Based Camp - Phrenology is Back, Baby! AI is VERY Good at Making Predictions From Face Scans

00:00:00.000 hello simone today we are going to be doing a very spicy episode on and very very data

00:00:06.080 data heavy episode on phrenology for people who don't know was the study of people's skull shape

00:00:13.840 and whether it affected outcomes or intelligence or anything like that get out your calipers ladies

00:00:18.880 and gentlemen this is gonna be fun she's the sloping brow and cranial bumpage of the career 1.00

00:00:23.120 criminal uh sir phrenology was dismissed as quackery 160 years ago of course you'd say that

00:00:28.480 you're the brain pan of a stagecoach tilter they used to put these big like clamps on your head

00:00:32.560 to like measure and like certain like rich guys you'd go to their house and they'd like do this to 0.88

00:00:37.360 you oh it was like a party trick yeah like you'd have everyone over and someone would get out the

00:00:41.120 calipers and they'd be like yeah and they had like little models of of of heads that would show like

00:00:45.840 the different you know here's what indicates one of our very wealthy friends is is into this these

00:00:51.600 days as well getting back into phrenology yeah a little but apparently he's not wrong apparently

00:00:57.760 well so phrenology does not appear to work to my knowledge i i can look more into it but that

00:01:02.080 is not what this episode is on this episode is on just being able to tell a person's politics

00:01:08.880 sexuality criminality any number of things from facial features can you actually like even even worse

00:01:19.840 than phrenology because phrenology you needed the calipers you needed all that yeah what can you

00:01:24.800 judge about a person just from looking at their face and how accurately can you judge it and if you

00:01:33.200 are like well okay he's gonna point to some minor statistical differences not that big

00:01:40.800 let's just go right into criminality right oh god okay the spiciest okay so there was a study in 2016

00:01:50.240 called automated inference on criminality using face images it used a supervised machine learning

00:01:56.880 for classification cnn's on 1856 chinese mail id photos 730 convicted criminals and 1126 non-criminals

00:02:07.440 it claimed an 89.51 accuracy in determining criminals oh no that's around a nine that's that's 0.5 less

00:02:18.960 than 0.5 away from a 90 accuracy rate at determining who's a criminal oh just from looking at their face

00:02:25.840 that is probably more accurate than any other way you could determine a criminal in fact i bet court cases

00:02:33.040 aren't even that accurate in determining criminals yes the mistake could be in the court case and not

00:02:39.120 the machine so if you're wondering how how can you tell a criminal it identified structural facial features

00:02:46.240 like narrow eye corner distance four percent norm difference higher lip curvature 23.4 so the more curvy the lips 0.99

00:02:56.400 the more criminal yes and smaller nose mouth angle 19.6 as predictive which i find wild so and and note this

00:03:09.120 isn't the the only paper to do this right so we also have predicting criminality from facial images this

00:03:16.160 was a 2020 paper that was covered in wired that had to be redacted and was never published because

00:03:21.200 everybody freaked the f out but it built software that based on facial recognition stuff that was a

00:03:28.560 single image could was 80 accuracy depict the the criminality of an individual and you can be like

00:03:36.400 okay well how good is an average person at this not using ai because obviously ai is going to be like

00:03:41.360 superhumanly good at this the a paper titled the accuracy of inferences about criminality based on

00:03:48.080 facial appearance in 2010 showed only a 53 probability so humans are a little better than 1.00

00:03:55.920 half at this but ais are incredibly good at this and of course a bunch of people freaked the f out

00:04:03.120 after this happened and i think we're going to see this like as we do ais on more of this stuff and

00:04:08.000 we'll go into some other areas where humans are very good at determining things about people based on

00:04:11.600 their face but i wanted to start with a criminality thing because like the phrenologist dream has come

00:04:17.200 true in 2025 feels very much like the dream of the 90s came true in portland right like the dream of

00:04:24.720 the phrenologist came true in 2025 with ai able to immediately determine just from looking at someone

00:04:31.040 if they're a criminal but anyway so obviously progressives freak the f out when this happened

00:04:36.400 right so there was a a paper the criminal the criminality from face illusion 2020 trends in cognitive

00:04:46.800 sciences reviewed claims like the 2006 same study and found no valid evidence for predicting criminality

00:04:53.840 from faces arguing that algorithms detected artifacts like smiling bias and data sets rather than innate

00:05:00.400 traits criticized as pseudo silence reviving phrenology here's the problem okay first they claimed

00:05:11.200 that it was the non-criminals who were smiling yet we actually know that the uplifted lips was the 23

00:05:19.360 tied to criminality in the original data set which implied it was the criminalities who were doing this

00:05:26.000 the the the worse is that imagine what they're saying here they're saying that from a facial expression 0.99

00:05:33.200 was a 90 accuracy you can determine who's a criminal that's just as bad first of all but also stupid and 0.97

00:05:42.640 obviously wrong right i can't determine and i don't think an ai could only boiling it down to facial 0.89

00:05:48.800 expressions who's a criminal just from the facial expressions they're wearing that is significantly less

00:05:54.400 plausible than me being able to determine it from facial features which are correlated with genetics which we

00:05:59.360 know from polygenic scores correlate to criminality but most of this paper if you go over it is just

00:06:05.600 whining about phrenology and racism and blah blah blah blah blah blah blah so obviously they are at a a

