WEBVTT - From the Vault: Flatus Ex Machina, Part 1 0:00:05.720 --> 0:00:07.720 Hey, welcome to Stuff to Blow Your Mind. My name 0:00:07.760 --> 0:00:10.760 is Robert Lamb and I'm Joe McCormick, and it's Saturday. 0:00:10.800 --> 0:00:12.640 Time to go into the vault. This time we are 0:00:13.119 --> 0:00:15.040 uh oh. This is going to be part one of 0:00:15.160 --> 0:00:17.640 the pair of episodes we did last year called flate 0:00:17.760 --> 0:00:23.040 Us x Mockina. This one originally published March nineteen, and 0:00:23.079 --> 0:00:26.320 it's about why it's so funny when machines fail, Yes, 0:00:26.480 --> 0:00:29.920 fart out of the machine. What is that? Uh, you 0:00:29.960 --> 0:00:32.440 know you're maybe wondering what that means. It's basically like 0:00:32.800 --> 0:00:36.440 what happens when when AI, despite all of its abilities, 0:00:36.520 --> 0:00:39.600 creates something that is just a flop? And why is 0:00:39.640 --> 0:00:41.640 it so amusing? And what can we learn from that? 0:00:41.720 --> 0:00:44.720 What does it reveal about artificial intelligence in general? Well, 0:00:44.760 --> 0:00:50.320 if you want to find out, we'll listen to this episode. 0:00:50.840 --> 0:00:53.680 Welcome to Stuff to Blow your Mind from how Stop 0:00:53.720 --> 0:01:02.000 Works dot com. Hey you, welcome to Stuff to Blow 0:01:02.000 --> 0:01:04.880 your Mind. My name is Robert Lamb and I'm Joe McCormick, 0:01:04.959 --> 0:01:06.760 and today we're gonna be talking about one of my 0:01:06.840 --> 0:01:11.839 favorite comedy subjects. What's funny about the way that machines fail? 0:01:12.440 --> 0:01:14.000 And here's just heads up. This is gonna be a 0:01:14.040 --> 0:01:16.680 two part episode because we ended up going super long. 0:01:16.800 --> 0:01:19.360 Like the machines, we can't stop. So I want to 0:01:19.400 --> 0:01:23.720 start with a particular example, probably my favorite funny thing 0:01:23.840 --> 0:01:28.840 on the Internet these days, the hilarious almost successes of 0:01:29.080 --> 0:01:35.000 artificial intelligence trying to generate examples of human language almost 0:01:35.040 --> 0:01:38.360 but not quite human. Yes, I don't know why this 0:01:38.480 --> 0:01:41.360 does it for me, but aside from Highlander to the Quickening, 0:01:41.440 --> 0:01:45.759 pretty much nothing makes me laugh harder than language generated 0:01:45.760 --> 0:01:49.120 by artificial neural networks and machine learning. And we'll explain 0:01:49.160 --> 0:01:51.600 a little bit more about exactly how this works in 0:01:51.640 --> 0:01:54.040 a minute, but first I just thought we should look 0:01:54.080 --> 0:01:56.440 at a few examples of what this is like. If 0:01:56.440 --> 0:01:59.480 you're on the Internet, you've probably encountered this at some point, 0:02:00.280 --> 0:02:03.560 because it's become popular in the past few years, especially 0:02:03.640 --> 0:02:06.160 for its comedic value. If you're not on the internet, boy, 0:02:06.160 --> 0:02:08.840 are you in for a treat um by far I 0:02:08.840 --> 0:02:11.240 would say the best of this stuff I've come across 0:02:11.440 --> 0:02:15.000 is traceable back to the blog of a person named 0:02:15.120 --> 0:02:18.080 Janelle Shane, who I think her day job as she 0:02:18.120 --> 0:02:21.800 works in industrial researching optics, but as a hobby she 0:02:21.919 --> 0:02:26.239 trains neural networks with text based data, uh, text based 0:02:26.320 --> 0:02:31.120 data sets to spit out these amazing simulations of types 0:02:31.160 --> 0:02:34.160 of human language, and so they'll they'll be in categories 0:02:34.200 --> 0:02:38.160 like it. She gets an AI program to write recipes 0:02:38.200 --> 0:02:41.120 for foods or to come up with the names of 0:02:41.320 --> 0:02:43.959 paint colors or something. And the way of course you 0:02:43.960 --> 0:02:45.920 would do this is and we'll explain more of the 0:02:45.919 --> 0:02:48.680 details in a bit, but you'd train it on existing 0:02:48.840 --> 0:02:51.880 names of paint colors, or you train it on existing 0:02:51.919 --> 0:02:54.280 recipes of