 00:30:49
     ◼ 
      
     ►  
     we'll talk about the previously famous Parisian art piece, 
     
     
  
 
 
	 00:30:53
     ◼ 
      
     ►  
     the Mona Lisa, right? 
     
     
  
 
 
 
	 00:30:55
     ◼ 
      
     ►  
     - And you wanna make your own Mona Lisa. 
     
     
  
 
 
	 00:30:57
     ◼ 
      
     ►  
     To be able to take the inspiration from that piece 
     
     
  
 
 
	 00:31:01
     ◼ 
      
     ►  
     and do it yourself, it is an imitation of previous work. 
     
     
  
 
 
	 00:31:04
     ◼ 
      
     ►  
     But you had to do it, you had to practice 
     
     
  
 
 
	 00:31:09
     ◼ 
      
     ►  
     and get the skill and build up your own level of skill 
     
     
  
 
 
	 00:31:13
     ◼ 
      
     ►  
     to perform that work. 
     
     
  
 
 
	 00:31:16
     ◼ 
      
     ►  
     Now, yes, all you have done is imitate it, 
     
     
  
 
 
	 00:31:18
     ◼ 
      
     ►  
     but if you get even 50% close, 
     
     
  
 
 
	 00:31:21
     ◼ 
      
     ►  
     you've made something that's interesting. 
     
     
  
 
 
	 00:31:22
     ◼ 
      
     ►  
     You've now learned the skills that you can go out maybe 
     
     
  
 
 
	 00:31:26
     ◼ 
      
     ►  
     and produce your own work, 
     
     
  
 
 
	 00:31:27
     ◼ 
      
     ►  
     but you've built the actual skill, the practice. 
     
     
  
 
 
	 00:31:31
     ◼ 
      
     ►  
     You've built the skill. 
     
     
  
 
 
	 00:31:33
     ◼ 
      
     ►  
     That is what I think is the thing that concerns me most 
     
     
  
 
 
	 00:31:37
     ◼ 
      
     ►  
     about this type of work. 
     
     
  
 
 
	 00:31:39
     ◼ 
      
     ►  
     is that I worry that the skills will get lost. 
     
     
  
 
 
	 00:31:43
     ◼ 
      
     ►  
     And I accept that some people do not value that 
     
     
  
 
 
	 00:31:48
     ◼ 
      
     ►  
     the same as me, but that's where I come from with this, 
     
     
  
 
 
	 00:31:52
     ◼ 
      
     ►  
     where I think that there is an inherent humanity 
     
     
  
 
 
	 00:31:56
     ◼ 
      
     ►  
     in these mostly inconsequential skills 
     
     
  
 
 
	 00:32:01
     ◼ 
      
     ►  
     that we value important as humans. 
     
     
  
 
 
	 00:32:04
     ◼ 
      
     ►  
     Practices, traditions, all of this kind of stuff 
     
     
  
 
 
	 00:32:09
     ◼ 
      
     ►  
     I hold those kinds of things dear and my concern with this a lot of this stuff is 
     
     
  
 
 
	 00:32:16
     ◼ 
      
     ►  
     we may lose this part in larger numbers like we may lose this part of our 
     
     
  
 
 
	 00:32:24
     ◼ 
      
     ►  
     humanity to more people if the creation of art is so simple. Like for example I 
     
     
  
 
 
	 00:32:30
     ◼ 
      
     ►  
     saw a comment on our YouTube video this morning when we put up and this 
     
     
  
 
 
	 00:32:36
     ◼ 
      
     ►  
     commenter had said that one of the things that they love about the idea of AI art is 
     
     
  
 
 
	 00:32:41
     ◼ 
      
     ►  
     that there are movies that people want to exist that don't currently exist and they 
     
     
  
 
 
	 00:32:49
     ◼ 
      
     ►  
     can type, they would maybe in the future be able to type a prompt into an AI art generator 
     
     
  
 
 
	 00:32:55
     ◼ 
      
     ►  
     and it would create that movie for them to watch. Honestly, I can't think of anything 
     
     
  
 
 
	 00:33:02
     ◼ 
      
     ►  
     worse than that. Why? Because there's no art in there, there's no passion in there, 
     
     
  
 
 
	 00:33:08
     ◼ 
      
     ►  
     there's no drive from a creator, from everyone involved in the creation of a 
     
     
  
 
 
	 00:33:14
     ◼ 
      
     ►  
     movie to come together and work towards something good. Now you may sit and think 
     
     
  
 
 
	 00:33:20
     ◼ 
      
     ►  
     to yourself, "Myke I don't agree with what you're saying right now" and that is 
     
     
  
 
 
	 00:33:23
     ◼ 
      
     ►  
     perfectly fine but I just want people to understand from my perspective the 
     
     
  
 
 
	 00:33:28
     ◼ 
      
     ►  
     The creation of art is as important, if not more important, than the piece that is at 
     
     
  
 
 
 
	 00:33:35
     ◼ 
      
     ►  
     And I think if all we end up with is just a bunch of pieces at the end, we will lose 
     
     
  
 
 
	 00:33:39
     ◼ 
      
     ►  
     so much of the humanity in this work. 
     
     
  
 
 
	 00:33:43
     ◼ 
      
     ►  
     The ideas that somebody might have that sparks off something to create a different shot in 
     
     
  
 
 
 
	 00:33:49
     ◼ 
      
     ►  
     Like that is what I actually hold to be so important to who we are as the human race, 
     
     
  
 
 
	 00:33:56
     ◼ 
      
     ►  
     rather than just here is media to consume. 
     
     
  
 
 
	 00:34:00
     ◼ 
      
     ►  
     So I don't know. 
     
     
  
 
 
	 00:34:01
     ◼ 
      
     ►  
     I don't know if I'm expressing myself clearly, 
     
     
  
 
 
	 00:34:05
     ◼ 
      
     ►  
     but I just want people to understand 
     
     
  
 
 
	 00:34:07
     ◼ 
      
     ►  
     that the thing that I care about is the creation of the art. 
     
     
  
 
 
	 00:34:12
     ◼ 
      
     ►  
     And it's not even just about jobs. 
     
     
  
 
 
	 00:34:15
     ◼ 
      
     ►  
     I just worry that we will lose this part of who we are. 
     
     
  
 
 
	 00:34:20
     ◼ 
      
     ►  
     And one of the things that makes us different 
     
     
  
 
 
	 00:34:25
     ◼ 
      
     ►  
     to every other species on this planet 
     
     
  
 
 
	 00:34:28
     ◼ 
      
     ►  
     is this kind of thing that we do, 
     
     
  
 
 
	 00:34:32
     ◼ 
      
     ►  
     sometimes for pleasure, mostly for pleasure. 
     
     
  
 
 
	 00:34:34
     ◼ 
      
     ►  
     There is business in it, but people like to make things 
     
     
  
 
 
	 00:34:38
     ◼ 
      
     ►  
     because they just like to make them. 
     
     
  
 
 
	 00:34:40
     ◼ 
      
     ►  
     That's something I find to be so beautiful 
     
     
  
 
 
	 00:34:42
     ◼ 
      
     ►  
     and I don't get the same sense of pleasure 
     
     
  
 
 
	 00:34:46
     ◼ 
      
     ►  
     out of typing six complex sentences into a text field 
     
     
  
 
 
	 00:34:50
     ◼ 
      
     ►  
     to then look at an image. 
     
     
  
 
 
	 00:34:52
     ◼ 
      
     ►  
     Maybe I'm old-fashioned. 
     
     
  
 
 
	 00:34:55
     ◼ 
      
     ►  
     If I'm trying to summarize your position, because I often have a hard time when 
     
     
  
 
 
	 00:35:01
     ◼ 
      
     ►  
     people use language like it's a fundamental part of our humanity. 
     
     
  
 
 
	 00:35:06
     ◼ 
      
     ►  
     Like I'm not, I'm never quite sure what that means, but I guess I'm trying to 
     
     
  
 
 
	 00:35:11
     ◼ 
      
     ►  
     summarize your position as like, you think it is just a fundamental good that humans 
     
     
  
 
 
	 00:35:16
     ◼ 
      
     ►  
     are producing art and that part of that process is the skills that are required to be learned 
     
     
  
 
 
	 00:35:26
     ◼ 
      
     ►  
     in order to make that art. 
     
     
  
 
 
	 00:35:28
     ◼ 
      
     ►  
     - I think that there is an importance in it. 
     
     
  
 
 
	 00:35:30
     ◼ 
      
     ►  
     I can't tell you why, but it just feels, I don't know, there's a lot of emotion in it. 
     
     
  
 
 
	 00:35:37
     ◼ 
      
     ►  
     And I just want people to understand that I'm not sitting here like, "Oh no, my job's 
     
     
  
 
 
	 00:35:42
     ◼ 
      
     ►  
     gonna go away." 
     
     
  
 
 
	 00:35:44
     ◼ 
      
     ►  
     Because realistically, it's not going to in my lifetime. 
     
     
  
 
 
	 00:35:47
     ◼ 
      
     ►  
     I feel pretty confident about that. 
     
     
  
 
 
	 00:35:49
     ◼ 
      
     ►  
     We'll get to why in a little bit. 
     
     
  
 
 
	 00:35:51
     ◼ 
      
     ►  
     I have an example. 
     
     
  
 
 
	 00:35:53
     ◼ 
      
     ►  
     I think we'll kind of like clear up why I'm not concerned about my own job. 
     
     
  
 
 
	 00:35:57
     ◼ 
      
     ►  
     I'm concerned more about creativity as an idea, something that I care greatly about, 
     
     
  
 
 
	 00:36:05
     ◼ 
      
     ►  
     about people being creative, even if it's just for fun. 
     
     
  
 
 
	 00:36:10
     ◼ 
      
     ►  
     And I'm just not sure that I like this idea of creativity will just ultimately become 
     
     
  
 
 
	 00:36:18
     ◼ 
      
     ►  
     the same thing, which is how good can you be at writing a prompt? 
     
     
  
 
 
	 00:36:22
     ◼ 
      
     ►  
     Like that doesn't feel creative to me. 
     
     
  
 
 
	 00:36:25
     ◼ 
      
     ►  
     The act of the act, the process doesn't feel like it exists to me in the same way anymore. 
     
     
  
 
 
	 00:36:32
     ◼ 
      
     ►  
     I don't know. 
     
     
  
 
 
	 00:36:33
     ◼ 
      
     ►  
     I agree. I find it strange this argument that the new artistry will be in creating the prompts. 
     
     
  
 
 
	 00:36:40
     ◼ 
      
     ►  
     There's something I don't know there's something very odd to me about that argument. 
     
     
  
 
 
	 00:36:43
     ◼ 
      
     ►  
     It kind of reminds me of a while ago when AI systems started becoming the best 
     
     
  
 
 
	 00:36:50
     ◼ 
      
     ►  
     chess players in the world. There was this what to me always seemed like an absolutely bizarre idea 
     
     
  
 
 
	 00:36:56
     ◼ 
      
     ►  
     that the chess that there was I forget what they called it, but it was it was this concept of like, 
     
     
  
 
 
	 00:37:03
     ◼ 
      
     ►  
     "Oh, the best chess player will be a hybrid chess player, that it will be a person who is being 
     
     
  
 
 
	 00:37:09
     ◼ 
      
     ►  
     advised by the computer." And there was a period in time where that was true, that a tag team of a 
     
     
  
 
 
	 00:37:15
     ◼ 
      
     ►  
     human and a computer could beat the best computer and they could beat the best human. That always 
     
     
  
 
 
	 00:37:21
     ◼ 
      
     ►  
     seemed to me like a strange artifact, like this won't exist forever. This just happens to be the 
     
     
  
 
 
	 00:37:25
     ◼ 
      
     ►  
     situation right now, but I see no reason why the computers won't just ultimately outclass the human 
     
     
  
 
 
	 00:37:32
     ◼ 
      
     ►  
     And the human will just be like a monkey, right, adding absolutely nothing to this incredibly 
     
     
  
 
 
	 00:37:37
     ◼ 
      
     ►  
     complicated game that's taking place. 
     
     
  
 
 
	 00:37:39
     ◼ 
      
     ►  
     And it's not the best comparison, but I feel that there is something in this concept that 
     
     
  
 
 
	 00:37:46
     ◼ 
      
     ►  
     people have of artistry will be the prompts that's the same. 
     
     
  
 
 
	 00:37:51
     ◼ 
      
     ►  
     If the computers are getting so good at text, at the same time that computers are getting 
     
     
  
 
 
	 00:37:56
     ◼ 
      
     ►  
     so good of interpreting text to create art, why will they not just meet? 
     
     
  
 
 
	 00:38:00
     ◼ 
      
     ►  
     Yeah, maybe that's what it is, is it's like, why do you think you will be the best 
     
     
  
 
 
	 00:38:07
     ◼ 
      
     ►  
     at coming up with the sequence of words that generates the most interesting art? 
     
     
  
 
 
 
	 00:38:13
     ◼ 
      
     ►  
     I just don't— 
     
     
  
 
 
	 00:38:14
     ◼ 
      
     ►  
     Because similarly, it's the same argument of like, "Oh, the computer will make better 
     
     
  
 
 
	 00:38:18
     ◼ 
      
     ►  
     art than any human ever can," right? 
     
     
  
 
 
	 00:38:21
     ◼ 
      
     ►  
     Like that's kind of the thinking of like, "Oh, or just as good or good enough." 
     
     
  
 
 
	 00:38:24
     ◼ 
      
     ►  
     Then why do you think computers won't be just as good as you are at creating prompts? 
     
     
  
 
 
 
	 00:38:30
     ◼ 
      
     ►  
     maybe that's, you've kind of sharpened it up there. I find that argument strange. And even if, 
     
     
  
 
 
	 00:38:37
     ◼ 
      
     ►  
     let's say that for whatever reason that was never – it turned out that just that wasn't true. 
     
     
  
 
 
	 00:38:41
     ◼ 
      
     ►  
     It's built into the laws of the universe that humans are just great at writing prompts in ways 
     
     
  
 
 
	 00:38:47
     ◼ 
      
     ►  
     that machines will never be, which I think is strange. But let's just say it was true. 
     
     
  
 
 
	 00:38:50
     ◼ 
      
     ►  
     At least in my experience of playing around with Diffusion Bee, I just, I agree with you. I don't 
     
     
  
 
 
	 00:38:57
     ◼ 
      
     ►  
     I don't think there was really anything creative 
     
     
  
 
 
	 00:38:59
     ◼ 
      
     ►  
     about what I was doing. 
     
     
  
 
 
	 00:39:00
     ◼ 
      
     ►  
     - It's problem solving, it's creative problem solving, 
     
     
  
 
 
	 00:39:03
     ◼ 
      
     ►  
     but that's like a completely different thing 
     
     
  
 
 
	 00:39:05
     ◼ 
      
     ►  
     to what I care about, which is the process of practicing 
     
     
  
 
 
	 00:39:10
     ◼ 
      
     ►  
     and getting good at a thing. 
     
     
  
 
 
	 00:39:12
     ◼ 
      
     ►  
     Like I think that that's really important. 
     
     
  
 
 
	 00:39:14
     ◼ 
      
     ►  
     I think I'm gonna do it again. 
     
     
  
 
 
	 00:39:16
     ◼ 
      
     ►  
     I think it's part of the human experience, I do. 
     
     
  
 
 
	 00:39:17
     ◼ 
      
     ►  
     And it doesn't need to be that everybody becomes a painter, 
     
     
  
 
 
	 00:39:20
     ◼ 
      
     ►  
     but we all have these things in our lives 
     
     
  
 
 
	 00:39:22
     ◼ 
      
     ►  
     that we practice and get better at, right? 
     
     
  
 
 
	 00:39:25
     ◼ 
      
     ►  
     It's like, why don't I just give my game controller 
     
     
  
 
 
	 00:39:28
     ◼ 
      
     ►  
     to a robot and then just watch what it does? 
     
     
  
 
 
 
	 00:39:32
     ◼ 
      
     ►  
     - That's where we're going. 
     
     
  
 
 
	 00:39:33
     ◼ 
      
     ►  
     It's like "WALL-E", right? 
     
     
  
 
 
	 00:39:36
     ◼ 
      
     ►  
     This is how we get to "WALL-E". 
     
     
  
 
 
	 00:39:37
     ◼ 
      
     ►  
     This is how "WALL-E" happens, 
     
     
  
 
 
	 00:39:40
     ◼ 
      
     ►  
     that we're all just sitting in these chairs. 
     
     
  
 
 
	 00:39:42
     ◼ 
      
     ►  
     Like that's, right now is the beginning 
     
     
  
 
 
	 00:39:45
     ◼ 
      
     ►  
     of the path to "WALL-E". 
     
     
  
 
 
	 00:39:47
     ◼ 
      
     ►  
     - Well, "WALL-E" if we're lucky, but yes. 
     
     
  
 
 
 
	 00:39:50
     ◼ 
      
     ►  
     - You know, it's just like, 
     
     
  
 
 
	 00:39:52
     ◼ 
      
     ►  
     I feel like we have to do things. 
     
     
  
 
 
	 00:39:54
     ◼ 
      
     ►  
     whatever it is, some kind of thing 
     
     
  
 
 
	 00:39:57
     ◼ 
      
     ►  
     that we enjoy the process of. 
     
     
  
 
 
	 00:39:59
     ◼ 
      
     ►  
     And I feel like if all we're doing is saying 
     
     
  
 
 
	 00:40:02
     ◼ 
      
     ►  
     that this stuff is just gonna replace filmmaking, 
     
     
  
 
 
	 00:40:05
     ◼ 
      
     ►  
     it's just like, I don't wanna watch those movies. 
     
     
  
 
 
	 00:40:08
     ◼ 
      
     ►  
     I really don't, but I like to be sold in something. 
     
     
  
 
 
	 00:40:11
     ◼ 
      
     ►  
     I like to believe that a human was involved 
     
     
  
 
 
	 00:40:13
     ◼ 
      
     ►  
     in the endeavors that I'm consuming. 
     
     
  
 
 
	 00:40:15
     ◼ 
      
     ►  
     Like, but again, maybe I'm old fashioned 
     
     
  
 
 
	 00:40:18
     ◼ 
      
     ►  
     and I'm fine to accept that, 
     
     
  
 
 
	 00:40:20
     ◼ 
      
     ►  
     but I'm just trying to get across 
     
     
  
 
 
	 00:40:22
     ◼ 
      
     ►  
     like why these things are important to me, 
     
     
  
 
 
	 00:40:24
     ◼ 
      
     ►  
     why I am so passionate about it. 
     
     
  
 
 
	 00:40:29
     ◼ 
      
     ►  
     Yeah, I sort of have a minor point and a major point 
     
     
  
 
 
	 00:40:33
     ◼ 
      
     ►  
     on that topic. 
     
     
  
 
 
	 00:40:34
     ◼ 
      
     ►  
     I think last time, I made some offhanded remark of, 
     
     
  
 
 
	 00:40:37
     ◼ 
      
     ►  
     like, very soon in major productions 
     
     
  
 
 
	 00:40:40
     ◼ 
      
     ►  
     where we're going to see stuff that's AI generated 
     
     
  
 
 
	 00:40:43
     ◼ 
      
     ►  
     and not know it. 
     
     
  
 
 
	 00:40:45
     ◼ 
      
     ►  
     And I didn't realize at the time, 
     
     
  
 
 
	 00:40:47
     ◼ 
      
     ►  
     but like, oh, that was already true. 
     
     
  
 
 
 
	 00:40:50
     ◼ 
      
     ►  
     Didn't know about this. 
     
     
  
 
 
	 00:40:51
     ◼ 
      
     ►  
     Yeah, I had watched, uh, the Obi-Wan Kenobi show on Disney and I didn't really think about it, 
     
     
  
 
 
	 00:40:58
     ◼ 
      
     ►  
     but when I had watched Rogue One, I was very aware of listening to James Earl Jones do Darth Vader's 
     
     
  
 
 
	 00:41:05
     ◼ 
      
     ►  
     lines of like, "He's still Darth Vader, but he's getting too old for this." Like, you can just hear 
     
     
  
 
 
	 00:41:12
     ◼ 
      
     ►  
     it in someone's voice that they're just older. That's just what happens, uh, as your vocal 
     
     
  
 
 
	 00:41:16
     ◼ 
      
     ►  
     chords physically change. And when I watched Obi-Wan Kenobi, I never thought about it. 
     
     
  
 
 
	 00:41:21
     ◼ 
      
     ►  
     And I also didn't think about the fact that I didn't think about it. And I realized, since 
     
     
  
 
 
	 00:41:26
     ◼ 
      
     ►  
     we recorded that episode, oh, I didn't think about it, because all of his lines were done 
     
     
  
 
 
	 00:41:31
     ◼ 
      
     ►  
     with AI. There's a program called Respeacher that will take a voice actor's lines and redo 
     
     
  
 
 
	 00:41:39
     ◼ 
      
     ►  
     them in the voice of someone else. And what's really quite remarkable about it is it is 
     
     
  
 
 
	 00:41:46
     ◼ 
      
     ►  
     isn't just what you sort of think like voice modulators on the phone like, oh, I you know, 
     
     
  
 
 
	 00:41:51
     ◼ 
      
     ►  
     I can increase or decrease my pitch, but it still sounds like me. Everybody has vocal 
     
     
  
 
 
	 00:41:56
     ◼ 
      
     ►  
     tics and things like no, no, no re speech or will put in the vocal tics of the other 
     
     
  
 
 
	 00:42:01
     ◼ 
      
     ►  
     person like it's not just making your words sound in their voice. It's making them sound 
     
     
  
 
 
	 00:42:08
     ◼ 
      
     ►  
     like them. Just as an interesting thing to note, though, I did I did go back on YouTube 
     
     
  
 
 
	 00:42:13
     ◼ 
      
     ►  
     and I watched some of the original line deliveries of James Earl Jones in the original Star Wars. 
     