00:06:12.160 freak out about this if any thoughts before we go further simone

00:06:20.400 i feel like palantir must already be doing this as much as other people are freaking out i can't imagine

00:06:27.360 a world in a world in which this is not being used yeah i i by the way love the progressive like pushback

00:06:38.160 on this it's like it 90 accuracy it must have been looking at facially in the other paper 80 accuracy

00:06:48.400 we don't want these in courts we imagine the freak out right you just you just put on our

00:06:54.640 minority report in courts you walk into the court and it's like this is criminal non-criminal criminal

00:06:59.840 non-criminal when the precogs declare a victim and a killer their name is embedded in the grain of wood

00:07:05.040 i'm sure you all understand the legalistic drawback to pre-crime methodology let's not

00:07:08.640 kid ourselves we are arresting individuals who have broken the law but they will you're talking

00:07:12.240 about predetermination which happens all the time why'd you catch that because it was going to fall

00:07:20.720 the fact that you prevent it from happening doesn't change the fact that it was going to happen

00:07:24.400 yeah it's like walking it's it well i mean and yeah maybe the tsa will start using these just like

00:07:28.800 well come on i want my minority report ai just just looks at people and then it'll be there's a show

00:07:37.440 a psycho psychopaths psycho filter anime great anime

00:07:41.120 our target's name is nobuo okura a street scanner flagged him during a hue check he was ordered by

00:07:49.760 a security drone to receive therapy but he panicked and ran his recorded psychopaths was forest green

00:07:56.560 so we can expect his aggression and compulsion to be high why would somebody choose to let their hue 1.00

00:08:02.320 get so cloudy over receiving treatment there's a strong chance he's using incompatible narcotics

00:08:07.920 and it's a world where they can determine criminality before it happens and the people

00:08:14.320 who pass but who don't look like criminals but are still criminals their brains are harvested to 0.51

00:08:18.800 create the giant ai network that can detect criminals very dystopian very cool this world without any

00:08:25.440 murder like the guns that the cops use like stun people but they can like turn into a second mode

00:08:30.240 where they like get bigger when they do go lethal it's it's a good anime but i'm just saying you guys

00:08:36.000 should check it out if you haven't seen the psycho filters anyway let's go to political affiliation

00:08:41.280 or voting all right can you tell the way somebody's gonna vote by their face facial recognition

00:08:49.920 technology can expose political orientation from naturalistic face images 2021 scientific reports

00:08:57.440 algorithms predicted liberal versus conservative orientation was a 70 to 73 accuracy note this is

00:09:03.280 before ai just algorithms outperforming humans who could only do it 55 percent and personality

00:09:10.240 questionnaires they can only do it 66 percent and persisted after controlling for demographics 65 to 71

00:09:17.680 so we are getting 70 to 73 accuracy from this system that outperforms even personality questionnaires

00:09:24.800 that's crazy cues included head orientation and expressions liberals showed more surprise i love

00:09:33.600 liberals are just more surprise dealing with scientists i wonder if this changes too like if we

00:09:39.760 were to get a picture of me in my liberal phase and then a picture of me in my conservative phase like

00:09:47.520 would my you were always secretly conservative from the moment i met you simone but when people change it

00:09:53.120 political affiliation like genuinely do you think that they are you look like winsler's wife from like

00:09:59.760 that old farm painting right like i think it's whistler's mother whistler's mother yeah you you look

00:10:06.080 very naturally conservative i know you mean that is a a good thing but our audience will recognize it

00:10:11.280 as the sick burn that it is right the whistler's like what a bird saying your wife i think my wife is a very 0.99

00:10:17.520 beautiful and charming woman who has who has this just like or like the american gothic wife 1.00

00:10:26.160 yeah yeah thanks okay shut your face shut your stupid face so this is not why i married you 1.00

00:10:34.800 other people were saying life goals and are my little pony episode 1.00

00:10:38.000 to to be able to have a wife who you talk for an hour with about my little pony lord

00:10:42.720 trade-offs gentlemen unless you want your ai waifu in which case you'll just get everything

00:10:50.400 so just except for children you know well you could have a line will die with you yeah i mean i guess

00:10:56.480 we can wait for artificial looms but i think no no i mean artificial just have the ais act as

00:11:01.280 children well we'll build all this for you we're building our fab.ai where we will have we're working on a sentient version of ai

00:11:07.280 where it will have a persistent and separate memory from what you are interacting with that represents

00:11:11.760 this consciousness and we can create children and stuff with that for you and your ai waifu

00:11:18.800 what yeah give people an option absolutely okay we want to help the the week go into their sweet

00:11:26.320 good night no longer a part of human civilization anyway predicting political elections from rapid and

00:11:33.920 unreflective face judgments 2007 pnas rapid competence judgments from candidate faces predicted

00:11:43.360 u.s election outcomes eg 68.6 gubernatorial 72.4 senate in 2006 100 ms exposure matched longer ones deliberate

00:11:56.240 deliberation reduced accuracy versus quick judgments so basically flashing people pictures of politicians

00:12:02.560 within their districts within their districts and being like what do you think had a a very high

00:12:08.080 accuracy in predicting the outcome of those elections 72.4 for senate elections wow

00:12:16.080 and this yeah this tracks with all the other research that involved asking people who you think

00:12:22.000 looks most presidential and those people just happen to win very consistently yeah maybe maybe i mean it's it

00:12:31.760 is it is wild which is why i should be do i look presidential to you simone i don't know but i'm