food. Right. So the end result here is 0:02:54.280 --> 0:02:57.919 you essentially have a machine trying to human but not 0:02:58.080 --> 0:03:01.720 quite pulling it off, yes, but but doing so in 0:03:01.880 --> 0:03:05.360 such hilarious fashion. Yeah, so you should absolutely look up 0:03:05.440 --> 0:03:08.880 Janell Shane's blog. It's called AI weirdness dot com. It's 0:03:08.919 --> 0:03:11.400 reliably such a source of joy. But I want to 0:03:11.440 --> 0:03:13.359 start in fact, I will say that if you were 0:03:13.480 --> 0:03:15.400 if you're scratching your head and you're thinking, oh, didn't 0:03:15.440 --> 0:03:19.400 I see something hilarious uh in this vein recently, there's 0:03:19.520 --> 0:03:22.280 there's a there's a high probability that it originated from 0:03:22.400 --> 0:03:26.120 AI weirdness dot com. Yes, that that blog is just awesome, 0:03:26.360 --> 0:03:28.000 But I wanted to look at a few examples of 0:03:28.040 --> 0:03:30.280 what this is like. So one thing is consider the work, 0:03:30.560 --> 0:03:33.560 uh that that Jenell Shane did training in neural network 0:03:33.639 --> 0:03:36.080 to come up with names for D and D spells. 0:03:36.840 --> 0:03:39.400 So you take the Dungeons and Dragons manual and you'd 0:03:39.440 --> 0:03:42.240 feed in all the actual names of spells to the 0:03:42.240 --> 0:03:44.280 sceneral network, so it gets a sense of what these 0:03:44.280 --> 0:03:46.640 things are like, and then it tries to come up 0:03:46.680 --> 0:03:49.360 with similar types of names on its own. Now, to 0:03:49.360 --> 0:03:51.600 give everybody just a quick idea first of what actual 0:03:51.720 --> 0:03:56.160 dungeons and dragon spells are named. You have everything from 0:03:56.320 --> 0:04:01.800 Magic Missile or Crown of Madness to Evard's Black Tentacles 0:04:01.960 --> 0:04:04.200 or anytime there's a wizard name in there, you know 0:04:04.240 --> 0:04:07.160 you're in for some good stuff. Um, well, there's you know, 0:04:07.160 --> 0:04:10.680 stuff like glyph of Warding or what's the what's the 0:04:10.720 --> 0:04:13.320 one I'm trying to think of. Oh, stuff like Leo 0:04:13.360 --> 0:04:19.000 Moon's Secret Chest or Leoman's Tiny Hut. That's another great one. Well, 0:04:19.040 --> 0:04:21.120 so here are the ones that it came up with. 0:04:21.440 --> 0:04:25.760 How about selections like Mr of Light, Confusing, Storm of 0:04:25.800 --> 0:04:32.400 the Giffling, Song of Goom, Song of the Darn, Ward 0:04:32.520 --> 0:04:36.760 of Snay to the pooda beast, primal rear. You've got 0:04:36.839 --> 0:04:39.720 to watch out for primal rear. Someone storm bear, Now, 0:04:39.720 --> 0:04:42.760 that one sounds legitimate. That's one of the beauties of 0:04:42.800 --> 0:04:47.080 these exercises is is when there's one that either almost 0:04:47.160 --> 0:04:50.120 works or actually does work. Because I think some in 0:04:50.279 --> 0:04:52.840 storm Bear, I can I can easily imagine. I can 0:04:52.880 --> 0:04:57.480 describe the storm bear blasting out of the portal and 0:04:57.760 --> 0:05:00.360 rushing into combat on behalf of your you know, your 0:05:00.440 --> 0:05:04.240 your storm mage. Yeah, it's almost kind of effortlessly evocative. 0:05:05.080 --> 0:05:07.720 A divine boom. How about that? That one sounds pretty 0:05:07.760 --> 0:05:11.119 good too, Soul of the bill. Now, now we're flying 0:05:11.160 --> 0:05:14.640 back down the hill. Farc mate, how about a charm 0:05:14.720 --> 0:05:20.839 of the CODs, death of the Sun. Okay that that's 0:05:21.080 --> 0:05:23.280 that would have to be a high level spell. But okay, okay, 0:05:23.360 --> 0:05:27.200 three more greater flick. Okay, that's that's just a can 0:05:27.240 --> 0:05:30.200 trip here, that's just a flick, A magical flick of 0:05:30.240 --> 0:05:34.599 the ear curse clam. I like it. Dav Ing fire 0:05:35.640 --> 0:05:38.800 is confusing. Now, while these phrases I think are mostly 0:05:39.000 --> 0:05:42.080 funny on their own, I think they're probably even funnier 0:05:42.120 --> 0:05:44.479 if you're an actual D and D player, because you 0:05:44.560 --> 0:05:47.600 not only get the pleasure of the nonsense words and 0:05:47.640 --> 0:05:50.839 the you know, the syllables that seem out of place 0:05:50.880 --> 0:05:54.200 in their context, like Dave does not really seem to 0:05:54.200 --> 0:05:56.679 go very well with some kind of magical fire spell. 