     
  
 
 
	 00:42:17
     ◼ 
      
     ►  
     I was like, "Ooh, he is better here!" 
     
     
  
 
 
	 00:42:20
     ◼ 
      
     ►  
     Like he has like funny little things that he does with a bunch of words that make those 
     
     
  
 
 
	 00:42:23
     ◼ 
      
     ►  
     line deliveries really great. 
     
     
  
 
 
	 00:42:25
     ◼ 
      
     ►  
     And there isn't as much of that in the Obi-Wan Kenobi show. 
     
     
  
 
 
	 00:42:28
     ◼ 
      
     ►  
     I will say when I watched Obi-Wan, there was something about Darth Vader's performance 
     
     
  
 
 
	 00:42:33
     ◼ 
      
     ►  
     that I didn't like. 
     
     
  
 
 
	 00:42:35
     ◼ 
      
     ►  
     And I assumed it was because James Earl Jones was getting older. 
     
     
  
 
 
	 00:42:40
     ◼ 
      
     ►  
     But it turned out it wasn't the case. 
     
     
  
 
 
	 00:42:42
     ◼ 
      
     ►  
     But there was something that felt missing to me, 
     
     
  
 
 
	 00:42:45
     ◼ 
      
     ►  
     and I just put it down to, 
     
     
  
 
 
	 00:42:46
     ◼ 
      
     ►  
     and I remember saying to Adina at the time, 
     
     
  
 
 
	 00:42:49
     ◼ 
      
     ►  
     "I can't believe they're still getting this guy to do this. 
     
     
  
 
 
	 00:42:51
     ◼ 
      
     ►  
     Why don't they just get like a voice actor?" 
     
     
  
 
 
 
	 00:42:55
     ◼ 
      
     ►  
     - Right, leave James Earl Jones alone. 
     
     
  
 
 
	 00:42:57
     ◼ 
      
     ►  
     - Yeah, well, it was just kind of just like, 
     
     
  
 
 
	 00:43:00
     ◼ 
      
     ►  
     this isn't gonna last forever with him, 
     
     
  
 
 
	 00:43:03
     ◼ 
      
     ►  
     so like they need a path. 
     
     
  
 
 
	 00:43:05
     ◼ 
      
     ►  
     And they created one, and it did a fine job, 
     
     
  
 
 
	 00:43:07
     ◼ 
      
     ►  
     but it still felt like it was missing something to me. 
     
     
  
 
 
	 00:43:10
     ◼ 
      
     ►  
     I can't put my finger on what it was. 
     
     
  
 
 
	 00:43:11
     ◼ 
      
     ►  
     It just didn't feel right. 
     
     
  
 
 
	 00:43:13
     ◼ 
      
     ►  
     I don't know if this was part, whatever. 
     
     
  
 
 
	 00:43:14
     ◼ 
      
     ►  
     But what I will say, this particular implementation of AI, 
     
     
  
 
 
	 00:43:19
     ◼ 
      
     ►  
     I'm fine with, because there are practical reasons for it. 
     
     
  
 
 
	 00:43:22
     ◼ 
      
     ►  
     Of like, if we want Darth Vader to sound the same, 
     
     
  
 
 
	 00:43:26
     ◼ 
      
     ►  
     which I think ideally we do, right? 
     
     
  
 
 
	 00:43:28
     ◼ 
      
     ►  
     Like you could just get someone else to do it, 
     
     
  
 
 
	 00:43:29
     ◼ 
      
     ►  
     but I would prefer it if he still sounded 
     
     
  
 
 
	 00:43:31
     ◼ 
      
     ►  
     like James Earl Jones, 
     
     
  
 
 
	 00:43:32
     ◼ 
      
     ►  
     'cause it's like an existing character. 
     
     
  
 
 
	 00:43:34
     ◼ 
      
     ►  
     And James Earl Jones signed this away 
     
     
  
 
 
	 00:43:37
     ◼ 
      
     ►  
     as an individual while still alive. 
     
     
  
 
 
	 00:43:39
     ◼ 
      
     ►  
     - Right, which is key. 
     
     
  
 
 
	 00:43:40
     ◼ 
      
     ►  
     you can do this, you can take my voice and do whatever you want 
     
     
  
 
 
	 00:43:44
     ◼ 
      
     ►  
     plus working voice actors continue to get jobs 
     
     
  
 
 
	 00:43:48
     ◼ 
      
     ►  
     being Darth Vader then they change the voice. For me there are enough 
     
     
  
 
 
	 00:43:52
     ◼ 
      
     ►  
     pieces of this puzzle where I'm like I am fine with this 
     
     
  
 
 
	 00:43:56
     ◼ 
      
     ►  
     because people are involved in it. It's if Disney just were like 
     
     
  
 
 
	 00:44:03
     ◼ 
      
     ►  
     we've just decided we don't want to hire him anymore 
     
     
  
 
 
	 00:44:06
     ◼ 
      
     ►  
     and we've created a thing and we're just gonna type 
     
     
  
 
 
	 00:44:10
     ◼ 
      
     ►  
     some text in and the AI's gonna spit out 
     
     
  
 
 
	 00:44:13
     ◼ 
      
     ►  
     the Darth Vader lines, I will be like, 
     
     
  
 
 
	 00:44:15
     ◼ 
      
     ►  
     "Mm, I don't like that, that doesn't feel good to me." 
     
     
  
 
 
	 00:44:18
     ◼ 
      
     ►  
     But the way they have done this whole thing going around, 
     
     
  
 
 
	 00:44:21
     ◼ 
      
     ►  
     I'm kind of fine with. 
     
     
  
 
 
	 00:44:23
     ◼ 
      
     ►  
     Like in the same way that I do find it kind of funny really, 
     
     
  
 
 
	 00:44:26
     ◼ 
      
     ►  
     that they continue to have Mark Hamill on set 
     
     
  
 
 
	 00:44:30
     ◼ 
      
     ►  
     during the Mandalorian stuff to be Luke Skywalker. 
     
     
  
 
 
	 00:44:35
     ◼ 
      
     ►  
     But then they completely digitally replace him with the younger version of him. 
     
     
  
 
 
	 00:44:39
     ◼ 
      
     ►  
     Well they do multiple things. They have him do a thing, then they have a look-alike actor 
     
     
  
 
 
	 00:44:43
     ◼ 
      
     ►  
     do a thing, and then they put a digital recreation of his face on the young actor's face. 
     
     
  
 
 
	 00:44:48
     ◼ 
      
     ►  
     "Really there is no point in him being here, right? We just like to have him around? I 
     
     
  
 
 
	 00:44:53
     ◼ 
      
     ►  
     don't know, right?" 
     
     
  
 
 
	 00:44:54
     ◼ 
      
     ►  
     But like, there is something funny to me about that, but at the same time, I'm like, at 
     
     
  
 
 
	 00:44:58
     ◼ 
      
     ►  
     least it is a respect of the person they are digitally recreating. Everyone is in on this 
     
     
  
 
 
	 00:45:04
     ◼ 
      
     ►  
     that's fine. You know? 
     
     
  
 
 
	 00:45:07
     ◼ 
      
     ►  
     I suspect that they're also using re-speacher to make Mark Hamill's voice sound like young 
     
     
  
 
 
	 00:45:12
     ◼ 
      
     ►  
     Mark Hamill, because that's again, like, he does not sound like he did when he was in 
     
     
  
 
 
 
	 00:45:16
     ◼ 
      
     ►  
     But he also doesn't sound like him now either, right? So they're doing something. 
     
     
  
 
 
	 00:45:20
     ◼ 
      
     ►  
     Yeah, something is happening there. But this is also one of these key things of, like, 
     
     
  
 
 
	 00:45:24
     ◼ 
      
     ►  
     okay, it's interesting to realize I had already watched an entire show where a major part 
     
     
  
 
 
	 00:45:28
     ◼ 
      
     ►  
     of it was AI created and I didn't even notice, or it didn't even cross my mind. But it also 
     
     
  
 
 
	 00:45:34
     ◼ 
      
     ►  
     touches on like what you're saying here is some of the key differences between what's 
     
     
  
 
 
	 00:45:40
     ◼ 
      
     ►  
     happening in different parts of the world. 
     
     
  
 
 
	 00:45:43
     ◼ 
      
     ►  
     And a lot of the AI art developments on the internet are just like a like a crazy Wild 
     
     
  
 
 
	 00:45:49
     ◼ 
      
     ►  
     West where people are just grabbing whatever they can. 
     
     
  
 
 
	 00:45:52
     ◼ 
      
     ►  
     And it's like, that makes me extremely uncomfortable. 
     
     
  
 
 
	 00:45:55
     ◼ 
      
     ►  
     Like, okay, so again, I will just say in general ways, I saw a demo of a Siri like voice assistant 
     
     
  
 
 
	 00:46:03
     ◼ 
      
     ►  
     that was significantly better than things that currently exist. 
     
     
  
 
 
	 00:46:07
     ◼ 
      
     ►  
     Oh, that's— it was very interesting, but it was also using the voice of a famous actress 
     
     
  
 
 
	 00:46:14
     ◼ 
      
     ►  
     just in their demo mode. 
     
     
  
 
 
	 00:46:16
     ◼ 
      
     ►  
     And I got very uncomfortable about that, 'cause it's like… 
     
     
  
 
 
	 00:46:21
     ◼ 
      
     ►  
     She's a— she's a person who doesn't know that you've made a machine that can make her 
     
     
  
 
 
	 00:46:26
     ◼ 
      
     ►  
     say anything. 
     
     
  
 
 
 
 
	 00:46:29
     ◼ 
      
     ►  
     That's what we want! 
     
     
  
 
 
 
	 00:46:32
     ◼ 
      
     ►  
     Yeah, and it's like, "No, no, no!" And so a lot of the AI art stuff just feels like that, 
     
     
  
 
 
	 00:46:37
     ◼ 
      
     ►  
     like it's just grabbed everything that exists. And the thing that I keep thinking of is, 
     
     
  
 
 
	 00:46:44
     ◼ 
      
     ►  
     like it makes a marionette of everyone who's ever put out any kind of art in the public. 
     
     
  
 
 
	 00:46:52
     ◼ 
      
     ►  
     And that's like a horror, right? That's just a completely horrifying concept that, 
     
     
  
 
 
	 00:47:00
     ◼ 
      
     ►  
     if you have a social media timeline where you've posted stuff, guess what? 
     
     
  
 
 
	 00:47:05
     ◼ 
      
     ►  
     There's enough information where someone can make a marionette of you, and it can do anything. 
     
     
  
 
 
	 00:47:10
     ◼ 
      
     ►  
     And right now, like, this can happen without someone's permission. 
     
     
  
 
 
	 00:47:13
     ◼ 
      
     ►  
     And yeah, it's totally why, like, the Darth Vader stuff I have no problem with, 
     
     
  
 
 
	 00:47:17
     ◼ 
      
     ►  
     because presumably James Earl Jones was, like, fantastic. 
     
     
  
 
 
	 00:47:20
     ◼ 
      
     ►  
     - I saw an article with that. That was, like, I read about that. 
     
     
  
 
 
	 00:47:22
     ◼ 
      
     ►  
     It was the case that he approved this, signed the rights, like, he's good with it. 
     
     
  
 
 
	 00:47:26
     ◼ 
      
     ►  
     There's a big distinction between giving permission for something like this to be done 
     
     
  
 
 
	 00:47:32
     ◼ 
      
     ►  
     and just the power of AI to make a marionette of anyone. 
     
     
  
 
 
	 00:47:39
     ◼ 
      
     ►  
     And there's something really horrifying about that. 
     
     
  
 
 
	 00:47:42
     ◼ 
      
     ►  
     Now I think it becomes, like other people brought this up as a kind of question, there's 
     
     
  
 
 
	 00:47:48
     ◼ 
      
     ►  
     a boundary that's crossed between doing it to a living person, which is super bad. 
     
     
  
 
 
	 00:47:54
     ◼ 
      
     ►  
     It's like, yeah, I just think that's really awful and immoral. It's less bad when the person's dead, 
     
     
  
 
 
	 00:48:02
     ◼ 
      
     ►  
     but it can still be bad if it's recent. I don't quite know what recent means in this context, but 
     
     
  
 
 
	 00:48:09
     ◼ 
      
     ►  
     clearly when we get to a point where we're talking about modifying Leonardo da Vinci's art, 
     
     
  
 
 
	 00:48:18
     ◼ 
      
     ►  
     I don't think there's anything bad there about like making a marionette of Leonardo da Vinci, 
     
     
  
 
 
	 00:48:23
     ◼ 
      
     ►  
     Making a machine that can spit out a ton of Leonardo da Vinci style paintings 
     
     
  
 
 
	 00:48:27
     ◼ 
      
     ►  
     I don't think it's possible to ascribe some kind of rule onto this 
     
     
  
 
 
	 00:48:31
     ◼ 
      
     ►  
     You just feel it when you feel it like I can't tell you what the right time period is 
     
     
  
 
 
	 00:48:37
     ◼ 
      
     ►  
     and it's also different in every case and 
     
     
  
 
 
	 00:48:39
     ◼ 
      
     ►  
     You know goes on and on forever like 
     
     
  
 
 
	 00:48:44
     ◼ 
      
     ►  
     The fast and furious movie right where Paul Walker had just died, right and it was like months later when the movie came out 
     
     
  
 
 
	 00:48:53
     ◼ 
      
     ►  
     out but they used some digital recreation of his face to be able to 
     
     
  
 
 
	 00:48:57
     ◼ 
      
     ►  
     give him like a tribute and send-off. And I think that there is a 
     
     
  
 
 
	 00:49:02
     ◼ 
      
     ►  
     little like it's awkward I think that movie came out too soon really but the 
     
     
  
 
 
	 00:49:08
     ◼ 
      
     ►  
     whole family kind of seemed to agree and there's something about like that was 
     
     
  
 
 
	 00:49:12
     ◼ 
      
     ►  
     the thing. Same as like Princess Leia right? Yeah that's what I was thinking. 
     
     
  
 
 
	 00:49:16
     ◼ 
      
     ►  
     Princess Leia's send-off was the next one after Carrie Fisher had died. 
     
     
  
 
 
	 00:49:20
     ◼ 
      
     ►  
     Whereas still I feel like we're on a fine line here because it's weird. 
     
     
  
 
 
	 00:49:24
     ◼ 
      
     ►  
     But it was recreated from stuff that she did participate in shooting and at least there's 
     
     
  
 
 
	 00:49:29
     ◼ 
      
     ►  
     like a tribute. 
     
     
  
 
 
	 00:49:31
     ◼ 
      
     ►  
     But then if like Princess Leia just continued being a character after that, be like no you've 
     
     
  
 
 
	 00:49:35
     ◼ 
      
     ►  
     pushed it too far now. 
     
     
  
 
 
 
	 00:49:37
     ◼ 
      
     ►  
     Because now that's the making of the marionette right? 
     
     
  
 
 
	 00:49:40
     ◼ 
      
     ►  
     She's dead now but you're continuing to use her. 
     
     
  
 
 
	 00:49:43
     ◼ 
      
     ►  
     That would be weird. 
     
     
  
 
 
	 00:49:46
     ◼ 
      
     ►  
     I don't think it's possible to put rules on the death of a person. I just know that I 
     
     
  
 
 
	 00:49:52
     ◼ 
      
     ►  
     don't find it particularly comfortable for someone to profit off, either financially 
     
     
  
 
 
	 00:49:59
     ◼ 
      
     ►  
     or publicity-wise, somebody who is dead. Like taking their actual work, and again, divorcing 
     
     
  
 
 
	 00:50:07
     ◼ 
      
     ►  
     the work required in the copying, right? And just straight up, like, "I took this thing 
     
     
  
 
 
	 00:50:14
     ◼ 
      
     ►  
     and I made this thing out of it. 
     
     
  
 
 
	 00:50:16
     ◼ 
      
     ►  
     And I didn't really have to do anything 
     
     
  
 
 
	 00:50:19
     ◼ 
      
     ►  
     other than put this image into the AI. 
     
     
  
 
 
	 00:50:23
     ◼ 
      
     ►  
     There's uncomfortableness for me in that. 
     
     
  
 
 
	 00:50:25
     ◼ 
      
     ►  
     - Yeah, I just wanted to highlight that there's a... 
     
     
  
 
 
	 00:50:28
     ◼ 
      
     ►  
     Like with many discussions, you can run into areas 
     
     
  
 
 
	 00:50:33
     ◼ 
      
     ►  
     where there's some kind of spectrum. 
     
     
  
 
 
	 00:50:34
     ◼ 
      
     ►  
     - Yeah, like someone taking the works of Shakespeare 
     
     
  
 
 
	 00:50:37
     ◼ 
      
     ►  
     and putting that into an AI and making more Shakespeare, 
     
     
  
 
 
	 00:50:39
     ◼ 
      
     ►  
     I don't care about that. 
     
     
  
 
 
	 00:50:40
     ◼ 
      
     ►  
     But I can't tell you why I don't care about it either. 
     
     
  
 
 
	 00:50:43
     ◼ 
      
     ►  
     I just don't. 
     
     
  
 
 
	 00:50:44
     ◼ 
      
     ►  
     Yeah, I mean, I do want to also add that this, this dance is around a concept 
     
     
  
 
 
	 00:50:49
     ◼ 
      
     ►  
     that I've been, I've been thinking about for years and I have, I have a hard time 
     
     
  
 
 
	 00:50:52
     ◼ 
      
     ►  
     articulating, but for now I will just call it as the importance of saying no in art. 
     
     
  
 
 
	 00:50:58
     ◼ 
      
     ►  
     And I think a lot of artistic projects or the life of creative people are defined as 
     
     
  
 
 
	 00:51:08
     ◼ 
      
     ►  
     much by what they did as what they didn't do. 
     
     
  
 
 
	 00:51:10
     ◼ 
      
     ►  
     A thousand no's for every yes. 
     
     
  
 
 
	 00:51:12
     ◼ 
      
     ►  
     Yeah, there is a way to put it, but I think that having spoken to creative people, a lot 
     
     
  
 
 
	 00:51:19
     ◼ 
      
     ►  
     of them have some internal set of things that they don't do that's not obvious in the 
     
     
  
 
 
	 00:51:27
     ◼ 
      
     ►  
     work that they create, but it lends a kind of character to their work. 
     
     
  
 
 
	 00:51:31
     ◼ 
      
     ►  
     I've also just recently went through this where I killed a what was going to probably 
     
     
  
 
 
	 00:51:37
     ◼ 
      
     ►  
     be like a 20 or 30 minute video that was all storyboarded and written out. 
     
     
  
 
 
	 00:51:42
     ◼ 
      
     ►  
     And part of the reason I killed it is because I realized this was a project to say no to, 
     
     
  
 
 
	 00:51:48
     ◼ 
      
     ►  
     that like, yes, I could put it out and I think people would like it, but it violated a couple 
     
     
  
 
 
	 00:51:54
     ◼ 
      
     ►  
     of my own internal, like, I don't want to do this sort of thing in this way, even if 
     
     
  
 
 
	 00:52:01
     ◼ 
      
     ►  
     no one in the audience would have noticed. 
     
     
  
 
 
	 00:52:05
     ◼ 
      
     ►  
     As like, man, that was a hard decision. 
     
     
  
 
 
	 00:52:06
     ◼ 
      
     ►  
     But I think that that's what matters. 
     
     
  
 
 
	 00:52:09
     ◼ 
      
     ►  
     And so even while I think that there is there's nothing morally wrong about making new Shakespeare 
     
     
  
 
 
	 00:52:16
     ◼ 
      
     ►  
     plays or new Leonardo da Vinci paintings with AI, I do think that there's something about 
     
     
  
 
 
	 00:52:23
     ◼ 
      
     ►  
     this concept of making an artist do anything that you want demeans the limits that that 
     
     
  
 
 
	 00:52:35
     ◼ 
      
     ►  
     person put on their own things. 
     
     
  
 
 
	 00:52:37
     ◼ 
      
     ►  
     - And I'm also like not sure why you even want it. 
     
     
  
 
 
	 00:52:41
     ◼ 
      
     ►  
     - Well, I think this is also just a side effect 
     
     
  
 
 
	 00:52:43
     ◼ 
      
     ►  
     of how easy it can be and will be to create this stuff. 
     
     
  
 
 
	 00:52:46
     ◼ 
      
     ►  
     Where just like out of curiosity, you type some prompts in. 
     
     
  
 
 
	 00:52:50
     ◼ 
      
     ►  
     I guess-- - Curiosity is different 
     
     
  
 
 
	 00:52:52
     ◼ 
      
     ►  
     though, right? - But I mean, it's like 
     
     
  
 
 
	 00:52:53
     ◼ 
      
     ►  
     when this gets good enough that it can just like 
     
     
  
 
 
	 00:52:55
     ◼ 
      
     ►  
     make a movie where some things happen. 
     