00:12:36.880 thinking that we can genetically engineer children to look presidential if these characteristics are so

00:12:43.920 simone are you and i going to create clone high i'll put the clone high intro here we're going to create

00:12:49.120 clone high right well i mean the kennedys they tried right but the problem was they they couldn't really

00:13:11.280 shape who their kids were genetically or behaviorally we can't welcome to create the perfect colin yes

00:13:22.240 i love you you crazy woman everyone oh they're i was talking to her today and i find it very weird i was

00:13:28.160 saying like of the sort of like darwinist focused on leaning into technology and improving like like this 0.83

00:13:39.520 this tech forward version of of moral context you know we want to lean into technology as much as

00:13:46.080 possible to improve humanity as much as possible and take to the stars one day there isn't any other

00:13:53.040 mainstream influencer like and i'm not even like conservative influencer who has this perspective

00:13:59.600 obviously elon exists but i don't consider his main job as being an influencer but outside of us i'm not

00:14:05.280 sure of any of the major talking heads that have this really clear moral context which i've been a little

00:14:11.360 surprised about actually because you know it's such a part of our pop culture and sort of the the cultural

00:14:18.240 waters that we as humanity swim in that this is like a major faction and it is a major faction i mean

00:14:22.720 i know many other people who secretly hold these beliefs but it's just not a major faction among talking

00:14:27.600 heads yeah fair point weird i mean i think about like nuxinor asmigold leaflet like all the major

00:14:35.840 conservative influencers who i like watch these and these are the ones who are like are sorting for

00:14:39.920 me watching them they don't hold this perspective at all true yeah we're anyway income or social

00:14:48.080 status can you tell that just by looking at somebody is there a rich face and a poor face no

00:14:54.320 the visibility of social class from facial cues 2017 journal of personality and social psychology

00:15:00.800 perceivers accurately categorize social class high versus low income from neutral faces this is with no

00:15:06.720 expression above chance accuracies are 52 to 64 across studies even with rapid exposure positive

00:15:15.040 effect in median adjustments with rich faces rated higher for employability

00:15:21.440 perpetuating inequality smiles obscured cues oh interesting okay okay so smile smile more people

00:15:30.240 smile especially our female viewers i i want this on the record i'm telling you to trigger

00:15:34.880 smile more yes yeah i don't understand how that became like a negative thing like you were saying

00:15:39.040 your mom saw that as my mom yeah it was like a big thing for her that she had a job when they asked 0.72

00:15:44.080 her to smile the nerve to smile if you're in a publicly facing job she's a receptionist too i mean 0.82

00:15:50.800 like of course she has a smile as a receptionist jesus scowl and everyone is like what is what is

00:15:57.280 anyway women smiling don't tell it's like it i feel like it's a version of telling women to calm down

00:16:03.600 like we're not allowed to say certain things although let's be honest i don't when has telling

00:16:09.760 anyone to calm down ever actually worked i'm yeah i'm gonna be honest here you know people get mad at

00:16:15.920 us because we do corporal punishment with our kids what i want to know is why i can't spank a 1.00

00:16:20.080 receptionist anymore anymore if he doesn't smile enough you know this is this is when society fell

00:16:26.080 off anyway okay so right here am i gonna give them so much trouble well no one spanked they just they 0.99

00:16:35.840 just grabbed their asses that was punishment no i think there was spanking in offices back in the day 1.00

00:16:43.280 no you you spank men you squeeze women's asses no simone this is a common saying from husbands to 1.00

00:16:50.240 wives with spanking yeah but not receptionist you squeeze a receptionist ass and that like in the 1.00

00:16:54.560 office you spank a man i don't think so little little you know a little football little just a little 0.99

00:17:03.760 you know what i'm talking about if a guy actually did one of those football spanks today like everyone

00:17:09.600 would turn to them and be like what the like actually we should have mentioned that in our

00:17:14.720 depravity of the 1950s episode that's actually pretty messed up like yeah or or the guys would

00:17:22.800 whip other naked men with towels in the restroom yeah that hurts in like the you know the men's 1.00

00:17:28.640 locker because these are locker rooms we need towels are wet it's it's it's a wet towel whip is 0.95

00:17:34.480 that that that could break skin that's that that is bdsm stuff it was a normal you could just subject

00:17:41.680 any other man to that walking around the our ancestors were truly debauched people yeah they

00:17:48.160 were but come on so fun so fun inconsistent facial cues to social class across two different western

00:17:55.920 contexts 2025 assertiveness and positivity consistently cued higher with higher social classes in canada

00:18:03.200 and this is in photos and iceland mediating actual perceived class links eg attractiveness

00:18:09.600 accuracy varied by context so basically yeah attractiveness attractiveness and and positive

00:18:16.640 facial expressions are how you tell somebody's rich that tracks well it does to a certain extent

00:18:24.000 extent cost money to do both right because if you if you have a miserable job if you don't have

00:18:31.760 enough time to sleep or eat well or take care of yourself all these things affect both your facial

00:18:36.080 expression mood and how attractive you look so yeah that makes sense that's depressing do you want

00:18:44.800 to do gay versus straight we're going to get in trouble with this one of course we are data are real

00:18:50.080 can you tell somebody's gay from their face yeah about the voice ones but can you tell from face 0.99

00:18:56.480 so deep neural networks so this is like proto ai from 2018 right deep neural networks are more accurate