0:05:57.440 --> 0:05:59.440 But if you actually play D and D you probably 0:05:59.480 --> 0:06:02.880 also get some humor from just like seeing the little 0:06:02.920 --> 0:06:06.560 resonances that these spells have with actual spells that you 0:06:06.600 --> 0:06:10.320 would recognize. I now want to run a gaming session 0:06:10.360 --> 0:06:13.800 where there's some sort of warp effect in place where 0:06:13.880 --> 0:06:18.600 suddenly all the magic users are forced to use spells 0:06:18.600 --> 0:06:21.120 from this list and they don't know what they're going 0:06:21.160 --> 0:06:24.400 to exactly do until they cast them. Even better, what 0:06:24.480 --> 0:06:27.120 if they had to What if your character suddenly had 0:06:27.160 --> 0:06:30.279 amnesia and then had to act as if they had 0:06:30.400 --> 0:06:33.680 bios that were also generated by an artificial neural network. 0:06:33.800 --> 0:06:35.919 That's right, because AI weirdness dot Com also has a 0:06:35.960 --> 0:06:39.679 wonderful piece titled D and D Character Bios. Now making 0:06:39.720 --> 0:06:42.479 slightly more sense. In fact, I would say this post 0:06:42.640 --> 0:06:44.720 from I think this was last week or something we 0:06:44.720 --> 0:06:46.840 were looking at. I think this inspired me to want 0:06:46.839 --> 0:06:48.960 to do this episode. Well, we should read a couple 0:06:49.000 --> 0:06:53.520 of these. Okay, I'll read the first one here quote. 0:06:54.000 --> 0:06:58.479 Frick found his old family's fortune and his curiosity, and 0:06:58.560 --> 0:07:00.640 he went to a small city to see if he 0:07:00.680 --> 0:07:04.440 could find a work in the goldfish. He heard stories 0:07:04.520 --> 0:07:07.719 of a goldfish, a goldfish, a sea monster, and a 0:07:07.760 --> 0:07:10.520 silver fish, a sea monster and a ship that was 0:07:10.560 --> 0:07:14.440 a ship of exploration. The ship was full of fish 0:07:14.480 --> 0:07:18.520 and evil some treasure, but it was not to be 0:07:19.120 --> 0:07:22.320 when Frick found the ship. He rushed back and found 0:07:22.360 --> 0:07:26.080 the ship full of treasure and full of fish. He 0:07:26.160 --> 0:07:29.760 wanted to be a pirate and fight it. Now, I 0:07:29.840 --> 0:07:32.160 like that because there's a there's a lot of silliness 0:07:32.360 --> 0:07:35.560 in there, but it does have the basic shape of 0:07:35.560 --> 0:07:38.000 a bio and and and and and it's weirdness, and 0:07:38.040 --> 0:07:42.800 it's the stuff that is more nonsensical actually feels suitably 0:07:43.040 --> 0:07:46.800 magical and fantastic. You know that there's this this fish 0:07:46.840 --> 0:07:49.000 that's not a fish, that's also a ship that's full 0:07:49.040 --> 0:07:51.600 of treasure and fish, and he's going to become a 0:07:51.600 --> 0:07:54.600 pirate and fight it. And I think you could run 0:07:54.640 --> 0:07:57.080 with that. This is one of the fascinating things about 0:07:57.120 --> 0:08:00.840 neural net generated text is that it often has the 0:08:00.880 --> 0:08:04.840 format of what you're going for, correct it just doesn't 0:08:04.880 --> 0:08:09.440 have the sense of it correct like it will esthetically 0:08:09.600 --> 0:08:12.679 and in shape be very much like what you're looking for, 0:08:13.000 --> 0:08:16.840 but keywords do not make any sense at all. Then again, 0:08:16.880 --> 0:08:19.320 another one that I thought was kind of interesting in 0:08:19.400 --> 0:08:22.400 what it showed about common themes in in D and 0:08:22.480 --> 0:08:25.280 D character bios, or I guess maybe fantasy more generally, 0:08:25.800 --> 0:08:29.240 was one bio that included the lines um the Orc 0:08:29.320 --> 0:08:32.040 Warlock was captured and killed by a group of Orcs. 0:08:32.040 --> 0:08:34.800 He was imprisoned and forced to work for a giant tome. 