     
  
 
 
	 00:52:57
     ◼ 
      
     ►  
     I don't know, this idea of the importance of saying now 
     
     
  
 
 
	 00:53:00
     ◼ 
      
     ►  
     is also connected to a thing that I see with a lot of, 
     
     
  
 
 
	 00:53:02
     ◼ 
      
     ►  
     particularly TV shows and long running creative projects, 
     
     
  
 
 
	 00:53:07
     ◼ 
      
     ►  
     where I think of this as you can see the audience influence the creators that never works. 
     
     
  
 
 
	 00:53:14
     ◼ 
      
     ►  
     And there's lots of shows where it's like, "Oh, this is the moment. I can see now that the 
     
     
  
 
 
	 00:53:19
     ◼ 
      
     ►  
     creators are aware of the audience." And with creative projects, the audience will ask for 
     
     
  
 
 
	 00:53:27
     ◼ 
      
     ►  
     things that they want that they don't realize in the long run actually destroys or makes worse the 
     
     
  
 
 
	 00:53:34
     ◼ 
      
     ►  
     thing that they want. And like that's sort of the I think that can sound super snobbish, 
     
     
  
 
 
	 00:53:39
     ◼ 
      
     ►  
     but I think it's true. And I've seen it in enough projects was like the audience wants 
     
     
  
 
 
	 00:53:42
     ◼ 
      
     ►  
     something and the creator then goes, Oh, I'll do that thing and the audience responds and 
     
     
  
 
 
	 00:53:48
     ◼ 
      
     ►  
     you get into this little positive feedback loop of doing the things that the audience 
     
     
  
 
 
	 00:53:51
     ◼ 
      
     ►  
     wants and none of those decisions mattered individually, but cumulatively they can make 
     
     
  
 
 
	 00:53:57
     ◼ 
      
     ►  
     a thing much worse in a way that's hard to pin down. And I just think there's something 
     
     
  
 
 
	 00:54:02
     ◼ 
      
     ►  
     in AI art where even when there's no problem, I think people can like destroy the things 
     
     
  
 
 
	 00:54:10
     ◼ 
      
     ►  
     that they love because there's no one to say no. There's no creator who says, "These 
     
     
  
 
 
	 00:54:16
     ◼ 
      
     ►  
     are the limits of my thing." And again, I keep thinking of the Miyazaki movies as a 
     
     
  
 
 
	 00:54:22
     ◼ 
      
     ►  
     particular example of this. Like if you can make a Miyazaki movie about anything you want, 
     
     
  
 
 
	 00:54:28
     ◼ 
      
     ►  
     It kind of destroys what those movies are unless moved by the argument about it's important 
     
     
  
 
 
	 00:54:34
     ◼ 
      
     ►  
     for humans to do these sorts of things. 
     
     
  
 
 
	 00:54:37
     ◼ 
      
     ►  
     But I can easily imagine a situation where even if the AI is able to make amazing art, 
     
     
  
 
 
	 00:54:44
     ◼ 
      
     ►  
     it's actually kind of worse for everyone involved, even though it's a thing that the audience 
     
     
  
 
 
	 00:54:50
     ◼ 
      
     ►  
     has asked for or that people go like, Oh man, I wish I could continue the series of movies 
     
     
  
 
 
	 00:54:55
     ◼ 
      
     ►  
     forever and I can do it by typing into the machine and it will make the movie for me. 
     
     
  
 
 
	 00:54:58
     ◼ 
      
     ►  
     It's like you'll ruin the thing you love by doing that, by getting what you want all the 
     
     
  
 
 
	 00:55:03
     ◼ 
      
     ►  
     time without a creative mind to say like, "No, this is the limits of this project" or "No, 
     
     
  
 
 
	 00:55:10
     ◼ 
      
     ►  
     this is when the story stopped." 
     
     
  
 
 
	 00:55:11
     ◼ 
      
     ►  
     Yeah, I mostly agree with that statement. Like, we've spoken about this a lot, right? 
     
     
  
 
 
	 00:55:17
     ◼ 
      
     ►  
     Like that idea of the point where the artist becomes aware of the audience. We speak about 
     
     
  
 
 
	 00:55:24
     ◼ 
      
     ►  
     this a lot in regards to like just personally TV show recommendations and like you know 
     
     
  
 
 
	 00:55:29
     ◼ 
      
     ►  
     like I know you are very sensitive to this like it's something that I'm familiar with 
     
     
  
 
 
	 00:55:33
     ◼ 
      
     ►  
     like you point this out and I'm like oh okay this is an interesting idea for why the show 
     
     
  
 
 
	 00:55:37
     ◼ 
      
     ►  
     may have gotten bad in that season like it's not really something I would have thought 
     
     
  
 
 
	 00:55:40
     ◼ 
      
     ►  
     of and I will say that I mostly agree with your thinking here but not completely I think 
     
     
  
 
 
	 00:55:48
     ◼ 
      
     ►  
     my kind of interpretation of this idea that you have is like, just to add the word like, 
     
     
  
 
 
	 00:55:55
     ◼ 
      
     ►  
     mostly to it. 
     
     
  
 
 
 
 
	 00:55:58
     ◼ 
      
     ►  
     Like that I don't, and I'm not saying that you say this, but the way that it sounds makes 
     
     
  
 
 
	 00:56:01
     ◼ 
      
     ►  
     it sound like a definitive, like that listening to the audience or seeing what the audience 
     
     
  
 
 
	 00:56:07
     ◼ 
      
     ►  
     want can never make something better, but I don't think it's that way, and I'm not sure 
     
     
  
 
 
	 00:56:12
     ◼ 
      
     ►  
     you think it's that way, but I just want to state that, right? 
     
     
  
 
 
 
 
	 00:56:17
     ◼ 
      
     ►  
     there is definitely feedback which helps make something better and I do this all 
     
     
  
 
 
	 00:56:23
     ◼ 
      
     ►  
     the time but there are sometimes people will ask for me to talk about a thing 
     
     
  
 
 
	 00:56:28
     ◼ 
      
     ►  
     and I just know it won't be interesting because I'm not that person to do that 
     
     
  
 
 
	 00:56:35
     ◼ 
      
     ►  
     right and that that tends to be a lot of it for me of like you're telling me you 
     
     
  
 
 
	 00:56:39
     ◼ 
      
     ►  
     want a thing I know you won't want it because I know it's not gonna be 
     
     
  
 
 
	 00:56:43
     ◼ 
      
     ►  
     interesting because either I don't find it interesting or I don't have the 
     
     
  
 
 
	 00:56:49
     ◼ 
      
     ►  
     knowledge or whatever it might be you know like that there is that part of the 
     
     
  
 
 
	 00:56:53
     ◼ 
      
     ►  
     artist of oh god I just called myself an artist there is that part of the 
     
     
  
 
 
	 00:56:58
     ◼ 
      
     ►  
     creative person right where they are aware of what they're good at and what 
     
     
  
 
 
	 00:57:03
     ◼ 
      
     ►  
     they're not and what they think will be ultimately enjoyable for this type of 
     
     
  
 
 
	 00:57:08
     ◼ 
      
     ►  
     stuff that they create yeah and then they go out and make it obviously like 
     
     
  
 
 
	 00:57:12
     ◼ 
      
     ►  
     this is a different kind of thing. 
     
     
  
 
 
	 00:57:14
     ◼ 
      
     ►  
     I also think it's the bounds of the project. 
     
     
  
 
 
	 00:57:18
     ◼ 
      
     ►  
     And I found, I just recently stumbled upon to me what is just the most perfect 
     
     
  
 
 
	 00:57:24
     ◼ 
      
     ►  
     example of the artist saying no, which is also a little bit heartbreaking for 
     
     
  
 
 
	 00:57:30
     ◼ 
      
     ►  
     reasons that will be obvious when I, when I mentioned who it is, J.R.R. 
     
     
  
 
 
 
	 00:57:34
     ◼ 
      
     ►  
     I just discovered, I can't believe I never knew this. 
     
     
  
 
 
	 00:57:37
     ◼ 
      
     ►  
     He started a book that was set after the Lord of the Rings. 
     
     
  
 
 
	 00:57:42
     ◼ 
      
     ►  
     And so he wrote the first couple of chapters of this story. 
     
     
  
 
 
	 00:57:48
     ◼ 
      
     ►  
     And it's quite interesting, but like he, 
     
     
  
 
 
	 00:57:51
     ◼ 
      
     ►  
     basically he then wrote about, 
     
     
  
 
 
	 00:57:56
     ◼ 
      
     ►  
     he wrote in one of his letters, 
     
     
  
 
 
	 00:57:57
     ◼ 
      
     ►  
     you can get these books of all of his private letters 
     
     
  
 
 
	 00:57:59
     ◼ 
      
     ►  
     where he discusses a bunch of his thoughts. 
     
     
  
 
 
	 00:58:01
     ◼ 
      
     ►  
     He wrote about why he didn't finish this. 
     
     
  
 
 
	 00:58:04
     ◼ 
      
     ►  
     And the reason he didn't finish it 
     
     
  
 
 
	 00:58:06
     ◼ 
      
     ►  
     is because, oh, in his timeline after The Lord of the Rings, 
     
     
  
 
 
	 00:58:10
     ◼ 
      
     ►  
     there's not really much magic left in the world. 
     
     
  
 
 
	 00:58:14
     ◼ 
      
     ►  
     He had an idea of who the story would be about, 
     
     
  
 
 
	 00:58:16
     ◼ 
      
     ►  
     but basically everybody magic is gone, 
     
     
  
 
 
	 00:58:18
     ◼ 
      
     ►  
     and it's just the world of men. 
     
     
  
 
 
	 00:58:20
     ◼ 
      
     ►  
     And he said, "Oh, I could have written 
     
     
  
 
 
	 00:58:22
     ◼ 
      
     ►  
     "a totally fine adventure book, 
     
     
  
 
 
	 00:58:26
     ◼ 
      
     ►  
     "but that's not what the world 
     
     
  
 
 
	 00:58:28
     ◼ 
      
     ►  
     "I was trying to create was about." 
     
     
  
 
 
	 00:58:31
     ◼ 
      
     ►  
     He was trying to create this mythology. 
     
     
  
 
 
	 00:58:34
     ◼ 
      
     ►  
     And writing another story that's what happens 
     
     
  
 
 
	 00:58:38
     ◼ 
      
     ►  
     after the mythology, he was like, 
     
     
  
 
 
	 00:58:39
     ◼ 
      
     ►  
     "Wait, why am I doing this project? 
     
     
  
 
 
	 00:58:41
     ◼ 
      
     ►  
     This isn't really what I want to do. 
     
     
  
 
 
	 00:58:43
     ◼ 
      
     ►  
     Sure, we can continue what happens 
     
     
  
 
 
	 00:58:44
     ◼ 
      
     ►  
     with a bunch of these characters, but I'm gonna say no." 
     
     
  
 
 
	 00:58:47
     ◼ 
      
     ►  
     And so he just like stopped writing it and said, 
     
     
  
 
 
	 00:58:49
     ◼ 
      
     ►  
     "No, I'm not gonna write anymore 
     
     
  
 
 
	 00:58:51
     ◼ 
      
     ►  
     the stories of what happens after. 
     
     
  
 
 
	 00:58:52
     ◼ 
      
     ►  
     I'll just fill in some of the details 
     
     
  
 
 
	 00:58:54
     ◼ 
      
     ►  
     of what happened before." 
     
     
  
 
 
	 00:58:55
     ◼ 
      
     ►  
     And I think that's an amazing example 
     
     
  
 
 
	 00:58:57
     ◼ 
      
     ►  
     in reading his letter about it. 
     
     
  
 
 
	 00:58:58
     ◼ 
      
     ►  
     He's very self-aware. 
     
     
  
 
 
	 00:59:01
     ◼ 
      
     ►  
     Sure, the audience would love this. 
     
     
  
 
 
	 00:59:03
     ◼ 
      
     ►  
     Everybody's asking for like a sequel to Lord of the Rings because it was so great, 
     
     
  
 
 
	 00:59:07
     ◼ 
      
     ►  
     but it's also so great because he knew where to stop. 
     
     
  
 
 
	 00:59:10
     ◼ 
      
     ►  
     I just think that's an interesting example and it would just make me very sad if people 
     
     
  
 
 
	 00:59:15
     ◼ 
      
     ►  
     were typing into AI generator machines like, "Give me Tolkien's next book!" 
     
     
  
 
 
	 00:59:20
     ◼ 
      
     ►  
     Like, "Oh no, please don't. 
     
     
  
 
 
	 00:59:22
     ◼ 
      
     ►  
     Please don't do that. 
     
     
  
 
 
 
	 00:59:24
     ◼ 
      
     ►  
     This episode of Cortex is brought to you by Squarespace, the all-in-one platform for building 
     
     
  
 
 
	 00:59:29
     ◼ 
      
     ►  
     your brand and growing your business online. 
     
     
  
 
 
	 00:59:31
     ◼ 
      
     ►  
     You can stand out with a beautiful website, engage with your audience and sell your products, 
     
     
  
 
 
	 00:59:36
     ◼ 
      
     ►  
     services, the content that you create, whatever it is, Squarespace has got you covered. 
     
     
  
 
 
	 00:59:41
     ◼ 
      
     ►  
     No matter what type of website you're looking to make, whether it's something for a personal 
     
     
  
 
 
	 00:59:45
     ◼ 
      
     ►  
     event, maybe you have a business site that you want to create, it doesn't matter, Squarespace 
     
     
  
 
 
	 00:59:49
     ◼ 
      
     ►  
     can help you. 
     
     
  
 
 
	 00:59:50
     ◼ 
      
     ►  
     They've got you covered. 
     
     
  
 
 
	 00:59:51
     ◼ 
      
     ►  
     And it starts with one of their beautiful templates. 
     
     
  
 
 
	 00:59:54
     ◼ 
      
     ►  
     They're best in class, they're award winning. 
     
     
  
 
 
	 00:59:57
     ◼ 
      
     ►  
     It is as easy as browsing the type of site or category of business you want to make a 
     
     
  
 
 
	 01:00:01
     ◼ 
      
     ►  
     website for and that will give you the perfect starting place for you to then customise just 
     
     
  
 
 
	 01:00:05
     ◼ 
      
     ►  
     a few clicks and really make it feel like your own. 
     
     
  
 
 
	 01:00:09
     ◼ 
      
     ►  
     And then once your website expands and maybe you want to add that online store, you can 
     
     
  
 
 
	 01:00:13
     ◼ 
      
     ►  
     sell whatever you want, physical, digital goods, it doesn't matter. 
     
     
  
 
 
	 01:00:17
     ◼ 
      
     ►  
     They have all of the tools that you need to start selling online. 
     
     
  
 
 
	 01:00:20
     ◼ 
      
     ►  
     And then as your website grows and your visitors start pouring in, you can use insights to 
     
     
  
 
 
	 01:00:25
     ◼ 
      
     ►  
     grow your business. 
     
     
  
 
 
	 01:00:26
     ◼ 
      
     ►  
     If you want to know where your site visitors are coming from, where your sales are generating 
     
     
  
 
 
	 01:00:30
     ◼ 
      
     ►  
     from, which channels are most effective for you, you can analyse all of this in Squarespace. 
     
     
  
 
 
	 01:00:35
     ◼ 
      
     ►  
     Then when you have the data, you can improve your website and build a marketing strategy 
     
     
  
 
 
	 01:00:39
     ◼ 
      
     ►  
     based on your top keywords or the most popular products and content you have available. 
     
     
  
 
 
	 01:00:43
     ◼ 
      
     ►  
     I love Squarespace. I have been using Squarespace for nearly 15 years now for various projects 
     
     
  
 
 
	 01:00:50
     ◼ 
      
     ►  
     and I wouldn't go anywhere else. When I want to put something online, they're the first 
     
     
  
 
 
	 01:00:54
     ◼ 
      
     ►  
     place I go. I know how to use it, it's so easy, everything looks great on every type 
     
     
  
 
 
	 01:00:58
     ◼ 
      
     ►  
     the device, they have awesome apps for you to manage everything. It's the full package, 
     
     
  
 
 
	 01:01:03
     ◼ 
      
     ►  
     I love it and I'm confident you're going to love it too. But you don't have to just take 
     
     
  
 
 
	 01:01:07
     ◼ 
      
     ►  
     my word for it. Go to squarespace.com/cortex and you can sign up for a free trial, no credit 
     
     
  
 
 
	 01:01:12
     ◼ 
      
     ►  
     card required. You can go in and build your whole website. Then when you want to launch 
     
     
  
 
 
	 01:01:16
     ◼ 
      
     ►  
     it to the world, use the offer code "CORTEX" and you will get 10% off your first purchase 
     
     
  
 
 
	 01:01:20
     ◼ 
      
     ►  
     of a website or domain. That's squarespace.com/cortex and then when you sign up, use the offer code 
     
     
  
 
 
	 01:01:26
     ◼ 
      
     ►  
     Cortex to get that 10% off your first purchase and show you support for the show. 
     
     
  
 
 
	 01:01:30
     ◼ 
      
     ►  
     Our thanks to Squarespace for their continued support of Cortex and Relay FM. 
     
     
  
 
 
	 01:01:34
     ◼ 
      
     ►  
     Hey, do you want to talk about this podcast that you found here, Myke? 
     
     
  
 
 
	 01:01:38
     ◼ 
      
     ►  
     So I referenced why I'm not worried about my job specifically. This has been going around 
     
     
  
 
 
	 01:01:44
     ◼ 
      
     ►  
     a lot. It is an AI Steve Jobs on an AI Joe Rogan episode. And it is a full podcast created 
     
     
  
 
 
	 01:01:54
     ◼ 
      
     ►  
     by an AI. I've not really bothered to look into this because I actually think it might 
     
     
  
 
 
	 01:01:59
     ◼ 
      
     ►  
     be part of a publicity stunt. But nevertheless, it is a thing that exists, it's 20 minutes 
     
     
  
 
 
	 01:02:05
     ◼ 
      
     ►  
     long and you can listen to it and I've like skipped around in it and you know what? It 
     
     
  
 
 
	 01:02:09
     ◼ 
      
     ►  
     really sounds like Steve Jobs. Like there are points where you can hear that it's not 
     
     
  
 
 
	 01:02:13
     ◼ 
      
     ►  
     real right? But like and ultimately my takeaway from this is who wants this? Like genuinely 
     
     
  
 
 
	 01:02:19
     ◼ 
      
     ►  
     like who wants this? Who wants to hear what Steve Jobs might have said to a fake Joe Rogan? 
     
     
  
 
 
	 01:02:25
     ◼ 
      
     ►  
     For me, I don't feel concerned about my job because you can take a version of me and a version of you 
     
     
  
 
 
	 01:02:33
     ◼ 
      
     ►  
     and we can have them make podcasts forever, but I feel like if that's what you want, I don't know 
     
     
  
 
 
	 01:02:43
     ◼ 
      
     ►  
     how much you could enjoy the content because it's not real. It's not real conversations. 
     
     
  
 
 
	 01:02:48
     ◼ 
      
     ►  
     That's what I do for a living, is real conversations between real people about 
     
     
  
 
 
	 01:02:52
     ◼ 
      
     ►  
     things that they care about and my assumption is the majority of people that listen to my shows 
     
     
  
 
 
	 01:02:58
     ◼ 
      
     ►  
     want to hear that rather than let's imagine what to ai might be talking about instead for me that's 
     
     
  
 
 
	 01:03:06
     ◼ 
      
     ►  
     just so broken from what i imagine people want the content for and look if you are that person you 
     
     
  
 
 
	 01:03:14
     ◼ 
      
     ►  
     don't need to tell me of course i know you exist right like i'm sure that there are people that 
     
     
  
 
 
	 01:03:18
     ◼ 
      
     ►  
     would like to just have us on in the background so they could go to sleep and it's just like we're 
     
     
  
 
 
	 01:03:22
     ◼ 
      
     ►  
     We're just going to have an infinite amount of episodes forever. 
     
     
  
 
 
	 01:03:25
     ◼ 
      
     ►  
     But realistically, I don't imagine that that's going to be a thing that people 
     
     
  
 
 
	 01:03:32
     ◼ 
      
     ►  
     would genuinely care about enough in the way that people might care 
     
     
  
 
 
	 01:03:35
     ◼ 
      
     ►  
     about the content that I make. 
     
     
  
 
 
	 01:03:36
     ◼ 
      
     ►  
     So just coming across this one episode, it honestly made me, even though it 
     
     
  
 
 
	 01:03:42
     ◼ 
      
     ►  
     showed me it can be done, it made me feel more secure in my own profession. 
     
     
  
 
 
	 01:03:45
     ◼ 
      
     ►  
     But honestly, in the last episode, I wasn't worrying about my own profession. 
     
     
  
 
 
	 01:03:50
     ◼ 
      
     ►  
     It was more about the idea of creativity in people. 
     
     
  
 
 
	 01:03:53
     ◼ 
      
     ►  
     That's what I get more about. 
     
     
  
 
 
	 01:03:54
     ◼ 
      
     ►  
     - You said this to me and I took a listen. 
     
     
  
 
 
	 01:03:56
     ◼ 
      
     ►  
     I did the same thing. 
     
     
  
 
 
	 01:03:57
     ◼ 
      
     ►  
     I didn't listen to the whole thing all the way through. 
     