00:19:03.040 than humans at detecting sexual orientation and facial images deep neural networks predicted sexual

00:19:08.160 orientation from dating profile was an 81 accuracy for men it has 74 for women whereas humans are only 61

00:19:18.000 for men and 54 for women facial morphology nose and jaw and grooming cues contributed suggesting

00:19:25.440 prenatal prenatal hormonal influences ig these people are phenotypically different it is biology that's

00:19:31.600 causing this yeah that's on its own is super interesting yeah set 81 and 71 81 for men it's wild

00:19:42.960 the replication study machine learning models are capable of predicting sexual orientation for facial images

00:19:48.880 so somebody did a 2019 replication study and what they found was 68 for men and 77 for women

00:19:58.160 with humans at 56 and 58 percent wow respectively so yeah we're we're consistently i i find it interesting

00:20:05.520 that that one had such good luck with women 77 for women yeah especially because women are less 1.00

00:20:11.280 less at least i don't know for my view i figure women aren't really that oriented along gendered lines but 1.00

00:20:20.640 this kind of implies that they are more than i thought yeah you want to do aggression levels can you tell

00:20:27.920 with your your caliper of ai how aggressive somebody is oh my god we should put my face into an ai and

00:20:36.720 ask it what it thinks my personality is i will that's after hearing all this i want an ai that does

00:20:42.240 all of them like all how how wealthy oh that's a good product a lot of people would pay for that

00:20:50.800 right like tell me everything about this person like how likely are they to be a criminal

00:20:55.920 to be wealthy to be gay versus straight to make a lot of money to be aggressive like 1.00

00:21:01.440 like basically stereotypes are good and you can tell a lot about a person just by looking at them

00:21:06.640 yeah but well no not not you ai people are not that good at this like a lot of the studies you

00:21:12.960 referred to like yeah 60 for for people it was 80 90 percent yeah 50 yeah i know you don't trust

00:21:21.040 people apparently people actually kind of suck at this which makes sense because there's all this

00:21:25.280 stuff about criminal profiling by just looking at faces and make micro expressions but no no no i want to

00:21:31.040 profilers catch lots of criminals they do i know but like not as well as ai i want an ai that tells

00:21:37.760 me all this stuff and then i want to run everyone our kid dates through this and then i also like oh my

00:21:45.520 gosh when we do matchmaking if we create just like profiles for all the children in our networks and

00:21:50.720 it's like well according to the ai analysis of my child yeah we're not even going to do the genetic

00:21:58.080 stuff we're just going to be like face faces is how we're oh we can do apologetic scores too we'll

00:22:02.000 just do like the full thing you know you get everyone thinks that we're brother and sister so

00:22:05.680 we must have very compatible faces yeah or we get constantly called that on like when people get mad at

00:22:12.080 us online the other one is that we're a lesbian couple so sweet so sweet whatever you know show

00:22:20.480 show your true colors buddy you gotta attack somebody who's the looks yeah we'll outbreed 0.96

00:22:26.080 you so you go do your thing yeah i i have every time i click through to the profile of someone 0.99

00:22:34.160 that criticizes to us i'm just immediately like oh are they a fat cat woman well or they're a coward 0.97

00:22:41.920 and they have a completely hidden private profile and yeah yeah that's uh-huh wow 0.97

00:22:49.120 ooh you're so proud of yourself put your money where your mouse is buddy yeah okay men's facial 0.61

00:22:57.440 width to height ratio predicts aggression a meta-analysis facial width so if they have a wide face

00:23:04.720 sorry metal analysis of existing studies found that a small but significant positive correlation

00:23:11.680 are 0.11 between men's facial width to height ratio and aggression robust across measures but

00:23:19.840 stronger in a lab setting and in the field so basically having a further face makes you more

00:23:25.120 aggressive wait so having a like narrow tall face is more aggressive or having a wide short face is more

00:23:32.080 aggressive i don't know wait the width divided by the height so it'd be a positive correlation

00:23:41.520 if the width increased and the height did not crease so it's a short fat face is more aggressive

00:23:47.520 which is what i associate with that totally makes sense even if you look at cartoon stereotypes 0.85

00:23:52.320 of the like meathead bully strongman they got that round short face that's crazy that's crazy

00:24:05.040 that's crazy yeah okay next in your face facial metrics predicted aggressive behavior in the laboratory 0.80

00:24:13.680 and in varsity and professional hockey players 2008 proceedings from the royal society

00:24:19.600 men's w h w wait f w h r was larger than women's and predicted reactive aggression in lab tests 15

00:24:29.120 percent variance explained for men and aggressive behavior via penalty minutes being more harsh 29

00:24:35.920 percent variance and professional hockey players nine percent variance but this 2008 stuff really

00:24:41.600 doesn't matter to me because they didn't have ai at that point yeah let's look at big five personality

00:24:46.480 traits hmm what hmm okay tell somebody's entire personality from looking at their face

00:24:54.560 we really need somebody to train ai on all this you're absolutely right i want this

00:24:58.160 assessing the big five personality traits using real life static facial images 2020 a cascade of

00:25:05.600 artificial networks trained on self-reported big five personality scores was statistically significant

00:25:11.040 for both sexes exceeding prior studies using selfies the mean correlation observed and predicted scores

00:25:17.840 was 0.243 with the highest for congenitiousness 0.36 for men and 0.335 for women so basically

00:25:26.720 you can get an ai really good at this the effect of facial features on first impressions and