0:08:35.200 --> 0:08:37.520 He was imprisoned and imprisoned for a while until he 0:08:37.600 --> 0:08:40.320 was rescued by a group of adventurers. He was imprisoned 0:08:40.360 --> 0:08:42.440 and imprisoned for a while until he was rescued by 0:08:42.440 --> 0:08:44.440 a group of adventurers who were looking for a group 0:08:44.480 --> 0:08:47.640 of adventurers to help eradicate the orch is ferocity. See 0:08:47.679 --> 0:08:49.880 now that's wonderful. I actually really like that, And it's 0:08:49.880 --> 0:08:52.920 a little you know, nonsensical. But uh, first of all, 0:08:53.040 --> 0:08:56.280 you know, being an orc is hard. It's it's it's 0:08:56.280 --> 0:09:00.560 a warlike h world for the orc. But but then 0:09:00.840 --> 0:09:03.080 the idea that he's working for a giant tone the 0:09:03.160 --> 0:09:06.320 idea that there's like a magical book employing him. And 0:09:06.400 --> 0:09:10.640 the repetitive imprisonment. Yeah, it's a hard knock life. You're 0:09:10.679 --> 0:09:13.959 always getting imprisoned and then escaping and encountering a band 0:09:14.000 --> 0:09:17.199 of adventurers. Yeah, and working for weird magic books. So 0:09:17.520 --> 0:09:19.640 that one, that one works. I think, of course. Another 0:09:19.679 --> 0:09:21.800 great one on this page is an injury that just 0:09:21.840 --> 0:09:25.920 seems to devolve into endless like dozens of repetitions of 0:09:26.040 --> 0:09:30.360 variations of the phrase big cat. Yes, little did he know. 0:09:30.600 --> 0:09:34.360 During the monastery's course of time, when the monastery's training 0:09:34.440 --> 0:09:37.920 and growth was complete, his mother told big cat, big cat, 0:09:38.000 --> 0:09:40.200 and big cat to drive their own path and test 0:09:40.280 --> 0:09:43.320 big cat with big cats, army of big cats, army 0:09:43.360 --> 0:09:46.800 of big cats and big cats and big cats, big cats, 0:09:46.840 --> 0:09:48.960 big cat, big cat, and it goes on, yes for 0:09:49.040 --> 0:09:53.280 something for for many many lines. I'm trying to picture 0:09:53.360 --> 0:09:56.720 big cat, though I can't. I don't actually go to 0:09:56.760 --> 0:09:59.560 like tiger or lie in there. I think more of 0:09:59.640 --> 0:10:03.400 a a kind of mental power, great basilisk type of 0:10:03.559 --> 0:10:06.200 fat house cat. I think that would work. I also 0:10:06.240 --> 0:10:09.240 can't help but think of cheshire cats, and of course 0:10:09.280 --> 0:10:12.200 cat Bus from Totoro is being sort of in the 0:10:12.280 --> 0:10:16.839 vein of big Cat, Big Cat, big Cat. Um. How 0:10:16.880 --> 0:10:19.439 about a neural networks trained on a corpus of more 0:10:19.480 --> 0:10:23.920 than forty three thousand questions and answer style jokes. I'll 0:10:23.960 --> 0:10:26.440 just read one of them. Why do you call a 0:10:26.480 --> 0:10:33.840 pastor across the road? He take the chicken? Yeah. Another 0:10:34.160 --> 0:10:37.880 great one from AI weirdness dot Com was a neural 0:10:37.920 --> 0:10:41.079 network designs Halloween costumes. Okay, so you just feeded a 0:10:41.080 --> 0:10:43.240 bunch of Halloween costume names. Yeah. And this one, this 0:10:43.280 --> 0:10:45.560 one was tremendous fun. I remember when it first came. 0:10:45.600 --> 0:10:49.080 I just made me laugh so hard. Highlights include uh, 0:10:49.160 --> 0:10:55.520 Saxy Pumpkins, Um, Disco Monster, Spartan Gandalf. I really like 0:10:55.600 --> 0:10:59.480 that one. Starfleet Shark, Yeah, that's good. Um, let's see 0:10:59.559 --> 0:11:03.960 Martian Devil. That was too believable. Panda clam though, that's 0:11:04.000 --> 0:11:09.200 that's pretty good. Shark cow. That's very interesting given some 0:11:09.280 --> 0:11:12.160 of our past discussions on was it a sharp cow 0:11:12.240 --> 0:11:13.800 that we were discussing, and I believe it was. Oh, 0:11:13.920 --> 0:11:17.520 that was That's Daniel Dennett's thought experiment, exposing what he 0:11:17.559 --> 0:11:22.319 believes to be flaws in Donald David's Donald Davidson's Swampman 0:11:22.800 --> 0:11:25.760 thought experiment it was the cow shark, where he says, 0:11:26.360 --> 0:11:29.120 is it actually meaningful if you posit a cow shark 0:11:29.200 --> 0:11:31.480 to ask whether it's a cow or a shark? A 0:11:31.520 --> 0:11:36.360 little accidental thought experiment here from this list. Uh, you'll 0:11:36.400 --> 0:11:39.920 find less thought experimentation in such entries though, as snape 0:11:39.960 --> 0:11:50.760 scarce snape scarecrow, or or how about lady garbage, Lady garbage. 