     
  
 
 
	 01:03:59
     ◼ 
      
     ►  
     I sort of skipped around. 
     
     
  
 
 
	 01:04:01
     ◼ 
      
     ►  
     I have a slightly different take on this sort of stuff, 
     
     
  
 
 
	 01:04:03
     ◼ 
      
     ►  
     which is it doesn't matter how good Steve Jobs 
     
     
  
 
 
	 01:04:07
     ◼ 
      
     ►  
     on Joe Rogan is to listen to. 
     
     
  
 
 
	 01:04:10
     ◼ 
      
     ►  
     It's more of just a demonstration of proof of concept. 
     
     
  
 
 
	 01:04:14
     ◼ 
      
     ►  
     This thing is possible. 
     
     
  
 
 
	 01:04:15
     ◼ 
      
     ►  
     And once you have a demonstration of proof of concept, 
     
     
  
 
 
 
	 01:04:18
     ◼ 
      
     ►  
     get better, they don't get worse. So this was just like the first real proof of concept 
     
     
  
 
 
	 01:04:24
     ◼ 
      
     ►  
     demonstration of two people having a podcast conversation. I do have to say, listening 
     
     
  
 
 
	 01:04:29
     ◼ 
      
     ►  
     to it, I think the Steve Jobs was less good, but like holy moly did it nail Joe Rogan's 
     
     
  
 
 
	 01:04:36
     ◼ 
      
     ►  
     way of talking. 
     
     
  
 
 
	 01:04:37
     ◼ 
      
     ►  
     That makes sense though, right? The amount of source material for Joe Rogan is almost 
     
     
  
 
 
	 01:04:41
     ◼ 
      
     ►  
     infinite really. 
     
     
  
 
 
	 01:04:42
     ◼ 
      
     ►  
     Yeah, I just thought like it's actually interesting to listen to because you can almost 
     
     
  
 
 
	 01:04:46
     ◼ 
      
     ►  
     hear the fact that there's what, 100 million hours of Joe Rogan talking and, you know, 
     
     
  
 
 
	 01:04:52
     ◼ 
      
     ►  
     the database for Steve Jobs is so much smaller. And you can hear that in the two voices. It's 
     
     
  
 
 
	 01:04:57
     ◼ 
      
     ►  
     like, especially because I think Rogan has a funny circular way of talking sometimes 
     
     
  
 
 
	 01:05:02
     ◼ 
      
     ►  
     that it's like, I would never not know that that wasn't him. If you told me it was Rogan, 
     
     
  
 
 
	 01:05:07
     ◼ 
      
     ►  
     but it was only just an AI Jobs, I would have believed it. But this to me is, again, a good 
     
     
  
 
 
	 01:05:13
     ◼ 
      
     ►  
     example of this thing where making a marionette of someone is just bad. It's like, "Hey 
     
     
  
 
 
	 01:05:19
     ◼ 
      
     ►  
     guys, whatever you think of Rogan, he's still a person and it's real bad to make 
     
     
  
 
 
	 01:05:23
     ◼ 
      
     ►  
     like a fake show where he's talking. And it's less bad for Steve Jobs because he 
     
     
  
 
 
	 01:05:31
     ◼ 
      
     ►  
     is dead, but it's still close enough that it makes me very uncomfortable." I just 
     
     
  
 
 
	 01:05:36
     ◼ 
      
     ►  
     think we're going to see an increasingly large number of these sorts of things where 
     
     
  
 
 
	 01:05:40
     ◼ 
      
     ►  
     people can make whatever they want and it's only going to become increasingly easy over 
     
     
  
 
 
	 01:05:45
     ◼ 
      
     ►  
     time and I don't know how this project was made in particular but yeah it's the first 
     
     
  
 
 
	 01:05:50
     ◼ 
      
     ►  
     example of oh it's an AI podcast with two people where we have enough information to 
     
     
  
 
 
	 01:05:57
     ◼ 
      
     ►  
     recreate them in some sense. 
     
     
  
 
 
	 01:05:59
     ◼ 
      
     ►  
     Yeah it's a publicist stunt for the company. The company that made this has an AI text 
     
     
  
 
 
	 01:06:05
     ◼ 
      
     ►  
     to voice generation system that they are trying to sell. 
     
     
  
 
 
	 01:06:08
     ◼ 
      
     ►  
     I mean that's not surprising. 
     
     
  
 
 
	 01:06:10
     ◼ 
      
     ►  
     But it's also interesting just how fast so much of this is being commercialized. 
     
     
  
 
 
	 01:06:15
     ◼ 
      
     ►  
     I stumbled across a – I thought, "Oh God, how brutal is this?" 
     
     
  
 
 
	 01:06:21
     ◼ 
      
     ►  
     But it's a company that makes AI people who will read quite convincingly scripts where 
     
     
  
 
 
	 01:06:27
     ◼ 
      
     ►  
     it looks like it's a talking head segment. 
     
     
  
 
 
	 01:06:29
     ◼ 
      
     ►  
     And the whole idea is, "Oh, you can have all of your corporate training material delivered 
     
     
  
 
 
	 01:06:33
     ◼ 
      
     ►  
     in this way where there's like an AI person who will talk through whatever it is you need 
     
     
  
 
 
	 01:06:37
     ◼ 
      
     ►  
     to onboard your new employees. There's something about that to me which is like, it's very 
     
     
  
 
 
	 01:06:41
     ◼ 
      
     ►  
     convincing but it also feels like what a horrible dystopian nightmare. It's like you as an 
     
     
  
 
 
	 01:06:46
     ◼ 
      
     ►  
     employee sign up with the company is like, "Oh hey, we didn't even take the time to 
     
     
  
 
 
	 01:06:52
     ◼ 
      
     ►  
     film a person going through our own training materials. We just gave it to an AI and they 
     
     
  
 
 
	 01:06:57
     ◼ 
      
     ►  
     made a fake person who you get to listen to explain your job to you." 
     
     
  
 
 
	 01:07:01
     ◼ 
      
     ►  
     Enjoy your training. 
     
     
  
 
 
	 01:07:02
     ◼ 
      
     ►  
     I know, it's just so awful. 
     
     
  
 
 
	 01:07:04
     ◼ 
      
     ►  
     We value you. 
     
     
  
 
 
	 01:07:06
     ◼ 
      
     ►  
     I just want to read from the podcast AI, their kind of description of the show. 
     
     
  
 
 
	 01:07:10
     ◼ 
      
     ►  
     "Whether you're a machine learning enthusiast, just want to hear your favorite topics covered 
     
     
  
 
 
	 01:07:16
     ◼ 
      
     ►  
     in a new way, or even just want to listen to voices from the past brought back to life, 
     
     
  
 
 
	 01:07:20
     ◼ 
      
     ►  
     this is the podcast for you." 
     
     
  
 
 
	 01:07:22
     ◼ 
      
     ►  
     - Right. - Alright. 
     
     
  
 
 
	 01:07:25
     ◼ 
      
     ►  
     - Do you want to hear our ghoulish marionettes say what we made them say? 
     
     
  
 
 
	 01:07:30
     ◼ 
      
     ►  
     - It's just like, if you are a fan of Steve Jobs, which I think is who they are pitching this to, 
     
     
  
 
 
	 01:07:35
     ◼ 
      
     ►  
     right? Why do you want to hear him talk about things he never spoke about? Like what do you 
     
     
  
 
 
	 01:07:39
     ◼ 
      
     ►  
     value from that other than just like hearing the voice? Like if you just want to hear his voice 
     
     
  
 
 
	 01:07:44
     ◼ 
      
     ►  
     just go to youtube and watch like commencement speeches or interviews or whatever. Like I don't 
     
     
  
 
 
	 01:07:49
     ◼ 
      
     ►  
     understand why you want to hear him talk about things he never spoke about. It's not his opinion. 
     
     
  
 
 
	 01:07:56
     ◼ 
      
     ►  
     There is no opinion of him in this. They've built all of this from things he said but you're just 
     
     
  
 
 
	 01:08:03
     ◼ 
      
     ►  
     taking words he used and just putting them together in a new way. It's not actually 
     
     
  
 
 
	 01:08:08
     ◼ 
      
     ►  
     his opinion. And I just I find it so strange. 
     
     
  
 
 
	 01:08:15
     ◼ 
      
     ►  
     Since the last episode, I have much more strongly onboarded the concept of trying to only read 
     
     
  
 
 
	 01:08:23
     ◼ 
      
     ►  
     and consume and listen to and watch media that you know has been produced by a human. 
     
     
  
 
 
	 01:08:29
     ◼ 
      
     ►  
     And how did that do for you with Obu? 
     
     
  
 
 
	 01:08:31
     ◼ 
      
     ►  
     You know what I mean? 
     
     
  
 
 
 
	 01:08:33
     ◼ 
      
     ►  
     - Yeah, it sneaks in there, but at least it's like, 
     
     
  
 
 
	 01:08:36
     ◼ 
      
     ►  
     oh, I know humans made the show. 
     
     
  
 
 
 
	 01:08:39
     ◼ 
      
     ►  
     - Yeah, at least you got that part, right? 
     
     
  
 
 
 
	 01:08:42
     ◼ 
      
     ►  
     - Maybe not all humans, all of it all the time, 
     
     
  
 
 
	 01:08:43
     ◼ 
      
     ►  
     but like enough. 
     
     
  
 
 
	 01:08:44
     ◼ 
      
     ►  
     - Yeah, and so again, this idea that went from like, 
     
     
  
 
 
	 01:08:48
     ◼ 
      
     ►  
     that's crazy to, oh, I guess this is wise advice to live by 
     
     
  
 
 
	 01:08:51
     ◼ 
      
     ►  
     has accelerated quite quickly in my life. 
     
     
  
 
 
	 01:08:54
     ◼ 
      
     ►  
     And yeah, it's kind of a weird, 
     
     
  
 
 
	 01:08:57
     ◼ 
      
     ►  
     oh, make sure you know it's a person 
     
     
  
 
 
	 01:08:59
     ◼ 
      
     ►  
     who wrote or made this thing. 
     
     
  
 
 
	 01:09:01
     ◼ 
      
     ►  
     So yeah, I am trying to onboard that as a concept 
     
     
  
 
 
	 01:09:05
     ◼ 
      
     ►  
     because I think not onboarding that as a concept 
     
     
  
 
 
	 01:09:07
     ◼ 
      
     ►  
     is part of what is going to lead to AI doom for us all. 
     
     
  
 
 
	 01:09:12
     ◼ 
      
     ►  
     - Oh God, we still want to talk about AI. 
     
     
  
 
 
	 01:09:14
     ◼ 
      
     ►  
     All right, yep, okay. 
     
     
  
 
 
	 01:09:16
     ◼ 
      
     ►  
     This episode of Cortex is brought to you by Fitbod. 
     
     
  
 
 
	 01:09:19
     ◼ 
      
     ►  
     Between balancing your work life, your family life, 
     
     
  
 
 
	 01:09:22
     ◼ 
      
     ►  
     and everything else going on in your life, 
     
     
  
 
 
	 01:09:24
     ◼ 
      
     ►  
     it can be hard to make fitness a priority. 
     
     
  
 
 
	 01:09:26
     ◼ 
      
     ►  
     Personal fitness shouldn't be about competing with people. 
     
     
  
 
 
	 01:09:30
     ◼ 
      
     ►  
     What you need is a program that is tailored for you, working with you and not against 
     
     
  
 
 
	 01:09:34
     ◼ 
      
     ►  
     you and that's why you need FitBud. 
     
     
  
 
 
	 01:09:37
     ◼ 
      
     ►  
     You don't want to have to look to other people and do what they're doing. 
     
     
  
 
 
	 01:09:40
     ◼ 
      
     ►  
     You don't want to have to be told exactly what to do. 
     
     
  
 
 
	 01:09:42
     ◼ 
      
     ►  
     You don't want to have to do these boring things every single day. 
     
     
  
 
 
	 01:09:45
     ◼ 
      
     ►  
     You don't want that rigid structure. 
     
     
  
 
 
	 01:09:47
     ◼ 
      
     ►  
     You want something that is flexible, something that adjusts, that is dynamic, that is your 
     
     
  
 
 
	 01:09:51
     ◼ 
      
     ►  
     own fitness plan that you can have access to from anywhere. In a beautiful app of wonderful 
     
     
  
 
 
	 01:09:58
     ◼ 
      
     ►  
     HD video tutorials shot from multiple angles so every exercise you want to learn is super 
     
     
  
 
 
	 01:10:04
     ◼ 
      
     ►  
     simple. This is all why you need Fitbod. They use data to create and adjust that plan for 
     
     
  
 
 
	 01:10:10
     ◼ 
      
     ►  
     you. It's crafted to be unique for you exactly. Their app is super easy to learn. It integrates 
     
     
  
 
 
	 01:10:18
     ◼ 
      
     ►  
     with your smartwatch, your Apple watch, your Wear OS watch, it integrates with apps like 
     
     
  
 
 
	 01:10:22
     ◼ 
      
     ►  
     Strava, Fitbit and Apple Health. Everything's super awesome. I love Fitbud. It really set 
     
     
  
 
 
	 01:10:27
     ◼ 
      
     ►  
     me on a different path to my fitness journey when I started using it. What I really liked 
     
     
  
 
 
	 01:10:32
     ◼ 
      
     ►  
     was that there was a variety there. Like that variety was super important to me. Previously 
     
     
  
 
 
	 01:10:37
     ◼ 
      
     ►  
     I'd used apps and videos and stuff. It's just the same thing every day on a phone that's 
     
     
  
 
 
	 01:10:40
     ◼ 
      
     ►  
     really boring. But I was learning new things, while at the same time it's also mixing up 
     
     
  
 
 
	 01:10:46
     ◼ 
      
     ►  
     with those variety of exercises so you're not overworking or underworking anything. 
     
     
  
 
 
	 01:10:50
     ◼ 
      
     ►  
     It's really clever. Personalized training of this quality can be expensive. Fitbod is just 
     
     
  
 
 
	 01:10:55
     ◼ 
      
     ►  
     $12.99 a month or $79.99 a year but you can get 25% off her membership by going and signing up today 
     
     
  
 
 
	 01:11:02
     ◼ 
      
     ►  
     at fitbod.me/cortex. So go now and get your customized fitness plan at fitbod.me/cortex. 
     
     
  
 
 
	 01:11:09
     ◼ 
      
     ►  
     That is 25% off that you will get by signing up at FITBOD.ME/CORTEX. 
     
     
  
 
 
 
	 01:11:16
     ◼ 
      
     ►  
     Our thanks to Fitbod for their support of this show and Relay FM. 
     
     
  
 
 
 
	 01:11:20
     ◼ 
      
     ►  
     Yes, I do want to talk about Doom for us all, but partly because it's clearing up a little 
     
     
  
 
 
	 01:11:26
     ◼ 
      
     ►  
     bit of confusion from last time as well. 
     
     
  
 
 
 
	 01:11:29
     ◼ 
      
     ►  
     So one of the things we discussed last time was stable diffusion making images like, 
     
     
  
 
 
	 01:11:36
     ◼ 
      
     ►  
     here is the filming of NASA landing on the moon 
     
     
  
 
 
	 01:11:40
     ◼ 
      
     ►  
     and creating fake stuff. 
     
     
  
 
 
	 01:11:42
     ◼ 
      
     ►  
     And I think a totally fair criticism of that 
     
     
  
 
 
	 01:11:44
     ◼ 
      
     ►  
     that I saw from people is, 
     
     
  
 
 
	 01:11:46
     ◼ 
      
     ►  
     hey, Photoshop exists, you know, 
     
     
  
 
 
	 01:11:50
     ◼ 
      
     ►  
     like we can make those images now, 
     
     
  
 
 
	 01:11:52
     ◼ 
      
     ►  
     we don't need an AI art system to generate them for us. 
     
     
  
 
 
	 01:11:56
     ◼ 
      
     ►  
     If you would be convinced by a good Photoshop, 
     
     
  
 
 
	 01:11:59
     ◼ 
      
     ►  
     what's the difference from being convinced 
     
     
  
 
 
	 01:12:01
     ◼ 
      
     ►  
     by an AI art system about something 
     
     
  
 
 
	 01:12:03
     ◼ 
      
     ►  
     that isn't real in the world? 
     
     
  
 
 
	 01:12:06
     ◼ 
      
     ►  
     So what I want to portray here is I think Doom comes in like three phases. 
     
     
  
 
 
	 01:12:11
     ◼ 
      
     ►  
     Phase one, I think will be just pollution of the public information space. 
     
     
  
 
 
	 01:12:18
     ◼ 
      
     ►  
     I think it's sort of a general confusion. 
     
     
  
 
 
	 01:12:20
     ◼ 
      
     ►  
     And what makes the difference between AI art and something like Photoshop is the scale 
     
     
  
 
 
	 01:12:29
     ◼ 
      
     ►  
     and the cost. So right now, if someone wants to try and put out a bunch of misinformation 
     
     
  
 
 
	 01:12:37
     ◼ 
      
     ►  
     or create evidence to back a conspiracy theory, there still needs to be effort that goes into 
     
     
  
 
 
	 01:12:43
     ◼ 
      
     ►  
     creating that thing. Like you were saying, Myke, you need to learn these skills as part of your art 
     
     
  
 
 
	 01:12:48
     ◼ 
      
     ►  
     for how to create a piece of misinformation to put out in the world. And I think what I kind of 
     
     
  
 
 
	 01:12:56
     ◼ 
      
     ►  
     of expect if I'm projecting a bunch of this stuff forward is the scale of this potential 
     
     
  
 
 
	 01:13:03
     ◼ 
      
     ►  
     misleading images and misleading text that were AI generated is just vastly beyond what 
     
     
  
 
 
	 01:13:10
     ◼ 
      
     ►  
     we can imagine now. And I think in the modern world, like a lot of people have been driven 
     
     
  
 
 
	 01:13:16
     ◼ 
      
     ►  
     kind of crazy just from the selection of what information they're presented. Like, you know, 
     
     
  
 
 
	 01:13:26
     ◼ 
      
     ►  
     people on social media can kind of like drive themselves crazy by just going down rabbit 
     
     
  
 
 
	 01:13:30
     ◼ 
      
     ►  
     holes and being continually presented with information that agrees with them. And that's 
     
     
  
 
 
	 01:13:34
     ◼ 
      
     ►  
     just by like, selecting the things that they're seeing that actual human beings have created. 
     
     
  
 
 
	 01:13:40
     ◼ 
      
     ►  
     But what I'm kind of thinking might happen here is that when you're able to generate 
     
     
  
 
 
	 01:13:45
     ◼ 
      
     ►  
     a huge amount of content. Just as right now, companies intentionally AB test what they 
     
     
  
 
 
	 01:13:53
     ◼ 
      
     ►  
     show you for engagement. Not even on purpose, but AI generated content will effectively 
     
     
  
 
 
	 01:14:01
     ◼ 
      
     ►  
     be unintentionally AB tested for convincing this. How convincing is this sequence of words 
     
     
  
 
 
	 01:14:09
     ◼ 
      
     ►  
     in whatever idea it's trying to spread. And I really do view a lot of this stuff as the 
     
     
  
 
 
	 01:14:17
     ◼ 
      
     ►  
     kind of concept of memes, of ideas. They evolve, and they spread, and they mutate, and their 
     
     
  
 
 
	 01:14:24
     ◼ 
      
     ►  
     spreading in the world doesn't have anything to do with how true they are, it has to do 
     
     
  
 
 
	 01:14:29
     ◼ 
      
     ►  
     with how convincing they are. 
     
     
  
 
 
	 01:14:31
     ◼ 
      
     ►  
     Imagine hooking up Twitter's algorithm to a text AI. 
     
     
  
 
 
	 01:14:36
     ◼ 
      
     ►  
     Yeah, I think that's going to happen. 
     
     
  
 
 
	 01:14:37
     ◼ 
      
     ►  
     every time you pull to refresh, it just gives you a bunch of other nonsense. 
     
     
  
 
 
	 01:14:43
     ◼ 
      
     ►  
     That is truly horrifying to me as a thought. 
     
     
  
 
 
	 01:14:47
     ◼ 
      
     ►  
     I think we're going to see that. 
     
     
  
 
 
	 01:14:48
     ◼ 
      
     ►  
     I think we're going to see that or we're going to see something very much like it. 
     
     
  
 
 
	 01:14:51
     ◼ 
      
     ►  
     I mean, look, the AI video stuff that we saw before. 
     
     
  
 
 
	 01:14:55
     ◼ 
      
     ►  
     Do you think TikTok won't start doing that the moment that it becomes engaging to people? 
     
     
  
 
 
	 01:15:01
     ◼ 
      
     ►  
     That's a really good point. 
     
     
  
 
 
	 01:15:02
     ◼ 
      
     ►  
     That's actually the most likely of all of them, I think, to occur. 
     
     
  
 
 
	 01:15:06
     ◼ 
      
     ►  
     to just automatically generate 20 second, 30 second videos every time a person refreshes 
     
     
  
 
 
	 01:15:12
     ◼ 
      
     ►  
     and just keep doing it for whatever keeps that person engaged. 
     
     
  
 
 
	 01:15:16
     ◼ 
      
     ►  
     I think we're going to see that. 
     