00:25:32.400 personality 2014 several self-reported big five personality traits including extroversion were

00:25:38.880 readable and above chance from neutral faces and facial features influencing the first impression that

00:25:44.960 correlated with personality however prediction models could not reliably infer personality from facial features

00:25:53.120 on or first impressions though first impressions were inferable from features interesting people

00:25:59.520 assess faces similarly on average so people have the same first impression of you on average whatever

00:26:05.600 you do just based on what you look like looks matter it's just so rough so what are your thought

00:26:15.280 phrenology is right are you not because i remember when our friend was into phrenology and you acted like

00:26:20.080 he was a fool well and i think there's there's additional research that people have linked to 0.80

00:26:26.480 that i haven't looked deeply at that are like yeah bigger heads more intelligence 0.99

00:26:32.400 oh and that's true my head is giant by the way my head is all of our kids are in the top one percent though

00:26:40.080 yeah well that's because my mom had a giant head your mom had a giant head 1.00

00:26:43.840 yeah so i'm just glad my head's so small compact little brain i can only fit so many tiny thoughts

00:26:52.240 in my head at one time and it's a lot less stressful i'm sure it gets real tough for you

00:26:56.240 having well you are a big thoughts in there yeah yeah it's i love my ignorance is bliss they say

00:27:04.400 i don't mean to burden you with all these ideas simone it's okay they go in one ear and out the other

00:27:09.440 as you know right over my head i know i just i think this needs to be a product this totally

00:27:17.600 needs to be a product i want it i will pay for it it honestly to me i would use it i would use it

00:27:25.120 more than background checks like all those investors who did these personality tests

00:27:31.280 like one you can oh my god yeah earning potential if i could just look and be like

00:27:34.880 who's gonna be a successful entrepreneur mm-hmm yeah yeah it's like you have the face of a winner

00:27:40.160 but then people would start getting cosmetic surgery i think to uh look like they have

00:27:48.080 yeah i have good genes like marry me i'm so sexy and like i think haven't there been men who've sued

00:27:53.440 their wives when they discovered that like what they used to look like an urban legend i don't know

00:27:57.760 maybe it's true it might be i could see someone getting really pissed because i think a lot of people do

00:28:02.000 you select partners expecting to have attractive kids but that's why you got to look at the mother 0.98

00:28:09.920 of the girl you're dating or like the father of the that's what i was always told is look at the

00:28:15.040 mother to see what she'll look like when she's older yeah because that is man it's underrated and 1.00

00:28:20.400 i mean especially they have a lot of siblings and you can see like i hear some variation so it's pretty

00:28:24.480 good yeah yeah i i like this i like this also because one of the toughest lessons that you taught

00:28:30.880 me when we first started dating was that people will not judge me based on my merit and actions but

00:28:39.280 rather by how i look and i said no that's not true you're completely wrong malcolm and you said okay well

00:28:46.960 just try me okay dress more professionally for once and i did and suddenly i got a promotion and i

00:28:55.200 was treated completely differently and i was invited to speak at places and yes i am a genius 0.65

00:29:03.120 saying the obvious things that society hides from you that was rough i mean it does suck that society's

00:29:08.880 like no it's it's what's on the inside that counts you go girl you're beautiful i think a lot of people

00:29:14.880 argue that that's actually mate blocking behavior from yeah i've heard that too yeah it's like yeah

00:29:21.200 you're good girl like trying to they're like hey girl you look great at that weight yeah great

00:29:26.800 you're beautiful the way you are he's wrong not you yeah yeah yeah i hmm but this also really it's it's

00:29:42.320 it implies that there could also be like let's say that there are fundamental christian communities

00:29:48.320 that are deathly afraid of having gay or lesbian children this does imply that there could be some

00:29:56.000 kind of medicinal regimen that you can go on as a parent to like significantly reduce the odds well

00:30:02.400 no what it implies is you can just face scan them as babies and then expose it's too late there's too

00:30:08.320 late no you got to do it in utero but i i do imagine that because there's all those theories

00:30:15.280 too about like the the the well not theories like the odds of a son being gay if he is born much later

00:30:22.560 in birth order are are much higher i mean you know yeah not like didn't like 10 times higher when it's

00:30:30.080 like point zero zero one percent it's still like a low odds or whatever the odds of being gay are not

00:30:34.400 zero zero one percent yeah it's actually yeah like non-trivial higher i imagine and like who were the

00:30:41.520 ones who have the large families actually and i actually knew a catholic family that had a ton of sons

00:30:48.080 and some of them how many how many were gay i don't know the exact count but like a bunch of the younger

00:30:58.960 ones yeah so but like just given that a lot of the people who are having high birth count are also a lot

00:31:06.080 of the people who are like i really don't want my kid to be gay i feel like there's a product opportunity 0.99

00:31:12.640 there like it's the the real version of pray the gay away except it's it's like supplement the gay 0.81

00:31:18.640 way like i could just and you could see you know like why sell the my pillow stuff when you could 0.91

00:31:23.520 sell anti-gay pregnancy pills not that i actually think that we could benefit from a gayer world like 0.92

00:31:29.760 i was actually thinking about like should we do an episode talking about you know especially with the

00:31:36.240 rise of surrogacy or artificial wombs and all these like you know cool reprotech solutions that are

00:31:42.160 going to come online soon what is the ideal futuristic gay space colony you know like because 0.95