0:11:50.800 --> 0:11:53.040 Lady garbage is good. So yeah, there are a lot 0:11:53.040 --> 0:11:57.760 of great, great entries on that list. Okay, one last 0:11:57.800 --> 0:12:00.160 thing from General Shane's work, I just must mean in 0:12:00.320 --> 0:12:03.319 some some titles for recipes that she wrote a script 0:12:03.400 --> 0:12:07.640 to come up with. Um. These include chocolate pickle sauce, 0:12:07.800 --> 0:12:16.480 whole chicken cookies, salmon beef style chicken bottom, artichoke, gelatin dogs, 0:12:16.520 --> 0:12:21.760 and crock pot cold water. These are where any of 0:12:21.800 --> 0:12:24.920 these attempted these recipes. Uh well, actually at one point 0:12:24.960 --> 0:12:27.600 she so those were just names at some point. They 0:12:27.640 --> 0:12:30.760 don't have these selected, but she did have it based 0:12:30.800 --> 0:12:34.720 on a corpus of recipes generate new recipes which are 0:12:34.760 --> 0:12:38.680 just nightmares, you know, like huge lists of ingredients that 0:12:38.679 --> 0:12:42.160 would be like, you know, a third cup smore goals, 0:12:42.400 --> 0:12:47.280 you know, a cup of Horseradishuh. Then there was one 0:12:47.320 --> 0:12:49.440 website I said, I don't remember who actually did it 0:12:49.520 --> 0:12:52.120 might have been Super Deluxe. There's somebody made a video 0:12:52.559 --> 0:12:55.480 where they took one of these AI generated recipes and 0:12:55.559 --> 0:12:58.679 just literally made it. Now, they had some trouble because 0:12:58.720 --> 0:13:01.160 some of the ingredients were not real words, they're just 0:13:01.480 --> 0:13:04.080 you know, yeah, like, so they had to I guess 0:13:04.080 --> 0:13:07.160 substitute something or they put in instead of smart goals 0:13:07.200 --> 0:13:09.520 it was coco powder or something. But they ended up 0:13:09.559 --> 0:13:13.600 with these like, uh, pasta shells. I think that had 0:13:13.679 --> 0:13:16.520 like chocolate and stuff all over them. But anyway, so 0:13:16.960 --> 0:13:19.400 that I think that's currently one of the best websites 0:13:19.440 --> 0:13:21.760 on the internet. Go to AI weirdness dot com if 0:13:21.800 --> 0:13:25.120 you want more joy and humor. But um, but I 0:13:25.160 --> 0:13:28.960 was wondering, why is this the funniest thing out there 0:13:29.040 --> 0:13:32.320 for me right now? What is so inherently funny about 0:13:32.360 --> 0:13:37.679 the ways machines create things that are sort of like 0:13:37.920 --> 0:13:41.160 real language, just close enough to be in the zone 0:13:41.200 --> 0:13:43.320 where they are funny, but at the same time far 0:13:43.440 --> 0:13:46.320 enough off that they're they're totally hilarious. And I think 0:13:46.360 --> 0:13:49.880 that there are at least two parts that make this 0:13:49.880 --> 0:13:53.120 this stuff so golden one is that there's something inherently 0:13:53.160 --> 0:13:58.040 funny about machines trying to behave like humans and failing. Specifically, 0:13:58.040 --> 0:14:01.280 it's the ways that they fail, and will definitely explore 0:14:01.320 --> 0:14:04.679 this more throughout the episode. But the way that they 0:14:04.679 --> 0:14:10.280 failed demonstrates a kind of pristine, oblivious quality of stupidity. 