     
  
 
 
	 01:15:18
     ◼ 
      
     ►  
     Even if it's not the companies directly doing it themselves, you'll have entities on Twitter 
     
     
  
 
 
	 01:15:23
     ◼ 
      
     ►  
     where it's like, "Oh, it's a bot, but it's, you know, it's acting like it's a person and 
     
     
  
 
 
	 01:15:27
     ◼ 
      
     ►  
     it's just existing in the world and it's doing this kind of unintentional A/B testing for 
     
     
  
 
 
	 01:15:32
     ◼ 
      
     ►  
     convincing this." 
     
     
  
 
 
	 01:15:33
     ◼ 
      
     ►  
     and the ones that are more convincing for whatever reason will just spread better. 
     
     
  
 
 
	 01:15:37
     ◼ 
      
     ►  
     So I really think that is a kind of like, I think a lot of the criticism is people want 
     
     
  
 
 
	 01:15:44
     ◼ 
      
     ►  
     to know, like, gray, what like, you love technology, why aren't you behind this stuff? I used to 
     
     
  
 
 
	 01:15:50
     ◼ 
      
     ►  
     be a real technological optimist. But I've changed my mind on a bunch of that stuff. 
     
     
  
 
 
	 01:15:55
     ◼ 
      
     ►  
     And this is one area in particular, where, and I want to be clear here, I don't think 
     
     
  
 
 
	 01:16:01
     ◼ 
      
     ►  
     It's malicious. 
     
     
  
 
 
	 01:16:02
     ◼ 
      
     ►  
     I don't think it's necessarily that someone's out there trying to do bad, but I think the 
     
     
  
 
 
	 01:16:07
     ◼ 
      
     ►  
     ability to create hundreds of millions of memes in the way of like just a concept that 
     
     
  
 
 
	 01:16:15
     ◼ 
      
     ►  
     can spread at the drop of a hat is just bad for the public information space. 
     
     
  
 
 
	 01:16:22
     ◼ 
      
     ►  
     And the world has barely survived social media in some ways, like that has made stuff so 
     
     
  
 
 
	 01:16:29
     ◼ 
      
     ►  
     so bad with how people think about the world. And this is that just taken to the next level 
     
     
  
 
 
	 01:16:34
     ◼ 
      
     ►  
     by a huge order of magnitude. So I suspect that'll be one of the first ways that this 
     
     
  
 
 
	 01:16:41
     ◼ 
      
     ►  
     becomes obviously bad over time. Like once it progresses out of the stage of just being 
     
     
  
 
 
	 01:16:49
     ◼ 
      
     ►  
     an interesting toy. My hope is that for whatever reason, we're in the part of the technological 
     
     
  
 
 
	 01:16:55
     ◼ 
      
     ►  
     development where it looks like it's an exponential graph, but it's rapidly going 
     
     
  
 
 
	 01:17:00
     ◼ 
      
     ►  
     to level off at an s-curve, and we discover "oh, there's parts of this that were way 
     
     
  
 
 
	 01:17:05
     ◼ 
      
     ►  
     harder than we thought." That's where I hope this is going, but… 
     
     
  
 
 
	 01:17:09
     ◼ 
      
     ►  
     MATT: You show me the signs of that one on the next episode, then. 
     
     
  
 
 
	 01:17:12
     ◼ 
      
     ►  
     BRIAN well, I'm in this position where I feel 
     
     
  
 
 
	 01:17:15
     ◼ 
      
     ►  
     a little bit like "oh, I'm like a crazy person talking into a microphone about this 
     
     
  
 
 
	 01:17:19
     ◼ 
      
     ►  
     stuff. But I have spoken to some of the top people in the world in this area and they're 
     
     
  
 
 
	 01:17:28
     ◼ 
      
     ►  
     very concerned is the way that I would put it. And talking to people, I was trying to 
     
     
  
 
 
	 01:17:34
     ◼ 
      
     ►  
     tease out this concept of where do we think we are on this curve? Is this the start of 
     
     
  
 
 
	 01:17:39
     ◼ 
      
     ►  
     the exponential or are there obvious problems ahead? And the answer was pretty universally, 
     
     
  
 
 
	 01:17:45
     ◼ 
      
     ►  
     Oh, the as far as we can tell, the exponential has barely begun. Because the thing part of 
     
     
  
 
 
	 01:17:51
     ◼ 
      
     ►  
     the reason it's making such fast progress is because the the work being done right now 
     
     
  
 
 
	 01:17:57
     ◼ 
      
     ►  
     is still in the realm of Oh, hey, you get into work. And like, what's the first thing 
     
     
  
 
 
	 01:18:01
     ◼ 
      
     ►  
     you think of that could make this better? And you try it, and it makes it better. That's 
     
     
  
 
 
	 01:18:05
     ◼ 
      
     ►  
     just like an indication that we're at the start of an exponential curve. So yeah, I 
     
     
  
 
 
	 01:18:09
     ◼ 
      
     ►  
     think we're gonna have like a bunch of confusion about this stuff. I think that transitions 
     
     
  
 
 
	 01:18:14
     ◼ 
      
     ►  
     into genuine economic problems as particularly language models get better and better and better 
     
     
  
 
 
	 01:18:23
     ◼ 
      
     ►  
     at doing all the kinds of work that humans do, which is largely knowledge work. 
     
     
  
 
 
	 01:18:30
     ◼ 
      
     ►  
     You know, and it's funny, like I made "Humans need not apply" eight years ago now, I think? 
     
     
  
 
 
	 01:18:37
     ◼ 
      
     ►  
     And it's been on the back of my mind about, you know, revisiting that at some point. 
     
     
  
 
 
	 01:18:41
     ◼ 
      
     ►  
     And I was kind of thinking like, oh, I don't know how relevant this video still is. But having 
     
     
  
 
 
	 01:18:48
     ◼ 
      
     ►  
     rewatched it, it's like, Oh, no, I put it back, like to be featured on my channel under one of 
     
     
  
 
 
	 01:18:52
     ◼ 
      
     ►  
     like the most watched videos, because it's like, no, no, all this AI stuff and all of the language 
     
     
  
 
 
	 01:18:56
     ◼ 
      
     ►  
     stuff like makes this way more concerning. I think the lesson learned there has simply been that 
     
     
  
 
 
	 01:19:02
     ◼ 
      
     ►  
     physical automation is slower to progress for a bunch of reasons. But all of the knowledge 
     
     
  
 
 
	 01:19:08
     ◼ 
      
     ►  
     worker stuff is coming along very very fast. I just know that there are companies that 
     
     
  
 
 
	 01:19:13
     ◼ 
      
     ►  
     are very explicitly targeting low-level knowledge work and then will be progressing further 
     
     
  
 
 
	 01:19:19
     ◼ 
      
     ►  
     and further up the chain as fast as they can with better and better language models to 
     
     
  
 
 
	 01:19:23
     ◼ 
      
     ►  
     do all sorts of things that people can do. I understand that lots of people just fundamentally 
     
     
  
 
 
	 01:19:28
     ◼ 
      
     ►  
     disagree with me on this point of economics that jobs can't be replaced because humans 
     
     
  
 
 
	 01:19:34
     ◼ 
      
     ►  
     have infinite needs and always want more things. I understand that argument, I just don't 
     
     
  
 
 
	 01:19:41
     ◼ 
      
     ►  
     agree with it, and I think AI just breaks some of the fundamental assumptions that are 
     
     
  
 
 
	 01:19:45
     ◼ 
      
     ►  
     built into that model of as we get better machines of all kinds, we just increase the 
     
     
  
 
 
	 01:19:51
     ◼ 
      
     ►  
     quality of life and increase our desires. Like, I just don't think that that's universally 
     
     
  
 
 
	 01:19:55
     ◼ 
      
     ►  
     true. So yeah, I think we're going to end up with some really major problems in the 
     
     
  
 
 
	 01:20:01
     ◼ 
      
     ►  
     the economy, particularly in the knowledge worker part of that economy. 
     
     
  
 
 
	 01:20:06
     ◼ 
      
     ►  
     And look, we don't need to talk about it today because we've talked about it enough, 
     
     
  
 
 
	 01:20:10
     ◼ 
      
     ►  
     but I think once you start encroaching on AI systems that are good enough to replace 
     
     
  
 
 
	 01:20:15
     ◼ 
      
     ►  
     most human work, you really start encroaching on the kinds of things that can lead to the 
     
     
  
 
 
	 01:20:21
     ◼ 
      
     ►  
     extinction of the human race. 
     
     
  
 
 
	 01:20:24
     ◼ 
      
     ►  
     Maybe that's a little too heavy for today, but that's kind of my having thought it 
     
     
  
 
 
	 01:20:28
     ◼ 
      
     ►  
     through of like, what are the three phases of where does this go? 
     
     
  
 
 
	 01:20:31
     ◼ 
      
     ►  
     – confusion of the public information space, destruction of the economy, extinction of 
     
     
  
 
 
	 01:20:36
     ◼ 
      
     ►  
     the species. It's one, two, three. 
     
     
  
 
 
 
	 01:20:40
     ◼ 
      
     ►  
     How you feeling, Myke? 
     
     
  
 
 
 
 
	 01:20:45
     ◼ 
      
     ►  
     No, I didn't want to do another hour and a half on this, to be honest. But I guess 
     
     
  
 
 
	 01:20:50
     ◼ 
      
     ►  
     this is who we are now. 
     
     
  
 
 
	 01:20:52
     ◼ 
      
     ►  
     Yeah. Can I just ask you though, like, how crazy does that sound to you? 
     
     
  
 
 
	 01:20:55
     ◼ 
      
     ►  
     No, it doesn't. And that's why I don't like it. I don't have optimism about this 
     
     
  
 
 
	 01:20:59
     ◼ 
      
     ►  
     this area of technology. I do not think that this is a thing that will produce much good. 
     
     
  
 
 
	 01:21:05
     ◼ 
      
     ►  
     We can leave it there for now then. On with the show. It's only been an hour and a half. 
     
     
  
 
 
	 01:21:17
     ◼ 
      
     ►  
     Just some quick follow up from the last thing we talked about on the previous episode. Okay, 
     
     
  
 
 
	 01:21:21
     ◼ 
      
     ►  
     on the lighter side of things, what is a podcast, Myke? 
     
     
  
 
 
	 01:21:25
     ◼ 
      
     ►  
     I don't know why. Why are we doing this one again, too? What is this? 
     
     
  
 
 
	 01:21:30
     ◼ 
      
     ►  
     We just relive the horrors of the previous episode? 
     
     
  
 
 
	 01:21:32
     ◼ 
      
     ►  
     I feel so badly for you, Myke. Last episode, you stumbled into a terrible mistake, 
     
     
  
 
 
	 01:21:39
     ◼ 
      
     ►  
     which was attempting to describe... but what is a podcast? 
     
     
  
 
 
	 01:21:43
     ◼ 
      
     ►  
     Well, the real issue was I put two topics together that seemed related, but they were unrelated. 
     
     
  
 
 
	 01:21:50
     ◼ 
      
     ►  
     Right, yeah. This is always the dangers of speaking 
     
     
  
 
 
	 01:21:53
     ◼ 
      
     ►  
     extemporaneously and you're just sort of like in the middle of a conversation and you say some 
     
     
  
 
 
	 01:21:57
     ◼ 
      
     ►  
     things and so anyway we ended up talking for a while about like but what is a podcast what must 
     
     
  
 
 
	 01:22:03
     ◼ 
      
     ►  
     it be in order for it to be the platonic ideal of a podcast and this this caught people's attention 
     
     
  
 
 
	 01:22:09
     ◼ 
      
     ►  
     and you can click the link in the show notes a cortex sent in what i absolutely adored is this 
     
     
  
 
 
	 01:22:16
     ◼ 
      
     ►  
     oh i saw this podcast alignment chart and i read this and was like i think i agree with absolutely 
     
     
  
 
 
	 01:22:22
     ◼ 
      
     ►  
     everything on this chart. This nails what I wanted, I think, to kind of get across, 
     
     
  
 
 
	 01:22:27
     ◼ 
      
     ►  
     but maybe didn't do a great job of doing, I don't know. 
     
     
  
 
 
	 01:22:29
     ◼ 
      
     ►  
     So in the great tradition of memes, there's a meme where people make the alignment chart for 
     
     
  
 
 
	 01:22:36
     ◼ 
      
     ►  
     various things, and I think this started with what is still the fantastic example of the sandwich 
     
     
  
 
 
	 01:22:42
     ◼ 
      
     ►  
     alignment chart. So you have two axes in the chart for the sandwich, which is ingredients 
     
     
  
 
 
	 01:22:49
     ◼ 
      
     ►  
     and structure. So it ranges from ingredient purist to ingredient rebel, and then you have 
     
     
  
 
 
	 01:22:56
     ◼ 
      
     ►  
     structure purist. A sandwich must have the classic sandwich shape, two pieces of bread 
     
     
  
 
 
	 01:23:01
     ◼ 
      
     ►  
     with toppings in between. And then like structure rebel, any kind of food enveloped in any way is a 
     
     
  
 
 
	 01:23:07
     ◼ 
      
     ►  
     sandwich. Oh, like a hot dog? Yeah, so this is like, is a hot dog a sandwich? And hot dog falls 
     
     
  
 
 
	 01:23:13
     ◼ 
      
     ►  
     on the sandwich alignment chart of ingredient neutral and structure neutral. Wait, this is a 
     
     
  
 
 
	 01:23:18
     ◼ 
      
     ►  
     - There's a very intriguing way that you've introduced this. 
     
     
  
 
 
	 01:23:21
     ◼ 
      
     ►  
     Isn't this just like the chaotic, evil, chaotic, good thing? 
     
     
  
 
 
	 01:23:25
     ◼ 
      
     ►  
     - Yes, that's where this comes from. 
     
     
  
 
 
	 01:23:26
     ◼ 
      
     ►  
     That's the origin. 
     
     
  
 
 
	 01:23:27
     ◼ 
      
     ►  
     - Oh, right. - It is like. 
     
     
  
 
 
	 01:23:28
     ◼ 
      
     ►  
     - I misunderstood you and thought you were saying 
     
     
  
 
 
	 01:23:29
     ◼ 
      
     ►  
     that the sandwich one was the origin. 
     
     
  
 
 
	 01:23:31
     ◼ 
      
     ►  
     I was like, I don't think you're right there. 
     
     
  
 
 
	 01:23:33
     ◼ 
      
     ►  
     - No, no, no, no, no, no, no, no, no, no, no, no. 
     
     
  
 
 
 
	 01:23:37
     ◼ 
      
     ►  
     I always forget 'cause I never played "Dea Deas." 
     
     
  
 
 
	 01:23:39
     ◼ 
      
     ►  
     - Chaotic, neutral, chaotic, good, that's the thing. 
     
     
  
 
 
	 01:23:41
     ◼ 
      
     ►  
     - No, but the two axes, it's evil and good 
     
     
  
 
 
	 01:23:43
     ◼ 
      
     ►  
     and then lawful and chaotic. 
     
     
  
 
 
	 01:23:45
     ◼ 
      
     ►  
     Is that, that's, that's, 
     
     
  
 
 
	 01:23:46
     ◼ 
      
     ►  
     those the two axes on the traditional one? 
     
     
  
 
 
 
	 01:23:48
     ◼ 
      
     ►  
     The sandwich alignment chart is the meme, right? 
     
     
  
 
 
	 01:23:51
     ◼ 
      
     ►  
     Which is like, oh, here is from which other things birth. 
     
     
  
 
 
	 01:23:54
     ◼ 
      
     ►  
     And so what I love on the internet is people will get into an argument about like, what 
     
     
  
 
 
 
	 01:24:00
     ◼ 
      
     ►  
     Is a hot dog a sandwich? 
     
     
  
 
 
	 01:24:01
     ◼ 
      
     ►  
     Is a pop tart a sandwich? 
     
     
  
 
 
	 01:24:03
     ◼ 
      
     ►  
     Is a chicken wrap a sandwich? 
     
     
  
 
 
	 01:24:04
     ◼ 
      
     ►  
     And someone will come up with like, here's the alignment chart to try to describe where 
     
     
  
 
 
	 01:24:09
     ◼ 
      
     ►  
     all of these things fit in. 
     
     
  
 
 
	 01:24:10
     ◼ 
      
     ►  
     And so someone did this for podcasts, the podcast alignment charts. 
     
     
  
 
 
	 01:24:15
     ◼ 
      
     ►  
     The two axes are distribution method and media type. 
     
     
  
 
 
	 01:24:20
     ◼ 
      
     ►  
     So do you wanna run through some of these, Myke? 
     
     
  
 
 
	 01:24:22
     ◼ 
      
     ►  
     Where do you fit on the podcast alignment chart? 
     
     
  
 
 
	 01:24:25
     ◼ 
      
     ►  
     - Oh man, I think I was... 
     
     
  
 
 
	 01:24:28
     ◼ 
      
     ►  
     - I feel like you were making an argument 
     
     
  
 
 
	 01:24:30
     ◼ 
      
     ►  
     for traditional distribution method, 
     
     
  
 
 
	 01:24:32
     ◼ 
      
     ►  
     traditional media type, 
     
     
  
 
 
	 01:24:33
     ◼ 
      
     ►  
     which is an audio RSS feed is a podcast. 
     
     
  
 
 
	 01:24:37
     ◼ 
      
     ►  
     - No, because I don't, I think I am a modernist. 
     
     
  
 
 
	 01:24:41
     ◼ 
      
     ►  
     - Distribution modernist, media type traditionalist, 
     
     
  
 
 
	 01:24:44
     ◼ 
      
     ►  
     - Any audio that's subscribable is a podcast? 
     
     
  
 
 
	 01:24:47
     ◼ 
      
     ►  
     - No, I think I'm modern modern. 
     
     
  
 
 
	 01:24:49
     ◼ 
      
     ►  
     - Any audio or video that's subscribable is a podcast? 
     
     
  
 
 
	 01:24:52
     ◼ 
      
     ►  
     Oh, so my YouTube channel is a podcast, 
     
     
  
 
 
	 01:24:54
     ◼ 
      
     ►  
     that's what you're saying? 
     
     
  
 
 
	 01:24:55
     ◼ 
      
     ►  
     - No, you see that's, yeah, no, 
     
     
  
 
 
	 01:24:56
     ◼ 
      
     ►  
     that I'm modern traditionalist. 
     
     
  
 
 
	 01:24:59
     ◼ 
      
     ►  
     'Cause that's the thing of like, 
     
     
  
 
 
	 01:25:00
     ◼ 
      
     ►  
     but it gets into an issue for me is that I watch videos 
     
     
  
 
 
	 01:25:04
     ◼ 
      
     ►  
     that I consider podcasts. 
     
     
  
 
 
	 01:25:06
     ◼ 
      
     ►  
     It's the any, I would see, I would phrase it as any audio 
     
     
  
 
 
	 01:25:10
     ◼ 
      
     ►  
     or video that's subscribable can be a podcast, 
     
     
  
 
 
	 01:25:14
     ◼ 
      
     ►  
     That would be my personal definition, I think. 
     
     
  
 
 
	 01:25:17
     ◼ 
      
     ►  
     Would be closer to how I feel. 
     
     
  
 
 
	 01:25:20
     ◼ 
      
     ►  
     But I think I'm more traditionalist modernist, 
     
     
  
 
 
	 01:25:22
     ◼ 
      
     ►  
     which is any audio that's subscribable is a podcast. 
     
     
  
 
 
	 01:25:25
     ◼ 
      
     ►  
     No, but then it's got audio books in here. 
     
     
  
 
 
	 01:25:27
     ◼ 
      
     ►  
     This is very complicated. 
     
     
  
 
 
 
 
	 01:25:30
     ◼ 
      
     ►  
     I would say any audio or video 
     
     
  
 
 
	 01:25:32
     ◼ 
      
     ►  
     that somebody wants to call a podcast 
     
     
  
 
 
	 01:25:34
     ◼ 
      
     ►  
     that you can subscribe to can be a podcast. 
     
     
  
 
 
	 01:25:36
     ◼ 
      
     ►  
     I think that's how I actually personally sit now. 
     
     
  
 
 
	 01:25:39
     ◼ 
      
     ►  
     I think that works for me. 
     
     
  
 
 
	 01:25:42
     ◼ 
      
     ►  
     I really love these charts go kind of crazy, right? 
     
     
  
 
 
	 01:25:45
     ◼ 
      
     ►  
     I think the problem with the podcast one is a two-dimensional surface is not enough to 
     
     
  
 
 
	 01:25:50
     ◼ 
      
     ►  
     express the entirety of what it wants to be. 
     
     
  
 
 
	 01:25:53
     ◼ 
      
     ►  
     You need a third axis, which I think is something like consumption intention. 
     
     
  
 
 
	 01:26:00
     ◼ 
      
     ►  
     So for me, like, I would be like a traditionalist here, right? 
     
     
  
 
 
	 01:26:03
     ◼ 
      
     ►  
     That the consumption intention is audio only. 
     
     
  
 
 
	 01:26:08
     ◼ 
      
     ►  
     That to me is a really key characteristic of what is a podcast. 
     
     
  
 
 
	 01:26:13
     ◼ 
      
     ►  
     It has to be intended as an audio first experience, which isn't quite captured on this chart. 
     
     
  
 
 
	 01:26:20
     ◼ 
      
     ►  
     But then it can be a video. 
     
     
  
 
 
	 01:26:22
     ◼ 
      
     ►  
     It can be a Spotify exclusive. 
     