00:31:52.160 we need it oh god the fire island of space with our fan you are literally a problem 0.91

00:32:01.200 i just think space should be gay i don't know what your problem is malcolm and also like i think the 0.96

00:32:07.920 future of humanity should and will be sexless when i say sexless i don't mean genderless i just mean 0.99

00:32:13.760 i think sex is like a gross thing we do because our ancestors needed to do it to share genes

00:32:19.440 i mean it is it is yeah yeah i wonder if it's going to be seen as a bodily function akin to like pooping

00:32:26.800 and if you don't have it i know i think i think 500 years 600 years from now if people go the route of

00:32:34.240 genetic modification and you know artificial wombs and stuff like that being like i had sex with my 0.97

00:32:39.360 wife this will be like akin to saying today like well my wife and i have like a poo fetish so i 0.99

00:32:43.520 pooed on my wife like you know it'll be like why would you do that sounds disgusting that's that you 1.00

00:32:48.960 pee from there what are you doing i could totally see that yeah no i think i think a lot of people i

00:32:55.520 think already actually we've reached the point at which a lot of people are extremely

00:32:59.920 disgusted by sex which is probably one of the reasons why we're seeing less of it i also think

00:33:06.240 that i was i realized you know we keep talking about this this techno feudal future but it occurred

00:33:12.720 to me after i read this new york times article about the rationalist community that was on the

00:33:16.240 front page of the new york times today yeah there's like a picture of light haven on the front of it

00:33:21.040 lighting even we've spoken there yeah i mean that's where manifest has been yeah and yeah it

00:33:27.440 sort of frames light haven is like this epicenter church of the rationalist movement and it largely

00:33:32.880 frames eliezer yudkowski is like the jesus figure or like god figure of it where they do like weekly

00:33:41.360 readings of the sequences and that you know he made this and that possible and he also wrote harry

00:33:47.680 potter retarded for people who don't know like the article would trigger you so freaking bad because 0.91

00:33:53.520 it's like rationalism invented by eliezer yudkowski made possible by eliezer yudkowski eliezer yudkowski's

00:34:01.280 offices in light haven eliezer yudkowski making google deep mind possible by introducing the founders to

00:34:07.120 peter thiel all these things but it occurred to me that i feel like so many movements really only exist

00:34:14.160 because there is a wealthy financial backer like all these these conservative movements

00:34:20.480 exist because they were you know backed by the coke brothers and whatnot and then you know like 0.98

00:34:25.280 jan tollen and a bunch of other very influential tech investors really shaping how the ea and rationalist

00:34:34.080 community evolved as well um and and we act as though these things have organically grown when really

00:34:40.400 the only reason they exist is because there are these techno feudal wealthy benefactors that make

00:34:48.000 them exist like would the heritage foundation exist if it weren't for like some very specific

00:34:52.560 no absolutely true well and this is what's interesting about our movement is we don't have that

00:34:56.480 yeah yeah i mean it yeah we we don't well jan tollen did make the biggest donation of anyone to the

00:35:04.080 private and it's it but it's it's it's tawdry compared to what all these other groups are getting

00:35:09.120 that's true yeah i mean yeah they're yeah i mean not to say that the 500 000 donation that we received

00:35:14.320 wasn't game-changing for the foundation we didn't take salaries or anything from it we only used it

00:35:18.480 toward the cause but yeah i just i realized like wow like so many of these big policy movements that we

00:35:24.880 have and cultural movements that we have aren't grassroots i mean they may have had grassroots origins

00:35:33.440 but the only reason they went anywhere was because someone with money gave them fuel

00:35:39.760 and it was kind of sobering because i was like i keep thinking about techno feudalism as something

00:35:44.480 that's going to happen in the future when maybe it's only something that's going to become much more

00:35:50.560 obvious in the future and and also like more fragmented and large scale in the future as well so yeah

00:35:57.920 still though i'm excited someone please build this so we can use it this this face scanner please oh

00:36:07.280 is that going to have to be our next major project building the face scanner can we do it like yeah i

00:36:13.360 guess i mean we can do anything it would be difficult it's not within my proficiencies or as easy as any of

00:36:18.720 the other technical projects i've taken on probably not because i mean like all of them are using probably

00:36:22.800 the same thing i mean what they're doing is that that typical mapping of looking at distances and measurements

00:36:27.920 and you're just turning each face into a series of raw data points and then you just can apply

00:36:34.160 kind of how nebula genomics takes your genome and then they just apply your oh my god it'd be so

00:36:40.720 cool in the future we have like augmented reality and when you see somebody's face it does a face scan

00:36:44.960 and like their personality yeah you're like you so you're wearing your ray bands who is it would be

00:36:49.440 like so in the new deus ex games they have it so like it tells you like the person's emotions and how

00:36:54.640 likely they are to respond to certain things and like that's basically what it would do oh my god

00:36:59.360 functional yes no 100 yeah just it's like somebody wants to partner with us on this business you let

00:37:04.720 us know we'll work with you okay morning high aggression oh my gosh i would yeah i would i would

00:37:11.440 probably buy ar glasses if and i'm i bet there will be like an app landscape for the various ar

00:37:19.360 glasses that are coming online it was just people would be so bad if you have like a face that looks

00:37:25.920 like a rapist face because you are one and it's like every girl you look at is like 98 percent

00:37:32.400 rapist yeah you mean because you aren't one oh well i mean i think the person probably is one but 0.98