0:14:10.320 --> 0:14:13.960 It's like a kind of platonic stupidity isolated from the 0:14:14.000 --> 0:14:18.520 ability to appreciate itself. It's like the funniness of you know, 0:14:18.559 --> 0:14:22.360 watching automatic doors repeatedly trying to close on something that's 0:14:22.400 --> 0:14:25.160 blocking them, Like that that's not itself so funny, but 0:14:25.200 --> 0:14:28.280 you see an inkling of the same thing there that 0:14:28.360 --> 0:14:32.000 comes through in these AI generated texts. But then we 0:14:32.080 --> 0:14:35.080 sort of mentioned this earlier. It's also funny because it 0:14:35.120 --> 0:14:39.040 tends to reveal something interesting about the human culture game 0:14:39.160 --> 0:14:41.760 that it's trying to play like. It brings this sort 0:14:41.760 --> 0:14:45.880 of cold objectivity to phenomena that we don't necessarily always bring, 0:14:46.280 --> 0:14:50.360 and it can identify and awkwardly replicate trends and behaviors 0:14:50.360 --> 0:14:52.320 that we might fail to notice in the same way 0:14:52.320 --> 0:14:55.680 that like you might notice funny things that are actually 0:14:55.720 --> 0:14:59.000 present in the names of D and D spells by 0:14:59.040 --> 0:15:02.760 watching a computer or try and replicate them. Yeah, yeah, 0:15:02.800 --> 0:15:05.400 I think these are these are these are two strong reasons. Okay, 0:15:05.400 --> 0:15:07.280 we're going to take a quick break. We'll be right 0:15:07.320 --> 0:15:13.280 back with more. Alright, we're back now with all things humor, 0:15:13.320 --> 0:15:14.880 And of course we'll get into this in this episode 0:15:14.880 --> 0:15:17.840 as we continue to discuss what humor is and then 0:15:17.920 --> 0:15:22.600 why machines in some cases achieve it, like the absurd 0:15:23.480 --> 0:15:29.200 is is funny, Like the absurdity is hilarious, And it 0:15:29.280 --> 0:15:31.440 seems it seems like we're in a situation where a 0:15:31.440 --> 0:15:34.400 lot of times, some of these, especially these neural net situations, 0:15:34.640 --> 0:15:40.280 where we were accidentally creating absurdity engines, we're creating machines 0:15:40.480 --> 0:15:46.200 that that produce absurdity. And uh, you know, well, you know, 0:15:46.200 --> 0:15:48.480 what can you say, like why is why is something 0:15:48.480 --> 0:15:52.280 that is absurd funny? You know, because we'll get into 0:15:52.280 --> 0:15:54.040 all that in a bit. But but sometimes I feel 0:15:54.040 --> 0:15:57.920 like the answer might be it's funny because it's funny. Well, yeah, 0:15:57.920 --> 0:16:01.520 I mean, there's definitely an ineffable quality of of of humor. 0:16:01.640 --> 0:16:04.320 That is one of the reasons there's so many different 0:16:04.360 --> 0:16:06.960 theories of humor. Then, as I said, we'll explore them 0:16:06.960 --> 0:16:09.880 more as the episode goes on. But um, yeah, it's 0:16:09.920 --> 0:16:13.040 obviously something that's really hard to to narrow down and 0:16:13.040 --> 0:16:15.080 put your finger on. It's that there seemed to be 0:16:15.120 --> 0:16:18.880 all these strange, conflicting, overlapping reasons we find things funny. 0:16:19.560 --> 0:16:24.000 But I do feel like this strain of machine humor 0:16:24.080 --> 0:16:27.400 machine failure humor being one of the funniest types of 0:16:27.480 --> 0:16:32.480 humor UH is bigger than just the AI text. Because 0:16:32.520 --> 0:16:35.360 it got me I started thinking about what are some 0:16:35.440 --> 0:16:38.280 of the funniest scenes in movies I can think of? 0:16:38.680 --> 0:16:40.520 And when I tried to think about that, I can't 0:16:40.520 --> 0:16:43.640 help but think of the dark humor of the hyper 0:16:43.760 --> 0:16:48.840 violent boardroom scene in RoboCop with ED two o nine. 0:16:49.080 --> 0:16:51.000 I would argue one of the I mean, it's not 0:16:51.080 --> 0:16:54.680 for children. This is like a hyper violent, horrible UH scene, 0:16:54.680 --> 0:16:57.080 but it's also in in a in a morbid way, 0:16:57.200 --> 0:16:59.640 one of the funniest scenes I think in film history, 0:17:00.640 --> 0:17:02.640 like and so why is it so funny? I think 0:17:02.640 --> 0:17:06.520 it hinges on parallels between humans and machines and the 0:17:06.520 --> 0:17:09.560 scene and the similarities in the ways they fail. So 0:17:10.080 --> 0:17:13.159 brief recap of the scene is, uh. You know Ronnie 0:17:13.200 --> 0:17:17.520 Cox plays this, you know self serious, bloviating businessman who's 0:17:17.520 --> 0:17:20.359 proudly proclaiming, you