     
  
 
 
	 01:26:24
     ◼ 
      
     ►  
     Yeah, see, I don't know. 
     
     
  
 
 
	 01:26:28
     ◼ 
      
     ►  
     I am very sympathetic to the position you were expressing last time, which is there's, 
     
     
  
 
 
	 01:26:33
     ◼ 
      
     ►  
     or at least I thought it was the position you're expressing last time, that there's 
     
     
  
 
 
	 01:26:37
     ◼ 
      
     ►  
     Something about it needs to be generally accessible. I think for me, the example in my mind that 
     
     
  
 
 
	 01:26:45
     ◼ 
      
     ►  
     really sticks out is Audible does these things that they call podcasts, which are like little 
     
     
  
 
 
	 01:26:50
     ◼ 
      
     ►  
     shows that you can subscribe to in the Audible app. 
     
     
  
 
 
	 01:26:52
     ◼ 
      
     ►  
     I don't know why they call them podcasts. I don't get that one at all. 
     
     
  
 
 
	 01:26:56
     ◼ 
      
     ►  
     I've listened to some of them, like they're fine, I haven't found one that's amazing 
     
     
  
 
 
	 01:26:59
     ◼ 
      
     ►  
     or anything, but they're also, I don't know, there's some part of my brain which 
     
     
  
 
 
	 01:27:03
     ◼ 
      
     ►  
     is just like, this is not a podcast at all. 
     
     
  
 
 
	 01:27:05
     ◼ 
      
     ►  
     It's just original audio content. 
     
     
  
 
 
	 01:27:06
     ◼ 
      
     ►  
     Yeah, it's an audio- 
     
     
  
 
 
	 01:27:07
     ◼ 
      
     ►  
     call them audible originals yeah and they call them podcasts and I just wonder if maybe 
     
     
  
 
 
	 01:27:12
     ◼ 
      
     ►  
     they should just stick to one of those descriptors but yeah but anyway like those are the ones 
     
     
  
 
 
	 01:27:18
     ◼ 
      
     ►  
     that really clang in my brain of like but what's the key thing there and I think the 
     
     
  
 
 
	 01:27:22
     ◼ 
      
     ►  
     fact that they're only available on audio really makes them not a podcast to me okay 
     
     
  
 
 
	 01:27:27
     ◼ 
      
     ►  
     well then Joe Rogan yeah is that a podcast? 
     
     
  
 
 
	 01:27:31
     ◼ 
      
     ►  
     - Ah, damn you, Myke. 
     
     
  
 
 
 
	 01:27:37
     ◼ 
      
     ►  
     'Cause I would say yes, it is, 
     
     
  
 
 
	 01:27:38
     ◼ 
      
     ►  
     even though it's a Spotify exclusive. 
     
     
  
 
 
	 01:27:40
     ◼ 
      
     ►  
     - Curse you. 
     
     
  
 
 
	 01:27:44
     ◼ 
      
     ►  
     That is the perfect counter example to my position. 
     
     
  
 
 
	 01:27:49
     ◼ 
      
     ►  
     - So that's why I feel like for me, 
     
     
  
 
 
	 01:27:51
     ◼ 
      
     ►  
     there is the third axis, which is like intent. 
     
     
  
 
 
	 01:27:54
     ◼ 
      
     ►  
     And that's when if we go on this third axis of intent, 
     
     
  
 
 
	 01:27:58
     ◼ 
      
     ►  
     I would sit in the modernist 
     
     
  
 
 
	 01:27:59
     ◼ 
      
     ►  
     where it could be like any audio or video that's 
     
     
  
 
 
	 01:28:01
     ◼ 
      
     ►  
     subscribable can be a podcast, 
     
     
  
 
 
	 01:28:04
     ◼ 
      
     ►  
     but it's on creators intent. 
     
     
  
 
 
	 01:28:06
     ◼ 
      
     ►  
     - Yeah, but see, look, just to make things hard, 
     
     
  
 
 
	 01:28:08
     ◼ 
      
     ►  
     it has to be phrased in order to fit the meme as, 
     
     
  
 
 
	 01:28:11
     ◼ 
      
     ►  
     is a podcast. 
     
     
  
 
 
	 01:28:13
     ◼ 
      
     ►  
     You can't put in these weasel words of can, right? 
     
     
  
 
 
	 01:28:15
     ◼ 
      
     ►  
     The purpose of the alignment chart 
     
     
  
 
 
	 01:28:17
     ◼ 
      
     ►  
     is to definitively answer what is a podcast. 
     
     
  
 
 
	 01:28:20
     ◼ 
      
     ►  
     - Well, I'll tell you now, it doesn't. 
     
     
  
 
 
	 01:28:22
     ◼ 
      
     ►  
     - I really love looking at this chart, 
     
     
  
 
 
	 01:28:24
     ◼ 
      
     ►  
     that traditionalist traditionalist is 
     
     
  
 
 
	 01:28:26
     ◼ 
      
     ►  
     an audio RSS feed is a podcast. 
     
     
  
 
 
	 01:28:29
     ◼ 
      
     ►  
     And then you start moving in these reasonable directions of any audio that's subscribable 
     
     
  
 
 
	 01:28:33
     ◼ 
      
     ►  
     is a podcast. 
     
     
  
 
 
	 01:28:35
     ◼ 
      
     ►  
     And then as you get into the radicalist ones, it just gets very funny, right? 
     
     
  
 
 
	 01:28:39
     ◼ 
      
     ►  
     Any audio available on the internet is a podcast. 
     
     
  
 
 
	 01:28:43
     ◼ 
      
     ►  
     Anything that you can subscribe to is a podcast. 
     
     
  
 
 
	 01:28:46
     ◼ 
      
     ►  
     And my personal favorite, the radicalist radicalist is that anything on the internet is a podcast, 
     
     
  
 
 
	 01:28:53
     ◼ 
      
     ►  
     which made me laugh so hard when I first saw it. 
     
     
  
 
 
	 01:28:57
     ◼ 
      
     ►  
     Because that got me. 
     
     
  
 
 
	 01:28:59
     ◼ 
      
     ►  
     Because you know what? 
     
     
  
 
 
	 01:29:02
     ◼ 
      
     ►  
     That fits with some people that I've come across in my life, you know what I mean? 
     
     
  
 
 
	 01:29:06
     ◼ 
      
     ►  
     I also enjoyed the cortexes giggling over this one, where people were like, "This 
     
     
  
 
 
	 01:29:11
     ◼ 
      
     ►  
     comment on this Reddit thread is a podcast, right? 
     
     
  
 
 
	 01:29:13
     ◼ 
      
     ►  
     Because I'm a radicalist? 
     
     
  
 
 
	 01:29:15
     ◼ 
      
     ►  
     Radicalist!" 
     
     
  
 
 
	 01:29:16
     ◼ 
      
     ►  
     I thought this was fantastic, and thanks to XD1936 for posting it, it got a really good 
     
     
  
 
 
	 01:29:22
     ◼ 
      
     ►  
     laugh out of it and I think yeah it was just a fun way to try to encapsulate the conversation 
     
     
  
 
 
	 01:29:27
     ◼ 
      
     ►  
     from last time. 
     
     
  
 
 
	 01:29:30
     ◼ 
      
     ►  
     This episode of Cortex is brought to you by Wealthfront. 
     
     
  
 
 
	 01:29:32
     ◼ 
      
     ►  
     Is your bank keeping money that could be yours? 
     
     
  
 
 
	 01:29:35
     ◼ 
      
     ►  
     If you're earning less than Wealthfront's 2.55% APY, they might be. 
     
     
  
 
 
 
	 01:29:41
     ◼ 
      
     ►  
     Federal interest rates have been going up this year which means banks have had the opportunity 
     
     
  
 
 
	 01:29:44
     ◼ 
      
     ►  
     to earn more on your savings. 
     
     
  
 
 
	 01:29:46
     ◼ 
      
     ►  
     Where's all the extra money going? 
     
     
  
 
 
	 01:29:48
     ◼ 
      
     ►  
     Well according to the FDIC, the average US bank has only raised their rates to 0.17% 
     
     
  
 
 
	 01:29:55
     ◼ 
      
     ►  
     this year, while Wealthrun is now offering their clients a rate that's about 15 times 
     
     
  
 
 
	 01:29:59
     ◼ 
      
     ►  
     higher with the Wealthrun Cash account. 
     
     
  
 
 
	 01:30:02
     ◼ 
      
     ►  
     Of course you have a choice. 
     
     
  
 
 
	 01:30:03
     ◼ 
      
     ►  
     You could let your bank keep the extra money they're making, or you could move your savings 
     
     
  
 
 
	 01:30:08
     ◼ 
      
     ►  
     to a high yield account like Wealthrun and earn 2.55% APY. 
     
     
  
 
 
	 01:30:13
     ◼ 
      
     ►  
     I think that having your savings in order is really just a great thing, it's one less 
     
     
  
 
 
	 01:30:17
     ◼ 
      
     ►  
     thing to have to think about, so check it out. It's super easy to sign up, plus you 
     
     
  
 
 
	 01:30:21
     ◼ 
      
     ►  
     get unlimited transfers that are completely fee free and up to $1 million in FDIC insurance 
     
     
  
 
 
	 01:30:27
     ◼ 
      
     ►  
     through partner banks. There are no account fees, no minimum balance, and if you sign 
     
     
  
 
 
	 01:30:31
     ◼ 
      
     ►  
     up at wealthfront.com/cortex you get a free $50 bonus of an initial deposit of $500. Don't 
     
     
  
 
 
	 01:30:37
     ◼ 
      
     ►  
     let your bank keep the interest you could be earning. Join nearly half a million people 
     
     
  
 
 
	 01:30:42
     ◼ 
      
     ►  
     who already use Wealthfront to earn 15 times more than the average US bank. That's wealthfront.com/cortex 
     
     
  
 
 
	 01:30:49
     ◼ 
      
     ►  
     to get started and earn your free $50 bonus. That's wealthfront.com/cortex. This has been 
     
     
  
 
 
	 01:30:55
     ◼ 
      
     ►  
     a paid endorsement for Wealthfront. Our thanks to Wealthfront for the support of this show 
     
     
  
 
 
	 01:30:59
     ◼ 
      
     ►  
     and Relay FM. 
     
     
  
 
 
	 01:31:01
     ◼ 
      
     ►  
     We are hurtling towards the end of the year. 
     
     
  
 
 
	 01:31:02
     ◼ 
      
     ►  
     Uh, yes, alarmingly fast. 
     
     
  
 
 
	 01:31:05
     ◼ 
      
     ►  
     And this is the final regular episode of Cortex this year. 
     
     
  
 
 
	 01:31:09
     ◼ 
      
     ►  
     Don't remind me, Myke. 
     
     
  
 
 
	 01:31:10
     ◼ 
      
     ►  
     So if you have more to say about AI, oh god, great, where's AI art gonna be in January? 
     
     
  
 
 
 
	 01:31:17
     ◼ 
      
     ►  
     By January we're gonna be at step two of my three steps to the apocalypse. 
     
     
  
 
 
	 01:31:21
     ◼ 
      
     ►  
     Excellent, can't wait for it. 
     
     
  
 
 
	 01:31:23
     ◼ 
      
     ►  
     So maybe that's why like we had to spend the best part of two hours doing follow-up, because 
     
     
  
 
 
	 01:31:28
     ◼ 
      
     ►  
     we're not gonna do it now. 
     
     
  
 
 
	 01:31:30
     ◼ 
      
     ►  
     Yeah I guess so, let's say that, that was the plan, we didn't just end up talking about 
     
     
  
 
 
	 01:31:34
     ◼ 
      
     ►  
     it for forever. 
     
     
  
 
 
	 01:31:35
     ◼ 
      
     ►  
     There is a possibility we might do something in Mortax, getmortax.com, I can't say that 
     
     
  
 
 
	 01:31:40
     ◼ 
      
     ►  
     for sure because I don't know what we're going to do over the next two months except 
     
     
  
 
 
	 01:31:43
     ◼ 
      
     ►  
     for what I know we're going to do which is state of the apps November yearly themes December. 
     
     
  
 
 
	 01:31:49
     ◼ 
      
     ►  
     So on that note for yearly themes I wanted to take this as a time to suggest a method 
     
     
  
 
 
	 01:31:56
     ◼ 
      
     ►  
     to people for preparing for their yearly theme. If you don't know what a yearly theme is I 
     
     
  
 
 
	 01:32:02
     ◼ 
      
     ►  
     will put a video in the show notes where Grey explains what yearly themes are but we're 
     
     
  
 
 
	 01:32:06
     ◼ 
      
     ►  
     going to get to it again obviously in December so I don't want to talk about it now. But 
     
     
  
 
 
	 01:32:11
     ◼ 
      
     ►  
     I am personally at the point of my year now where I'm preparing my yearly theme. The 
     
     
  
 
 
	 01:32:17
     ◼ 
      
     ►  
     way I do this is I have it in my mind to start noting down things that I'm happy about 
     
     
  
 
 
	 01:32:25
     ◼ 
      
     ►  
     in my work life and personal life and things that I find frustrating in those as well and 
     
     
  
 
 
	 01:32:31
     ◼ 
      
     ►  
     things I would like to change. So I keep this as an Apple note, it's in my yearly themes 
     
     
  
 
 
	 01:32:36
     ◼ 
      
     ►  
     Apple note, but I have it just playing in my mind that as I'm bumping into things that 
     
     
  
 
 
	 01:32:42
     ◼ 
      
     ►  
     are good or bad that I would like to continue or stop, I write them all down. I just write 
     
     
  
 
 
	 01:32:47
     ◼ 
      
     ►  
     them down. I don't need to draw any comparisons to them yet, I just start writing them down. 
     
     
  
 
 
	 01:32:52
     ◼ 
      
     ►  
     So I end up with a list of things, maybe 20 things or so, over the space of a few weeks 
     
     
  
 
 
	 01:32:58
     ◼ 
      
     ►  
     or a month that are in my head. Some of them are not even things I bump into, but because 
     
     
  
 
 
	 01:33:02
     ◼ 
      
     ►  
     I'm in this kind of reviewing mode, things just pop up in my head randomly. I'm like 
     
     
  
 
 
	 01:33:07
     ◼ 
      
     ►  
     "Oh, that's a thing, I'll write that down." Then as we get towards November or so, I start 
     
     
  
 
 
	 01:33:13
     ◼ 
      
     ►  
     to review this list of things and note the similarities where I can tie some similarities 
     
     
  
 
 
	 01:33:18
     ◼ 
      
     ►  
     between things. Be like "Oh, that's actually related to this" or "That is similar to this." 
     
     
  
 
 
	 01:33:24
     ◼ 
      
     ►  
     I then use these links as the basis to build my theme. 
     
     
  
 
 
	 01:33:27
     ◼ 
      
     ►  
     So what do I think I could do next year to address some of these common issues that I 
     
     
  
 
 
	 01:33:34
     ◼ 
      
     ►  
     have or these common things that I would like to do more of? 
     
     
  
 
 
	 01:33:38
     ◼ 
      
     ►  
     So if you struggle to come up with creating yearly theme, this is my top tip. 
     
     
  
 
 
	 01:33:43
     ◼ 
      
     ►  
     It's how I do mine every year. 
     
     
  
 
 
	 01:33:44
     ◼ 
      
     ►  
     I don't know if this is asking for spoilers for the theme episode, which I don't want, 
     
     
  
 
 
	 01:33:49
     ◼ 
      
     ►  
     but I just I feel like I would like some concrete examples of the kinds of things that you're 
     
     
  
 
 
	 01:33:52
     ◼ 
      
     ►  
     down that you end up thinking about and they don't they don't have to be 
     
     
  
 
 
	 01:33:56
     ◼ 
      
     ►  
     examples for this year. I will give you some examples for this year but I'm not 
     
     
  
 
 
	 01:34:00
     ◼ 
      
     ►  
     gonna give you so many that I think was born my theme. Good. I want to spend more 
     
     
  
 
 
	 01:34:04
     ◼ 
      
     ►  
     time on product creation and design. That's one. I want to listen to more 
     
     
  
 
 
	 01:34:09
     ◼ 
      
     ►  
     music. I'm happy with how I've looked after my health and want to do more of 
     
     
  
 
 
	 01:34:13
     ◼ 
      
     ►  
     that. I want to be smarter about my scheduling and the days that I'm in the 
     
     
  
 
 
	 01:34:17
     ◼ 
      
     ►  
     office. These are the kinds of things I write down. 
     
     
  
 
 
	 01:34:20
     ◼ 
      
     ►  
     Okay, so presumably then the way you have this happen is there's some frustration 
     
     
  
 
 
	 01:34:27
     ◼ 
      
     ►  
     about the scheduling and then I guess you notice that and add it to this file? 
     
     
  
 
 
	 01:34:30
     ◼ 
      
     ►  
     Is that the system? Mm-hmm. Okay. Like all these are at the moment 
     
     
  
 
 
	 01:34:35
     ◼ 
      
     ►  
     it's just an outline like bullet points of things, right? And I write a 
     
     
  
 
 
	 01:34:39
     ◼ 
      
     ►  
     bunch of notes down, I write some context down, and so then I start reading through 
     
     
  
 
 
	 01:34:43
     ◼ 
      
     ►  
     these and I'm like, "Oh hang on a minute, this can relate to this and this. 
     
     
  
 
 
	 01:34:47
     ◼ 
      
     ►  
     Or like, "Hey, I've written a bunch of times here that I'm unhappy about this kind of thing 
     
     
  
 
 
	 01:34:52
     ◼ 
      
     ►  
     in like a bunch of different ways. So can I address that thing?" 
     
     
  
 
 
	 01:34:58
     ◼ 
      
     ►  
     - Hmm, okay, actually, interesting. I think I should frame this thought a bit differently, 
     
     
  
 
 
	 01:35:03
     ◼ 
      
     ►  
     because I have a notes file where when it occurs to me things that I may want to discuss 
     
     
  
 
 
	 01:35:08
     ◼ 
      
     ►  
     on the theme episode, I make a note of them. I guess this is just a different kind of 
     
     
  
 
 
	 01:35:13
     ◼ 
      
     ►  
     the same idea of like things about my own personal theme there. 
     
     
  
 
 
	 01:35:17
     ◼ 
      
     ►  
     But no, I like that as a have a place to collect what you notice about your life, 
     
     
  
 
 
	 01:35:25
     ◼ 
      
     ►  
     which also then just encourages you to notice more. 
     
     
  
 
 
	 01:35:28
     ◼ 
      
     ►  
     Might be a way to try to pitch it to people who are, 
     
     
  
 
 
	 01:35:32
     ◼ 
      
     ►  
     as the end of the year comes hurtling towards us, 
     
     
  
 
 
	 01:35:35
     ◼ 
      
     ►  
     and if they've never tried a theme before, 
     
     
  
 
 
	 01:35:37
     ◼ 
      
     ►  
     might want to think about that over the next two and a half months. 
     
     
  
 
 
	 01:35:41
     ◼ 
      
     ►  
     They're going to start a theme at the beginning of the year. 
     
     
  
 
 
	 01:35:43
     ◼ 
      
     ►  
     It's like just have a note somewhere to put down things that you notice about your life as a way to get started 
     
     
  
 
 
	 01:35:49
     ◼ 
      
     ►  
     And then you have something to look over and try to synthesize later. It's not about any particular moment 
     
     
  
 
 
	 01:35:56
     ◼ 
      
     ►  
     Okay. Yeah, that's interesting 
     
     
  
 
 
	 01:35:57
     ◼ 
      
     ►  
     I think that this is make something that's more likely to stick on because it's it's actually related to things that you've experienced 
     
     
  
 
 
	 01:36:04
     ◼ 
      
     ►  
     Mmm, and pushes it further away from the New Year's resolution idea because New Year's resolutions are typically created from whole cloth 
     
     
  
 
 
	 01:36:12
     ◼ 
      
     ►  
     like this just like I have this aspirational idea of myself that I would like to be this different kind of person and I'll 
     
     
  
 
 
	 01:36:17
     ◼ 
      
     ►  
     Just go and live that life 
     
     
  
 
 
	 01:36:18
     ◼ 
      
     ►  
     Where what I'm suggesting that you do is how do you live right now? 
     
     
  
 
 
	 01:36:22
     ◼ 
      
     ►  
     What frustrates you about things that are happening right now? What would you like to be better in your actual life right now? 
     
     
  
 
 
	 01:36:29
     ◼ 
      
     ►  
     Mmm and try and think about what that might be 
     
     
  
 
 
	 01:36:33
     ◼ 
      
     ►  
     It's related to something that's actually happening rather than something that you look forward to just imagine yourself as a different person 
     
     
  
 
 
	 01:36:39
     ◼ 
      
     ►  
     Yeah, that's a good distinction because, again, also the New Year's resolutions have 
     
     
  
 
 
	 01:36:44
     ◼ 
      
     ►  
     the feeling to me of homework assignments that you forgot about until the morning of 
     
     
  
 
 
	 01:36:48
     ◼ 
      
     ►  
     and it's like, "Oh no, I need to write an essay about what I'm going to do this 
     
     
  
 
 
	 01:36:50
     ◼ 
      
     ►  
     year." Whereas this makes me think of—it's actually kind of popping into my head the 
     
     
  
 
 
	 01:36:55
     ◼ 
      
     ►  
     way some researchers have tried to study how happy people are by messaging them at random 
     
     
  
 
 
	 01:37:01
     ◼ 
      
     ►  
     points throughout the weeks or the years of just like, "Hey, right now, how do you feel?" 
     