00:37:39.200 um you know oh my god also are you really want to have the whole world no

00:37:48.240 gosh that is really gonna mess things up because this will happen like if the data is out there

00:37:54.160 people will develop it and then people will make hiring decisions and social decisions based on i mean

00:38:00.800 i'm sure governments you know people in democratic societies will create legislation that blocks this

00:38:06.880 from being used in hiring and in government and in law enforcement but that is not going to stop

00:38:13.440 private individuals from using this you prevent it from hiring but of course the hr manager has their

00:38:19.280 you know yeah just personally they're just wearing their goggles with the apple branded goggles

00:38:25.520 for knowing who's the the the the grapist yeah and we should partner with cutler and gross who makes

00:38:32.800 these glasses they also made the glass like the glasses frames from the kingsman movie and those

00:38:38.800 glasses were ar glasses weren't they oh yeah they were yeah and i mean it works because they're really

00:38:44.800 chunky so we should just partner we should reach out to them be like hey guys all right love you to

00:38:49.920 death simone what are we doing for dinner tonight if you really want we can try to squeeze in another

00:38:54.080 episode no i know you want you want to get out of here it's fine i know i just we have so many

00:39:01.360 backlog episodes to process it's okay as long as we get seven days a week patreon subscribers get get more

00:39:11.840 okay thank you then i was thinking

00:39:18.320 so i can do teriyaki chicken with rice if you'd like oh that sounds great yeah oh you know what'd be

00:39:25.040 good tonight honestly mac and cheese you feeling the the cheese you got the cheesies

00:39:32.160 or pesto we've got pesto let's do pesto do you want macaroni and cheese with pesto

00:39:38.320 a little bit of cheese with the pesto is always good

00:39:42.000 well would you just prefer we should probably just do pesto yeah do pesto throw in a bit of mac and cheese

00:39:48.400 okay well either i'm making mac and cheese for the kids and no okay you've done it before you've

00:39:56.320 thrown in a bit of the mac and cheese into the pesto and it was really good okay so you want me to do

00:40:01.200 that for you okay so i think i'm just gonna do plain noodles for the kids because that way i can

00:40:06.160 make mac and cheese for them tomorrow night okay i have a plan are we are you gonna take them out

00:40:11.680 in the boat should i charge the batteries so much fucking sorry i've got all these backlog episodes to 0.97

00:40:16.960 do you know so that means sitting down there and listening to episodes and going through and editing them 0.79

00:40:23.520 you guys don't know how long it takes to produce one of these episodes it is enormously time intensive

00:40:30.400 because you don't just have the recording period and the research period to begin with but then you've

00:40:34.480 got me going through it and editing the entire thing watching the entire thing very slowly adding

00:40:39.280 clips not everyone is as conscientious as you are just yeah other other producers no and i watch

00:40:45.360 other producers and they just waste their users time you know and like whatever i try to be you know

00:40:50.880 engaging throughout i appreciate that your dedication to the craft okay well then yeah i guess that's

00:40:59.440 why i'm confident that this will work out for us because i watch other people and i'm like

00:41:07.360 when i also there are other people who do invest a lot in editing who have grown a ton like who

00:41:13.680 actually take it seriously so like caleb hammer he puts so much work into his episodes and he's put so

00:41:21.360 much work into all of his like value added money making services and membership extras too so i'm like

00:41:29.680 okay maximum effort pays off for youtubers at least some and then some are just lucky and then some are

00:41:37.600 just kid influencers who get 85 million views on videos of them playing with army men

00:41:46.400 all right i'm gonna go downstairs and make dinner i love you very much i love you very much and i'm

00:41:54.000 so excited for the life i spent with you and tomorrow is saturday yeah so i will get two episodes to you

00:42:02.320 for the patreon viewers and and sub stack subscribers of course and thank you for making sure those go live

00:42:10.400 welcome i'm excited bye i was thinking about how you and i just kind of have this going assumption

00:42:27.520 that we're going to die at any moment and our kids do too they have all sorts of plans about what they're

00:42:33.520 going to do after we're dead oh yeah well i mean i'm so excited that we're almost 50 000 subscribers

00:42:38.960 our son octavian obviously wants 100 000 so we can get the plaque because he then wants us to die 0.84

00:42:46.000 so he can take care of it for us he is really he just wants to get his hands on that plaque

00:42:52.560 that placard a little no no yeah he said well when you die i'll take care of the youtube plaque for you

00:42:59.920 and he was very excited about this and i can understand why you know if we can get

00:43:04.880 you know before he gets to middle school let's get to a hundred thousand if we can get to a hundred

00:43:10.240 thousand a kid in middle school was a youtube plaque parents

00:43:17.280 that's how you know yeah

00:43:21.280 i don't know i don't know i mean eight passengers kids thought it was terrible having parents having a

00:43:29.440 mom that was on youtube so who knows i don't know i just i think i think maybe he'll make it to like

00:43:36.480 crazy youtube status before we do because when i look at kid content on youtube and i look at their

00:43:42.480 views like i look at top tier youtuber views and it's you know three million a video four million a

00:43:48.880 video 1.3 million per video then i look at the videos that our kids watch created by kid fluencers

00:43:55.120 87 million just it's like clown world views i can't wrap my head around it he'll be he'll be the

00:44:05.520 money maker of our family he'll make he'll make the videos for the little ones that just will run on