know, I've got the technology that's the 0:17:20.400 --> 0:17:23.880 future of policing. And he brings out this robot called 0:17:23.960 --> 0:17:26.040 ED two oh nine. He's got these big guns on it, 0:17:26.320 --> 0:17:28.119 and they're saying that it's going to take over the 0:17:28.160 --> 0:17:31.360 police force and it's this great new technology. And then 0:17:31.400 --> 0:17:35.120 they demonstrate it on a guy and it malfunctions and uh, 0:17:35.400 --> 0:17:37.920 it tells him to drop his weapon in the demonstration, 0:17:38.000 --> 0:17:40.040 and he does, and it doesn't seem to notice he 0:17:40.080 --> 0:17:42.680 has and then it shoots him like a hundred times, 0:17:42.720 --> 0:17:45.760 just a ridiculous amount of times. And but it's something 0:17:45.840 --> 0:17:49.919 about the way that the people in the room fail 0:17:50.040 --> 0:17:51.919 in the same way the machine does. Like he just 0:17:52.000 --> 0:17:56.119 plows through this this horrible violent encounter, and then afterwards 0:17:56.119 --> 0:17:58.280 somebody in the background is like, can we get a 0:17:58.280 --> 0:18:02.160 paramedic after this guy has been shot like a hundred times? 0:18:02.720 --> 0:18:05.000 As if like the machine, they're just sort of like 0:18:05.119 --> 0:18:09.680 carrying on their like route behaviors without understanding what they're 0:18:09.680 --> 0:18:13.239 doing or thinking about them. Yeah, the end to oh 0:18:13.320 --> 0:18:17.080 nine is such a great design in the original RoboCop 0:18:17.359 --> 0:18:22.120 because it it resembles it has animal qualities to it. 0:18:22.119 --> 0:18:26.480 It looks kind of like a bipedal dinosaur. Uh, And 0:18:26.840 --> 0:18:30.200 yet it's also smooth and abstract in so many ways 0:18:30.240 --> 0:18:33.480 that it looks like either a highly designed piece of technology, 0:18:33.480 --> 0:18:35.000 which of course it is. It has you know, it 0:18:35.000 --> 0:18:38.640 looks like it's a nice piece of stereo equipment. But 0:18:38.640 --> 0:18:42.240 but then it also lacks any additional details, like it's 0:18:42.240 --> 0:18:46.320 almost a silhouette of a lie of an animal. Yes, um, 0:18:46.960 --> 0:18:49.600 and the growls and it growls as well. But then 0:18:49.680 --> 0:18:52.000 also later a really funny scene in the movie is 0:18:52.000 --> 0:18:56.560 the discovery that this horrible violent killer robot is can 0:18:56.560 --> 0:18:59.439 be defeated by stairs. Like you can't use stairs. It 0:18:59.480 --> 0:19:03.200 has these one for like all terrain looking um legs 0:19:03.200 --> 0:19:06.480 that it walks on, and that it can't manipulate stairs 0:19:06.520 --> 0:19:09.080 it falls down them and then it's it's like a 0:19:09.600 --> 0:19:13.120 ridiculous upside down duckling. I mean it's so hilarious too, 0:19:13.119 --> 0:19:15.520 because I mean ed to A nine is highly effective 0:19:15.600 --> 0:19:19.840 in other situations if it's just filling a guy with bullets, 0:19:20.280 --> 0:19:25.480 highly effective, um, just battling RoboCop also highly effective for 0:19:25.680 --> 0:19:29.000 this part um and this falls in line I think 0:19:29.000 --> 0:19:32.640 pretty well with our experiences of machines and of AI. 0:19:32.720 --> 0:19:35.879 We can create highly effective specialist in many areas of 0:19:35.920 --> 0:19:39.320 AI and robotics. So you create a machine that just 0:19:39.359 --> 0:19:41.879 puts bullets into this guy. Uh, you know, it does 0:19:41.880 --> 0:19:44.600 a great job. But when it comes to creating a 0:19:44.640 --> 0:19:48.560 general AI or a machine that can navigate the complex 0:19:48.720 --> 0:19:51.679 natural or even or the human created world such as 0:19:51.760 --> 0:19:56.000 the stairs, Uh, there's continual challenge there. Like that's kind 0:19:56.000 --> 0:19:58.160 of That's that's what a lot of of what's going 0:19:58.160 --> 0:20:00.800 on in robotics and AI is about. Yeah, or the 0:20:01.000 --> 0:20:03.919 just recognizing that it's not actually supposed to shoot the 0:20:03.960 --> 0:20:07.720 board executive during the demonstration exactly. Yeah, but then also 0:20:07.800 --> 0:20:11.600 just the idea that it's it's wonderful, it's something and 0:20:11.680 --> 0:20:15.080 terrible at another thing, that imbalance is often where we 0:20:15.119 --> 0:20:17.840 find a lot a lot of hilarity in other um, 0:20:18.280 --> 0:20:21.679 you know, another comedic stories and bits of fiction or 0:20:21.680 --> 0:20:25.439 just situations like one of my favorite cut scenes of 0:20:25.440 --> 0:20:28.800 all times from Conan the Barbarian. So there's a scene 0:20:28.800 --> 0:20:31.040 in it where it coming this is the Arnold Schwarzenegger original. 