     
  
 
 
	 01:37:06
     ◼ 
      
     ►  
     And this is closer to that. 
     
     
  
 
 
	 01:37:07
     ◼ 
      
     ►  
     It's like a continual process of, hey, notice in your life 
     
     
  
 
 
	 01:37:10
     ◼ 
      
     ►  
     how you're feeling and just make a little note of it. 
     
     
  
 
 
	 01:37:12
     ◼ 
      
     ►  
     And then later, you can have some stuff to review. 
     
     
  
 
 
	 01:37:14
     ◼ 
      
     ►  
     So I like that. 
     
     
  
 
 
	 01:37:15
     ◼ 
      
     ►  
     I think that's good. 
     
     
  
 
 
	 01:37:16
     ◼ 
      
     ►  
     I think that's a good suggestion for people looking 
     
     
  
 
 
	 01:37:18
     ◼ 
      
     ►  
     for a place to onboard where to start instead of just trying 
     
     
  
 
 
	 01:37:23
     ◼ 
      
     ►  
     to create a theme out of whole cloth 
     
     
  
 
 
	 01:37:25
     ◼ 
      
     ►  
     after they've listened to the theme episode. 
     
     
  
 
 
 
 
	 01:37:29
     ◼ 
      
     ►  
     There's something I've got to tell you about before we wrap 
     
     
  
 
 
 
 
	 01:37:33
     ◼ 
      
     ►  
     I didn't know about this. 
     
     
  
 
 
	 01:37:35
     ◼ 
      
     ►  
     somehow seemed to have slipped by everyone. In iOS 16 there is now a Dvorak software keyboard 
     
     
  
 
 
	 01:37:44
     ◼ 
      
     ►  
     on the phone. Oh yeah? Let me grab my phone here. It's in the settings app right? You 
     
     
  
 
 
	 01:37:52
     ◼ 
      
     ►  
     have to go to settings and then I think it's general keyboard and then whatever your setting 
     
     
  
 
 
	 01:38:00
     ◼ 
      
     ►  
     as like English UK or English US or whatever. If you tap that you get the option of QWERTY, 
     
     
  
 
 
	 01:38:06
     ◼ 
      
     ►  
     AZERTY, QWERTS and DIVORAC. 
     
     
  
 
 
	 01:38:07
     ◼ 
      
     ►  
     Oh I see so it's not a new keyboard it's under English I would never have found that. 
     
     
  
 
 
	 01:38:12
     ◼ 
      
     ►  
     You could probably add it as a new keyboard? 
     
     
  
 
 
	 01:38:14
     ◼ 
      
     ►  
     No it's not that's what I was just trying to do. 
     
     
  
 
 
	 01:38:16
     ◼ 
      
     ►  
     Oh god I hate it. 
     
     
  
 
 
	 01:38:18
     ◼ 
      
     ►  
     Oh interesting. 
     
     
  
 
 
	 01:38:19
     ◼ 
      
     ►  
     Wow that's horrifying. 
     
     
  
 
 
	 01:38:20
     ◼ 
      
     ►  
     Do you still use DIVORAC? 
     
     
  
 
 
	 01:38:24
     ◼ 
      
     ►  
     Of course I use DIVORAC Myke. 
     
     
  
 
 
	 01:38:26
     ◼ 
      
     ►  
     So your keyboard on your desk is set in the Dvorak layout? 
     
     
  
 
 
	 01:38:30
     ◼ 
      
     ►  
     I feel like we haven't spoken about that 
     
     
  
 
 
	 01:38:31
     ◼ 
      
     ►  
     in a really long time. 
     
     
  
 
 
	 01:38:32
     ◼ 
      
     ►  
     I've kind of forgotten. 
     
     
  
 
 
	 01:38:34
     ◼ 
      
     ►  
     - Yes, well, my keyboard, if you look at it, 
     
     
  
 
 
	 01:38:37
     ◼ 
      
     ►  
     the current one I'm using is, it has a QWERTY layout, 
     
     
  
 
 
	 01:38:40
     ◼ 
      
     ►  
     but the keys are mapped to the Dvorak settings. 
     
     
  
 
 
	 01:38:43
     ◼ 
      
     ►  
     - You are a monster. 
     
     
  
 
 
	 01:38:45
     ◼ 
      
     ►  
     - No, no, no, I actually, I've decided that's best. 
     
     
  
 
 
	 01:38:48
     ◼ 
      
     ►  
     That's the better way to go. 
     
     
  
 
 
	 01:38:49
     ◼ 
      
     ►  
     I've stopped getting keyboards. 
     
     
  
 
 
	 01:38:51
     ◼ 
      
     ►  
     Look, I think there's only two ways to go. 
     
     
  
 
 
	 01:38:53
     ◼ 
      
     ►  
     QWERTY visual layout or blank, right? 
     
     
  
 
 
	 01:38:58
     ◼ 
      
     ►  
     Like nothing on any of the keys. 
     
     
  
 
 
	 01:39:01
     ◼ 
      
     ►  
     - Why can't you use the Dvorak visual layout? 
     
     
  
 
 
	 01:39:03
     ◼ 
      
     ►  
     I think I could build you a keyboard one day 
     
     
  
 
 
	 01:39:08
     ◼ 
      
     ►  
     that would be laid out this way. 
     
     
  
 
 
	 01:39:10
     ◼ 
      
     ►  
     - Yeah, okay, I guess here's, 
     
     
  
 
 
	 01:39:12
     ◼ 
      
     ►  
     I was just trying to articulate why. 
     
     
  
 
 
	 01:39:15
     ◼ 
      
     ►  
     After years and years of doing all sorts of different things, 
     
     
  
 
 
	 01:39:17
     ◼ 
      
     ►  
     why have I settled on this? 
     
     
  
 
 
	 01:39:19
     ◼ 
      
     ►  
     The answer is that keyboard shortcuts are funny 
     
     
  
 
 
	 01:39:24
     ◼ 
      
     ►  
     in a lot of applications. 
     
     
  
 
 
	 01:39:27
     ◼ 
      
     ►  
     So there's something in the system level 
     
     
  
 
 
	 01:39:31
     ◼ 
      
     ►  
     where a keyboard shortcut will either trigger based 
     
     
  
 
 
	 01:39:34
     ◼ 
      
     ►  
     on the letter that the key represents 
     
     
  
 
 
	 01:39:37
     ◼ 
      
     ►  
     or it will trigger based on the location that the key is. 
     
     
  
 
 
	 01:39:41
     ◼ 
      
     ►  
     You can see this explicitly in some things like Final Cut 
     
     
  
 
 
	 01:39:45
     ◼ 
      
     ►  
     where you can reprogram the key based on letter 
     
     
  
 
 
	 01:39:49
     ◼ 
      
     ►  
     or location. So I guess the way I use it is that it is I'm just not having I'm struggling 
     
     
  
 
 
	 01:39:55
     ◼ 
      
     ►  
     trying to think of a specific example that's not final cut, but I know it comes up where 
     
     
  
 
 
	 01:39:59
     ◼ 
      
     ►  
     it's useful to be able to see what does the keyboard look like for everyone in the whole 
     
     
  
 
 
	 01:40:06
     ◼ 
      
     ►  
     wide world, sometimes for keyboard shortcuts like that does come up where I am glad that 
     
     
  
 
 
	 01:40:11
     ◼ 
      
     ►  
     the layout is a QWERTY layout. And since I touch type, I don't look at the keyboard, 
     
     
  
 
 
	 01:40:16
     ◼ 
      
     ►  
     It doesn't matter when I'm actually writing what the keys look like at all, which is why 
     
     
  
 
 
	 01:40:20
     ◼ 
      
     ►  
     I have gone with a completely blank keyboard sometimes, which I do think is cool, but can 
     
     
  
 
 
	 01:40:25
     ◼ 
      
     ►  
     be annoying for keyboard shortcuts. 
     
     
  
 
 
	 01:40:27
     ◼ 
      
     ►  
     I have a question for you. 
     
     
  
 
 
 
	 01:40:29
     ◼ 
      
     ►  
     Well, because one, I'll come back to the keyboard shortcuts thing because I know how to fix 
     
     
  
 
 
 
	 01:40:33
     ◼ 
      
     ►  
     Depends on what you use. 
     
     
  
 
 
	 01:40:34
     ◼ 
      
     ►  
     Anyway, when you are using your Dvorak layout, what is the copy command? 
     
     
  
 
 
	 01:40:40
     ◼ 
      
     ►  
     Okay, so for me, it's command I is what it would look like on a query keyboard. That's 
     
     
  
 
 
	 01:40:47
     ◼ 
      
     ►  
     copy. Oh, okay. Yes, I know what you're saying. Now you are actually hitting still command 
     
     
  
 
 
	 01:40:54
     ◼ 
      
     ►  
     C though is what I'm asking. That's more work, but it's in a different place. Okay, it's 
     
     
  
 
 
	 01:40:58
     ◼ 
      
     ►  
     in a different place. So one of the things that exists now, which I don't know if it's 
     
     
  
 
 
	 01:41:03
     ◼ 
      
     ►  
     actually a good idea for any other lunatics who might want to switch their keyboard layouts. 
     
     
  
 
 
	 01:41:09
     ◼ 
      
     ►  
     One of the things that computers have gotten better at over years is there are now explicit 
     
     
  
 
 
	 01:41:13
     ◼ 
      
     ►  
     layouts which are on the Mac, it's called something like Dvorak preserve shortcuts. 
     
     
  
 
 
	 01:41:20
     ◼ 
      
     ►  
     So it's like you type in Dvorak, but it totally ignores the keyboard layout for all keyboard 
     
     
  
 
 
 
	 01:41:26
     ◼ 
      
     ►  
     It would be like command J looking at the Dvorak layout right in that scenario where 
     
     
  
 
 
	 01:41:33
     ◼ 
      
     ►  
     it's preserving the location of the key. 
     
     
  
 
 
	 01:41:36
     ◼ 
      
     ►  
     Yeah, it's funny, it's like I can't – look, I is C. What's the complication here? 
     
     
  
 
 
	 01:41:42
     ◼ 
      
     ►  
     I'm looking at a Dvorak layout on Wikipedia and like, you know, J is in the location that 
     
     
  
 
 
	 01:41:48
     ◼ 
      
     ►  
     C is in a QWERTY keyboard. 
     
     
  
 
 
	 01:41:51
     ◼ 
      
     ►  
     And so you would hit, I guess, command J, because what it's saying is, the way I've 
     
     
  
 
 
	 01:41:56
     ◼ 
      
     ►  
     understood the way you've explained that, is like, it's keeping the physical location 
     
     
  
 
 
	 01:42:00
     ◼ 
      
     ►  
     of the key no matter what the key actually says it is. 
     
     
  
 
 
	 01:42:03
     ◼ 
      
     ►  
     Yeah, okay. So I just looked it up like what is the actual thing called in the system setting? 
     
     
  
 
 
	 01:42:08
     ◼ 
      
     ►  
     So on the Mac, you can set it as something called Dvorak QWERTY command. So I presume 
     
     
  
 
 
	 01:42:14
     ◼ 
      
     ►  
     that what it's doing there is whenever you hit command, it ignores the Dvorak layout 
     
     
  
 
 
	 01:42:17
     ◼ 
      
     ►  
     and reverts to the QWERTY layout. 
     
     
  
 
 
	 01:42:19
     ◼ 
      
     ►  
     That seems horribly complicated. Yeah, all of this to say, one day, because I want to 
     
     
  
 
 
	 01:42:24
     ◼ 
      
     ►  
     build your keyboard one day, just for fun. I would choose a keyboard where I can change 
     
     
  
 
 
	 01:42:30
     ◼ 
      
     ►  
     the programming of the keyboard. The boards that I build these days tend to all support 
     
     
  
 
 
	 01:42:36
     ◼ 
      
     ►  
     a piece of software called VIA which overrides the layout on the keyboard. So it's not relying 
     
     
  
 
 
	 01:42:44
     ◼ 
      
     ►  
     on the Mac. So if you hit any shortcut in any app it's going to register as the key 
     
     
  
 
 
	 01:42:53
     ◼ 
      
     ►  
     that is on the Dvorak layout. It doesn't matter that you would try and do some kind of key 
     
     
  
 
 
	 01:42:59
     ◼ 
      
     ►  
     binding and software. Does that make sense? Yes. Right? So like one of the 
     
     
  
 
 
	 01:43:04
     ◼ 
      
     ►  
     things you're saying is like for example if you hit some kind of keyboard 
     
     
  
 
 
	 01:43:08
     ◼ 
      
     ►  
     shortcut in Final Cut it's going to assume that's QWERTY even if you've set 
     
     
  
 
 
	 01:43:12
     ◼ 
      
     ►  
     to Dvorak because there's like a software in between the two of them 
     
     
  
 
 
	 01:43:15
     ◼ 
      
     ►  
     that's trying to communicate it. But I'm saying the hardware of the keyboard 
     
     
  
 
 
	 01:43:20
     ◼ 
      
     ►  
     would communicate to the computer so I am convinced it would get it correctly 
     
     
  
 
 
	 01:43:27
     ◼ 
      
     ►  
     it would always hit correctly. 
     
     
  
 
 
	 01:43:29
     ◼ 
      
     ►  
     - Okay, I feel like I'm not 100% understanding, 
     
     
  
 
 
	 01:43:32
     ◼ 
      
     ►  
     but there's also so many layers here. 
     
     
  
 
 
	 01:43:35
     ◼ 
      
     ►  
     One of the other problems is, 
     
     
  
 
 
	 01:43:36
     ◼ 
      
     ►  
     I don't know if any of the cortexes 
     
     
  
 
 
	 01:43:37
     ◼ 
      
     ►  
     are able to help me out with this, 
     
     
  
 
 
	 01:43:39
     ◼ 
      
     ►  
     if anyone's already done this. 
     
     
  
 
 
	 01:43:40
     ◼ 
      
     ►  
     Here's the maximum level of crazy, right? 
     
     
  
 
 
	 01:43:43
     ◼ 
      
     ►  
     I use the Dvorak layout, 
     
     
  
 
 
	 01:43:45
     ◼ 
      
     ►  
     but I've learned all the keyboard commands, 
     
     
  
 
 
	 01:43:48
     ◼ 
      
     ►  
     all the basic system commands, 
     
     
  
 
 
	 01:43:50
     ◼ 
      
     ►  
     with the Dvorak layout, 
     
     
  
 
 
	 01:43:51
     ◼ 
      
     ►  
     so I'm not pressing the regular buttons 
     
     
  
 
 
	 01:43:53
     ◼ 
      
     ►  
     that everybody knows. 
     
     
  
 
 
	 01:43:55
     ◼ 
      
     ►  
     However, in Final Cut, the way that the programmers laid out all the keyboard shortcuts that do 
     
     
  
 
 
	 01:44:03
     ◼ 
      
     ►  
     things, it makes sense physically where they are on the keyboard. 
     
     
  
 
 
	 01:44:08
     ◼ 
      
     ►  
     Whereas like Command Copy, Command Paste, like it doesn't really matter if those two 
     
     
  
 
 
	 01:44:11
     ◼ 
      
     ►  
     are next to each other. 
     
     
  
 
 
	 01:44:13
     ◼ 
      
     ►  
     But there's tons of stuff in Final Cut, which is like trim from the start of the clips, 
     
     
  
 
 
	 01:44:17
     ◼ 
      
     ►  
     trim from the end of the clip. 
     
     
  
 
 
	 01:44:19
     ◼ 
      
     ►  
     And you want to have those keys like next to each other on opposite sides. 
     
     
  
 
 
	 01:44:22
     ◼ 
      
     ►  
     There's a lot of like physicality of this. 
     
     
  
 
 
	 01:44:24
     ◼ 
      
     ►  
     So in Final Cut, I've been trying to slowly build up my own custom mapping, which is to 
     
     
  
 
 
	 01:44:31
     ◼ 
      
     ►  
     make it so that when I am using the Dvorak layout, it's still acting as though it's a 
     
     
  
 
 
	 01:44:37
     ◼ 
      
     ►  
     QWERTY layout just for the Final Cut shortcuts. 
     
     
  
 
 
	 01:44:41
     ◼ 
      
     ►  
     So every time I try to learn a new Final Cut shortcut, I try to go in and like, change 
     
     
  
 
 
 
	 01:44:47
     ◼ 
      
     ►  
     But I've wondered like, has someone just done this? 
     
     
  
 
 
	 01:44:49
     ◼ 
      
     ►  
     is there a QWERTY for Dvorak remapping of all the Final Cut shortcuts that someone has just done? 
     
     
  
 
 
	 01:44:55
     ◼ 
      
     ►  
     Because I could tell like I'm getting into a situation of, "Oh, this is a little bit 
     
     
  
 
 
	 01:44:59
     ◼ 
      
     ►  
     inconsistent." So anyway, I've just wondered if like a person has done that work for me. 
     
     
  
 
 
	 01:45:02
     ◼ 
      
     ►  
     And there's also some weird system stuff that I've done with a few of the shortcuts. 
     
     
  
 
 
	 01:45:07
     ◼ 
      
     ►  
     Anyway, there's a bunch of these weird little problems. And I also have this 
     
     
  
 
 
	 01:45:10
     ◼ 
      
     ►  
     minor annoyance. Listen, before I say this annoyance, I understand. I understand why 
     
     
  
 
 
	 01:45:15
     ◼ 
      
     ►  
     Apple did it. You don't have to leave me comments for why they did it. I 
     
     
  
 
 
	 01:45:18
     ◼ 
      
     ►  
     understand. It's still annoying. But it used to be that Apple, when you had 
     
     
  
 
 
	 01:45:23
     ◼ 
      
     ►  
     different keyboard layouts, it had a little country flag to represent each of 
     
     
  
 
 
	 01:45:28
     ◼ 
      
     ►  
     the layouts. And so the US keyboard layout had a little US flag. And the 
     
     
  
 
 
	 01:45:33
     ◼ 
      
     ►  
     Dvorak one was just DV. And they made the change so that they don't use country 
     
     
  
 
 
	 01:45:41
     ◼ 
      
     ►  
     flags for languages. Now again, I get it, but it actually causes a huge annoyance because in my menu 
     
     
  
 
 
	 01:45:48
     ◼ 
      
     ►  
     bar I want to see which layout the keyboard is using. Is it using Dvorak or is it using US? 
     
     
  
 
 
	 01:45:54
     ◼ 
      
     ►  
     Because sometimes I do switch between those two layouts and now I can't visually see instantly 
     
     
  
 
 
	 01:46:00
     ◼ 
      
     ►  
     that the keyboard is on US layout versus Dvorak layout and it's maddening because it's just a 
     
     
  
 
 
	 01:46:06
     ◼ 
      
     ►  
     a little box and the box either says DV or it says US. And like, no, it was so much better when it was 
     
     
  
 
 
	 01:46:13
     ◼ 
      
     ►  
     black DV and that's Dvorak. And then there was a little American flag when it's the US layout. 
     
     
  
 
 
	 01:46:18
     ◼ 
      
     ►  
     So it's super frustrating. But anyway, all of that aside, I'm looking at this Dvorak layout on my 
     
     
  
 
 
	 01:46:23
     ◼ 
      
     ►  
     phone, and I hate everything about it. I don't think Dvorak was made for phones. I think all 
     
     
  
 
 
	 01:46:29
     ◼ 
      
     ►  
     All of the advantages for typing Dvorak with two hands, I don't feel like they translate 
     
     
  
 
 
	 01:46:36
     ◼ 
      
     ►  
     for two thumbs on a phone. 
     
     
  
 
 
	 01:46:38
     ◼ 
      
     ►  
     I think maybe the QWERTY layout is actually superior for the phone. 
     
     
  
 
 
	 01:46:41
     ◼ 
      
     ►  
     Yeah, all of the vowels are next to each other on the Dvorak one. 
     
     
  
 
 
	 01:46:45
     ◼ 
      
     ►  
     I'm just trying to type some words here. 
     
     
  
 
 
	 01:46:47
     ◼ 
      
     ►  
     By the way, I have two applications for you that can restore those flags. 
     
     
  
 
 
 
	 01:46:53
     ◼ 
      
     ►  
     Yeah, one is called Keyboard Switcheroo. 
     
     
  
 
 
	 01:46:56
     ◼ 
      
     ►  
     The other is called Colorful Input Menu Flags. 
     
     
  
 
 
	 01:46:59
     ◼ 
      
     ►  
     'cause they're both in the Mac App Store. 
     
     
  
 
 
 
 
	 01:47:03
     ◼ 
      
     ►  
     - Two enterprising developers who created applications 
     
     
  
 
 
	 01:47:06
     ◼ 
      
     ►  
     to bring that back. 
     
     
  
 
 
	 01:47:08
     ◼ 
      
     ►  
     I'd like to thank Glenn Fleischman at Macworld 
     
     
  
 
 
	 01:47:11
     ◼ 
      
     ►  
     for writing the article too. 
     
     
  
 
 
	 01:47:12
     ◼ 
      
     ►  
     'Cause I know I'd seen these, 
     
     
  
 
 
	 01:47:14
     ◼ 
      
     ►  
     and so I just did a quick Google while you were upset. 
     