0:20:31.080 --> 0:20:33.640 There's it's like a blooper outtake. It's a blueper out take. 0:20:33.680 --> 0:20:35.560 You find it, I don't think most of the special DVDs. 0:20:35.560 --> 0:20:39.560 It's available on YouTube as well. But basically Conan has 0:20:39.640 --> 0:20:44.280 just escaped or been released from servitude and he's running 0:20:44.320 --> 0:20:48.520 across the you know, the wasteland. Essentially, wild dogs are 0:20:48.600 --> 0:20:51.399 chasing him in order to to to eat his flesh. 0:20:51.560 --> 0:20:54.119 He manages in the finished film to scramble up some 0:20:54.240 --> 0:20:57.280 rocks and that begins this new story arc him discovering 0:20:57.320 --> 0:21:00.720 this great old sword. But this is like young corporal 0:21:00.760 --> 0:21:03.800 beef Body, giant Arnold Schwartzen. This is the this is 0:21:03.840 --> 0:21:06.680 the Arnold Schwartzenegger that was so ripped he had to 0:21:06.720 --> 0:21:09.679 like lose muscle so that he could actually hold the 0:21:09.720 --> 0:21:13.520 sword correctly. Uh So there's this out take though, where 0:21:13.520 --> 0:21:17.480 he's running in chains. The dogs are chasing him, the 0:21:17.520 --> 0:21:20.720 movie dogs, and then he's scrambling up the stones and 0:21:20.760 --> 0:21:24.320 the dogs catch him and drag him back down and 0:21:24.359 --> 0:21:27.800 he's just screamed on and and cursing the whole time. Yeah, 0:21:27.800 --> 0:21:31.359 it's I watched it after you linked. It very very funny, 0:21:31.400 --> 0:21:32.720 and you see the p a s come in to 0:21:32.760 --> 0:21:36.760 like get the dogs off the dogs and he's like, yeah, 0:21:37.160 --> 0:21:40.120 it's it's funny because the idea of Conan the barbarian 0:21:40.280 --> 0:21:43.879 or even just prime Arnold failing like this, it's just 0:21:44.520 --> 0:21:48.359 start contrast to actual or perceived strength of the character 0:21:48.640 --> 0:21:52.080 and or individual. And it's also funny because he wasn't 0:21:52.119 --> 0:21:55.200 actually mauled to death by movie dogs, right, of course, 0:21:55.200 --> 0:21:58.280 he wasn't actually seriously harmed in this incident, but he 0:21:58.560 --> 0:22:02.359 was apparently convenienced and had his pride wounded. Right. And 0:22:02.400 --> 0:22:04.760 it will come back to that idea, to the degree 0:22:04.920 --> 0:22:09.359 to which you know, the severity of the outcome um 0:22:09.720 --> 0:22:12.680 comes into play and determining if something is funny or not. Yeah, 0:22:12.720 --> 0:22:15.080 I think I can definitely see what you're talking about 0:22:15.119 --> 0:22:19.120 that the failures of technology are especially funny when there 0:22:19.119 --> 0:22:23.119 are other ways that the technology is highly advanced or 0:22:23.200 --> 0:22:26.480 presented as highly advanced. And as long as like nobody 0:22:26.520 --> 0:22:29.919 of course dies, you know, um. But but I do 0:22:30.000 --> 0:22:31.800 wonder too if there's a darker streak and all of 0:22:31.800 --> 0:22:35.320 this too, you know, something that ties into a deeply 0:22:35.520 --> 0:22:39.639 rooted human disdain for the other, especially for others of species. 0:22:40.400 --> 0:22:42.720 There is a quote from C. S. Lewis's The Lion, 0:22:42.880 --> 0:22:45.119 The Which the Wardrobe and the Four Children. That's my 0:22:45.200 --> 0:22:48.040 son suggested alternate title for it. He's like, like, why 0:22:48.040 --> 0:22:50.800 don't they call it the land which Wardrobe and the 0:22:50.840 --> 0:22:53.320