     
  
 
 
	 01:47:17
     ◼ 
      
     ►  
     And actually on keyboard switcheroo, 
     
     
  
 
 
	 01:47:20
     ◼ 
      
     ►  
     they have in their app screenshots, 
     
     
  
 
 
	 01:47:23
     ◼ 
      
     ►  
     US, French, German, Dvorak. 
     
     
  
 
 
	 01:47:25
     ◼ 
      
     ►  
     So you get a little US flag 
     
     
  
 
 
	 01:47:28
     ◼ 
      
     ►  
     a little DV for Dvorak. 
     
     
  
 
 
	 01:47:29
     ◼ 
      
     ►  
     Oh good, I'm glad to know I'm not the only person who was annoyed by this. 
     
     
  
 
 
	 01:47:32
     ◼ 
      
     ►  
     So I'll definitely check those out. 
     
     
  
 
 
 
	 01:47:35
     ◼ 
      
     ►  
     You can always rely on Mac developers to solve the little annoyances. 
     
     
  
 
 
	 01:47:38
     ◼ 
      
     ►  
     Yeah you can. 
     
     
  
 
 
	 01:47:39
     ◼ 
      
     ►  
     What I was just thinking here is with iOS, I think Qwerty on the phone, I think it does 
     
     
  
 
 
 
	 01:47:46
     ◼ 
      
     ►  
     I feel like the weirdness of the way Qwerty is laid out to sort of slow you down, although 
     
     
  
 
 
	 01:47:52
     ◼ 
      
     ►  
     I think that that's overplayed as a story of why it exists. 
     
     
  
 
 
	 01:47:55
     ◼ 
      
     ►  
     I think that actually works as an advantage for typing with your thumbs on a tiny phone 
     
     
  
 
 
	 01:47:59
     ◼ 
      
     ►  
     screen, but I am glad that this exists for the iPad. 
     
     
  
 
 
	 01:48:03
     ◼ 
      
     ►  
     I might actually switch it on the iPad, because that's the place where it's been annoying. 
     
     
  
 
 
	 01:48:06
     ◼ 
      
     ►  
     Big keyboard. 
     
     
  
 
 
	 01:48:07
     ◼ 
      
     ►  
     And that's more like a larger keyboard there is what you're more used to for Dvorak, right? 
     
     
  
 
 
	 01:48:12
     ◼ 
      
     ►  
     Bigger keyboards like a computer keyboard. 
     
     
  
 
 
	 01:48:15
     ◼ 
      
     ►  
     Can you get MacBooks in Dvorak? 
     
     
  
 
 
	 01:48:17
     ◼ 
      
     ►  
     I don't think so. 
     
     
  
 
 
	 01:48:18
     ◼ 
      
     ►  
     I don't think they sell that. 
     
     
  
 
 
	 01:48:21
     ◼ 
      
     ►  
     I'm pretty sure you have to just pick between American layout and the English layout when 
     
     
  
 
 
	 01:48:26
     ◼ 
      
     ►  
     I'm buying the keyboards, which is the thing I'm always very careful about. 
     
     
  
 
 
	 01:48:30
     ◼ 
      
     ►  
     Like, please give me the US layout one. 
     
     
  
 
 
	 01:48:32
     ◼ 
      
     ►  
     I do not want the UK layout one. 
     
     
  
 
 
	 01:48:34
     ◼ 
      
     ►  
     There's no little "huh" there. 
     
     
  
 
 
	 01:48:35
     ◼ 
      
     ►  
     Like the American enter is clearly superior. 
     
     
  
 
 
	 01:48:38
     ◼ 
      
     ►  
     I build all of my keyboards in NC. 
     
     
  
 
 
	 01:48:40
     ◼ 
      
     ►  
     My laptop is in the British layout, but I think in the future I might try and remember 
     
     
  
 
 
	 01:48:46
     ◼ 
      
     ►  
     to just order them in US layout because that's what I'm much more used to now. 
     
     
  
 
 
	 01:48:49
     ◼ 
      
     ►  
     Wait a second. 
     
     
  
 
 
	 01:48:50
     ◼ 
      
     ►  
     - You build your keyboards in the US layout, NC layout. 
     
     
  
 
 
 
	 01:48:55
     ◼ 
      
     ►  
     - But you get the MacBook keyboards with the English layout? 
     
     
  
 
 
	 01:48:58
     ◼ 
      
     ►  
     - I don't think about it when I'm buying them. 
     
     
  
 
 
	 01:49:01
     ◼ 
      
     ►  
     I think I will now in the future, right? 
     
     
  
 
 
	 01:49:03
     ◼ 
      
     ►  
     'Cause I'm getting more annoyed because I'm switching 
     
     
  
 
 
	 01:49:06
     ◼ 
      
     ►  
     between something I'm not used to. 
     
     
  
 
 
	 01:49:08
     ◼ 
      
     ►  
     And the keyboards that I build, 
     
     
  
 
 
	 01:49:10
     ◼ 
      
     ►  
     most of them are just more available 
     
     
  
 
 
	 01:49:13
     ◼ 
      
     ►  
     and easier to build in NC layout. 
     
     
  
 
 
	 01:49:14
     ◼ 
      
     ►  
     So I'm just used to it now, so the US English layout. 
     
     
  
 
 
	 01:49:18
     ◼ 
      
     ►  
     But I never think about it. 
     
     
  
 
 
	 01:49:19
     ◼ 
      
     ►  
     and obviously the default on the British Mac store 
     
     
  
 
 
	 01:49:23
     ◼ 
      
     ►  
     is the British layout, so I don't think about it 
     
     
  
 
 
	 01:49:25
     ◼ 
      
     ►  
     and I just buy what's the default, 
     
     
  
 
 
	 01:49:28
     ◼ 
      
     ►  
     but I think I wanna start changing it. 
     
     
  
 
 
	 01:49:30
     ◼ 
      
     ►  
     It doesn't look like they do Dvorak, 
     
     
  
 
 
	 01:49:31
     ◼ 
      
     ►  
     it's not in their options. 
     
     
  
 
 
	 01:49:32
     ◼ 
      
     ►  
     - I don't think I'll be using this on my phone. 
     
     
  
 
 
	 01:49:35
     ◼ 
      
     ►  
     I think it's worse on the phone, 
     
     
  
 
 
	 01:49:36
     ◼ 
      
     ►  
     but I will definitely try it out on the iPad 
     
     
  
 
 
	 01:49:38
     ◼ 
      
     ►  
     when the iPad comes out, 
     
     
  
 
 
	 01:49:39
     ◼ 
      
     ►  
     'cause that's where it's more of a problem, 
     
     
  
 
 
	 01:49:41
     ◼ 
      
     ►  
     but I mostly don't really think about it very much 
     
     
  
 
 
	 01:49:44
     ◼ 
      
     ►  
     'cause it's just been so long, 
     
     
  
 
 
	 01:49:46
     ◼ 
      
     ►  
     like this has been my entire life. 
     
     
  
 
 
	 01:49:48
     ◼ 
      
     ►  
     The last time it caused me problems was when I worked at a school and had to switch between 
     
     
  
 
 
	 01:49:53
     ◼ 
      
     ►  
     the QWERTY and Dvorak typing keyboards a bunch for using different computers, but even then 
     
     
  
 
 
	 01:49:57
     ◼ 
      
     ►  
     it was not the worst. 
     
     
  
 
 
	 01:49:58
     ◼ 
      
     ►  
     Like, I just learned to touch type while not looking and then sort of type while looking 
     
     
  
 
 
	 01:50:03
     ◼ 
      
     ►  
     at the keyboard as two different modes that my brain could switch between. 
     
     
  
 
 
	 01:50:06
     ◼ 
      
     ►  
     But I wonder why on earth they added Dvorak to the— 
     
     
  
 
 
 
	 01:50:11
     ◼ 
      
     ►  
     —to iOS 16 now. 
     
     
  
 
 
	 01:50:12
     ◼ 
      
     ►  
     It's a very funny, like— 
     
     
  
 
 
	 01:50:13
     ◼ 
      
     ►  
     It's strange. 
     
     
  
 
 
	 01:50:14
     ◼ 
      
     ►  
     Why did it take 16 attempts? 
     
     
  
 
 
	 01:50:15
     ◼ 
      
     ►  
     You know what I mean? 
     
     
  
 
 
	 01:50:17
     ◼ 
      
     ►  
     why 16 versions in? Is Dvorak having some resurgence that we're not aware of? Like why? 
     
     
  
 
 
	 01:50:23
     ◼ 
      
     ►  
     It is odd, right? Yeah, I'm not even sure if I was suggesting to someone now who had RSI problems if 
     
     
  
 
 
	 01:50:31
     ◼ 
      
     ►  
     Dvorak is the way to go. I always forget, I think it's Colmak or something? There's another one 
     
     
  
 
 
	 01:50:37
     ◼ 
      
     ►  
     which is definitely worth investigating if you're learning now versus Dvorak. And it's also a 
     
     
  
 
 
	 01:50:42
     ◼ 
      
     ►  
     a feature that I don't know why but it took all the way until now for people to notice 
     
     
  
 
 
	 01:50:47
     ◼ 
      
     ►  
     it it seems like. There's been a bunch of articles written because it was kind of discovered 
     
     
  
 
 
	 01:50:51
     ◼ 
      
     ►  
     but that means it went through the whole beta process without people seeming to know about 
     
     
  
 
 
	 01:50:56
     ◼ 
      
     ►  
     it. I don't know. 
     
     
  
 
 
	 01:50:57
     ◼ 
      
     ►  
     The thing that's also confusing about it is if they did add... why didn't they add something 
     
     
  
 
 
	 01:51:03
     ◼ 
      
     ►  
     like Colmac? I mean look I don't know a lot about programming iOS to add additional keyboard 
     
     
  
 
 
	 01:51:08
     ◼ 
      
     ►  
     layouts but if you're going to add something like Dvorak how hard is it to also add the 
     
     
  
 
 
	 01:51:14
     ◼ 
      
     ►  
     other popular one of these if you're gonna do it? 
     
     
  
 
 
	 01:51:17
     ◼ 
      
     ►  
     Well isn't Dvorak like the popular one though? 
     
     
  
 
 
	 01:51:20
     ◼ 
      
     ►  
     I don't know if that's true or not. 
     
     
  
 
 
	 01:51:22
     ◼ 
      
     ►  
     I feel like it must be or at least in the requests that they get or some high up executive 
     
     
  
 
 
	 01:51:27
     ◼ 
      
     ►  
     like Tim Cook is a Dvorak guy and like he'd had enough you know? 
     
     
  
 
 
	 01:51:31
     ◼ 
      
     ►  
     Yeah I mean with Dvorak I just always feel like there's dozens of us. 
     
     
  
 
 
 
 
	 01:51:38
     ◼ 
      
     ►  
     - You've seen the Godfather, right? 
     
     
  
 
 
	 01:51:40
     ◼ 
      
     ►  
     - No. - What? 
     
     
  
 
 
	 01:51:42
     ◼ 
      
     ►  
     - I've not seen the Godfather. 
     
     
  
 
 
	 01:51:44
     ◼ 
      
     ►  
     I've just never really been that interested. 
     
     
  
 
 
	 01:51:46
     ◼ 
      
     ►  
     - Okay, well, in the Godfather, 
     
     
  
 
 
	 01:51:49
     ◼ 
      
     ►  
     this is pointless for you now, 
     
     
  
 
 
	 01:51:50
     ◼ 
      
     ►  
     but in the Godfather, there is a moment, 
     
     
  
 
 
	 01:51:53
     ◼ 
      
     ►  
     I think Godfather 2, I think it's Godfather 2, 
     
     
  
 
 
	 01:51:56
     ◼ 
      
     ►  
     which I assume you also have not seen, 
     
     
  
 
 
	 01:51:58
     ◼ 
      
     ►  
     because why would you have, right? 
     
     
  
 
 
	 01:51:59
     ◼ 
      
     ►  
     - Yeah, that's a correct, yeah, I've seen the third one, 
     
     
  
 
 
	 01:52:01
     ◼ 
      
     ►  
     but not the first or the second one. 
     
     
  
 
 
	 01:52:03
     ◼ 
      
     ►  
     - You're joking, right? 
     
     
  
 
 
	 01:52:04
     ◼ 
      
     ►  
     - I am joking, yes. - Good, good, good, good. 
     
     
  
 
 
	 01:52:06
     ◼ 
      
     ►  
     There is a moment where a bunch of things happen 
     
     
  
 
 
	 01:52:08
     ◼ 
      
     ►  
     and the line is, "We're settling all family business." 
     
     
  
 
 
	 01:52:13
     ◼ 
      
     ►  
     There's a line about settling all family business. 
     
     
  
 
 
	 01:52:15
     ◼ 
      
     ►  
     And I feel like that is today's episode. 
     
     
  
 
 
	 01:52:17
     ◼ 
      
     ►  
     We have just settled a bunch of business, right? 
     
     
  
 
 
	 01:52:20
     ◼ 
      
     ►  
     A lot of follow-up has been dealt with, right? 
     
     
  
 
 
	 01:52:23
     ◼ 
      
     ►  
     We've spoken about Dvorak, right? 
     
     
  
 
 
	 01:52:25
     ◼ 
      
     ►  
     Like we're just settling the business 
     
     
  
 
 
	 01:52:27
     ◼ 
      
     ►  
     before the year ends for us. 
     
     
  
 
 
	 01:52:28
     ◼ 
      
     ►  
     Like very good episodes, now we go to the specials. 
     
     
  
 
 
	 01:52:31
     ◼ 
      
     ►  
     - Yeah, that's true. 
     
     
  
 
 
	 01:52:32
     ◼ 
      
     ►  
     - There are two more pieces of family business 
     
     
  
 
 
	 01:52:34
     ◼ 
      
     ►  
     that I would like to settle before we finish today. 
     
     
  
 
 
	 01:52:37
     ◼ 
      
     ►  
     One is to thank every Cortexan who donated 
     
     
  
 
 
	 01:52:40
     ◼ 
      
     ►  
     to our Saint Jude campaign. 
     
     
  
 
 
	 01:52:41
     ◼ 
      
     ►  
     Overall, throughout the month of September, 
     
     
  
 
 
	 01:52:44
     ◼ 
      
     ►  
     we raised $706,397.10. 
     
     
  
 
 
	 01:52:49
     ◼ 
      
     ►  
     All of us absolutely astounded. 
     
     
  
 
 
	 01:52:53
     ◼ 
      
     ►  
     This is the most money we've ever raised. 
     
     
  
 
 
	 01:52:55
     ◼ 
      
     ►  
     - It's unbelievable. 
     
     
  
 
 
	 01:52:57
     ◼ 
      
     ►  
     It's an incomprehensible amount of money. 
     
     
  
 
 
	 01:53:00
     ◼ 
      
     ►  
     So we have now not only passed $2 million raised 
     
     
  
 
 
	 01:53:05
     ◼ 
      
     ►  
     in the last four years, 
     
     
  
 
 
	 01:53:06
     ◼ 
      
     ►  
     we have now hit $2.2 million raised in the last four years. 
     
     
  
 
 
	 01:53:11
     ◼ 
      
     ►  
     - It's incredible. 
     
     
  
 
 
	 01:53:15
     ◼ 
      
     ►  
     It's just incredible to think that the relay listenership 
     
     
  
 
 
	 01:53:18
     ◼ 
      
     ►  
     has raised that much money for St. Jude. 
     
     
  
 
 
	 01:53:22
     ◼ 
      
     ►  
     So yeah, thank you to everyone who donated to the campaign. 
     
     
  
 
 
	 01:53:27
     ◼ 
      
     ►  
     That's completely mind blowing. 
     
     
  
 
 
	 01:53:29
     ◼ 
      
     ►  
     You know, it seemed for a while that it was like, 
     
     
  
 
 
	 01:53:31
     ◼ 
      
     ►  
     okay, we're gonna raise a lot of money, 
     
     
  
 
 
	 01:53:33
     ◼ 
      
     ►  
     we're gonna meet our goal, 
     
     
  
 
 
	 01:53:34
     ◼ 
      
     ►  
     but we're probably not going to exceed. 
     
     
  
 
 
	 01:53:36
     ◼ 
      
     ►  
     Because there is definite economic challenges 
     
     
  
 
 
	 01:53:39
     ◼ 
      
     ►  
     right now around the world, 
     
     
  
 
 
	 01:53:41
     ◼ 
      
     ►  
     and we were thinking, okay, that's gonna be that. 
     
     
  
 
 
	 01:53:43
     ◼ 
      
     ►  
     And then it just, the last couple of days even, 
     
     
  
 
 
	 01:53:47
     ◼ 
      
     ►  
     like the last couple of weeks, it just exploded 
     
     
  
 
 
	 01:53:49
     ◼ 
      
     ►  
     'cause people were just getting their final totals in. 
     
     
  
 
 
	 01:53:51
     ◼ 
      
     ►  
     We had to, we were gonna end it on September 30th, 
     
     
  
 
 
	 01:53:54
     ◼ 
      
     ►  
     but we extended it to October 3rd, 
     
     
  
 
 
	 01:53:56
     ◼ 
      
     ►  
     because the money was just piling in in the last day. 
     
     
  
 
 
	 01:54:00
     ◼ 
      
     ►  
     It was like, all right, 
     
     
  
 
 
	 01:54:01
     ◼ 
      
     ►  
     we're gonna leave this open a couple of days. 
     
     
  
 
 
	 01:54:02
     ◼ 
      
     ►  
     I think like in the last day or two, 
     
     
  
 
 
	 01:54:04
     ◼ 
      
     ►  
     we raised like an extra $110,000 or something. 
     
     
  
 
 
	 01:54:07
     ◼ 
      
     ►  
     So it was like, probably should keep this open. 
     
     
  
 
 
	 01:54:10
     ◼ 
      
     ►  
     - Yeah, that's worth leaving open an extra day. 
     
     
  
 
 
	 01:54:13
     ◼ 
      
     ►  
     - But overall, this year's campaign was fantastic. 
     
     
  
 
 
	 01:54:16
     ◼ 
      
     ►  
     We learned a lot and for me, 
     
     
  
 
 
	 01:54:18
     ◼ 
      
     ►  
     it was just so incredibly rewarding and fun 
     
     
  
 
 
	 01:54:22
     ◼ 
      
     ►  
     to be able to be back in Memphis for the Podcastathon, 
     
     
  
 
 
	 01:54:24
     ◼ 
      
     ►  
     which was a great success. 
     
     
  
 
 
	 01:54:26
     ◼ 
      
     ►  
     The whole video is on YouTube. 
     
     
  
 
 
	 01:54:27
     ◼ 
      
     ►  
     I'll put a link in the show notes 
     
     
  
 
 
	 01:54:28
     ◼ 
      
     ►  
     if people don't want to watch it. 
     
     
  
 
 
	 01:54:30
     ◼ 
      
     ►  
     A very kind comment to timestamp the whole thing. 
     
     
  
 
 
	 01:54:33
     ◼ 
      
     ►  
     So if you want to jump around to different segments 
     
     
  
 
 
	 01:54:35
     ◼ 
      
     ►  
     or whatever, that's all in there 
     
     
  
 
 
	 01:54:37
     ◼ 
      
     ►  
     in one of the comments in the YouTube video. 
     
     
  
 
 
	 01:54:39
     ◼ 
      
     ►  
     But yeah, it was truly fantastic 
     
     
  
 
 
	 01:54:42
     ◼ 
      
     ►  
     and we achieved something that I just did not think 
     
     
  
 
 
	 01:54:45
     ◼ 
      
     ►  
     was going to be possible. 
     
     
  
 
 
	 01:54:46
     ◼ 
      
     ►  
     And once again, the Relay FM community 
     
     
  
 
 
	 01:54:48
     ◼ 
      
     ►  
     has gone and surprised us. 
     
     
  
 
 
	 01:54:51
     ◼ 
      
     ►  
     Cortexmerch.com. 
     
     
  
 
 
	 01:54:52
     ◼ 
      
     ►  
     This is the second thing, final thing. 
     
     
  
 
 
	 01:54:54
     ◼ 
      
     ►  
     And what's the final piece? 
     
     
  
 
 
	 01:54:56
     ◼ 
      
     ►  
     - Is subtletys and subtle sweaters. 
     
     
  
 
 
	 01:54:58
     ◼ 
      
     ►  
     We're reminding you, maybe you have now finished your commute 
     
     
  
 
 
	 01:55:00
     ◼ 
      
     ►  
     and you're walking into the office or whatever, 
     
     
  
 
 
	 01:55:02
     ◼ 
      
     ►  
     and you didn't do what I asked you to do earlier, 
     
     
  
 
 
	 01:55:04
     ◼ 
      
     ►  
     so now you can get your phone out 
     
     
  
 
 
	 01:55:06
     ◼ 
      
     ►  
     and you can go to cortexmerch.com 
     
     
  
 
 
	 01:55:08
     ◼ 
      
     ►  
     and buy yourself a subtlety or subtle sweater or both 
     
     
  
 
 
	 01:55:10
     ◼ 
      
     ►  
     or whatever you want. 
     
     
  
 
 
	 01:55:12
     ◼ 
      
     ►  
     - Unlike the St. Jude fundraiser, this will not be extended. 
     
     
  
 
 
	 01:55:16
     ◼ 
      
     ►  
     So for realsies, cortexmerch.com.