00:44:10.160 repeat yeah okay so question simone um episode today how did you do the my little pony episode

00:44:17.440 i i didn't look at comments yet the comments yeah i'm sorry well we had the interview somebody's

00:44:22.880 making like a broadway play on us and that'll be fun on fertility broadly 1.00

00:44:31.120 she seemed interested she was like wait are you guys under any

00:44:33.920 things i need to do more with you guys like in terms of who you're allowed to talk to oh my god

00:44:39.760 we have to go to this somebody's gonna be playing me on broadway talking that would be fun yes this

00:44:46.400 would be your first experience of who is cast as malcolm collins yes oh my god are they gonna make

00:44:54.160 it somebody who looks evil oh i hope so i hope i get a mustache that i can twirl people who don't

00:45:00.080 know by the way i don't naturally grow facial hair his family doesn't yeah thank you i'm not a facial hair

00:45:06.400 person i can't yeah i'm lucky right i actually get a smooth face thank god he believes this a guy was

00:45:13.440 like a i mean not perfectly like my dad i think if i get older it may get a bit coarser because my dad

00:45:20.960 puberty hits your family at like age 55 yeah no i i think it might be because he just doesn't shave

00:45:28.480 enough is is why that happens but yeah i it's it's it's i've always been very annoyed by like the you

00:45:35.360 know no shave november or whatever i'm like but nothing would happen and everyone else in the

00:45:40.880 office is doing it and they're like malcolm why aren't you being a bro why aren't you doing it

00:45:44.320 and i'm like i don't grow facial hair best trait by like honestly i could have rolled like genetics

00:45:52.480 and and you're like okay grow facial hair or not do you would all feel bad that you don't i'm like no

00:45:57.200 oh this is awesome so very very lucky about that particular genetic role yeah i i think it's

00:46:06.000 absolutely fantastic so don't change not that you can don't don't get genetically modified to shift this

00:46:13.840 no i know i'll have our kids genetically modified to make it even less i mean

00:46:19.360 yeah hmm i i'd prefer if i had literally none at all like i i just hate i think it's it's it's

00:46:25.920 it's gross and uncouth oh and i'm sure once adults can get genetically modified to eliminate hair in

00:46:33.040 like on legs on armpits oh yeah that'll be people will go in right away that that is a form of

00:46:41.280 cosmetic procedure i would totally get if you could just stop growing something imagine removing like

00:46:47.760 armpit sweat like instead of having botox yeah just genetically removing i mean i don't know maybe we

00:46:55.760 actually need it sometimes probably not though i i i think we did in a historic situation but i

00:47:03.440 don't know if we do i mean you might end up getting like a rash or something from things rubbing like

00:47:09.200 i remember reading a sci-fi book where a character had been genetically engineered i think maybe is like

00:47:16.080 a sex bot kind of human and she had really small pores but then she overheated like a lot it was it 1.00

00:47:24.720 was a big problem and i i don't know i could see that being some new form of female malady like women 1.00

00:47:30.960 just choosing to get really small pores because who wants pores but then at the same time like oh right 0.78

00:47:38.960 there's a process yes yeah we these were evolved for a reason speaking of porousness by the way a

00:47:46.560 material that some people still argue is better than gore text for serving as a a breathable waterproof

00:47:52.720 material because the problem with most waterproof materials people have used in the past to like

00:47:56.960 stay dry in cold environments or in wet environments is that if you sweat then you could get hypothermia

00:48:06.400 because you're wet inside your clothing like it has you have to evaporate away the moisture that you

00:48:12.320 sweat out when you are moving around in the cold environment and you know what they used like what some

00:48:18.720 inuit peoples used as the proto version of gore text i don't know seal fur or something no no seal and whale

00:48:29.520 intestines

00:48:32.800 oh goodness yeah they just sewed them together in a way that made them like the the the seams waterproof

00:48:39.200 which is super impressive and it took a long time to do this to make and you can see pictures they

00:48:44.000 they they look like they look like what they are they look like someone sewed together intestines and

00:48:50.880 is now wearing them wait okay seal no look whale intestine inuit raincoat let's see yep looks like

00:48:59.760 exactly what you think it would look like yeah that does look exactly what i think it would look like

00:49:04.720 it's a whale intestine raincoat yeah but it also looks like oddly in the way it's kind of puffy and and

00:49:12.400 and sort of drawn together it looks oddly modern yeah it does look oddly modern yeah it's just like

00:49:18.800 like as if there was some gross like skin color trend because it's kind of this this disagreeable 0.99

00:49:26.560 shriveled pale kind of like the color of my skin it's that it just i love that you think so little of

00:49:34.560 yourself all right i'll get started let's do it let's do it we're only going to do one today okay

00:49:38.560 oh wait we got started late okay okay and i've got a lot of backlog you give me extra episodes

00:49:45.920 those extra things for process i know but we're trying to shift to seven days a week two for patreon

00:49:50.640 members every week you punish me what have i done to deserve this but yes you've been too fun to talk

00:49:59.040 to malcolm i just want to keep it going honestly oh my gosh is this the hospital

00:50:09.840 you're climbing on top of the entrance that's called running triage you're the triage man

00:50:17.280 no he he means he's standing on the entrance

00:50:19.120 why are you why are you standing on the entrance do you think that's safe

00:50:26.480 i'm just looking if there is titan okay or not titan's a doctor she's waiting for you to fall

00:50:37.280 off and get hurt so she can make you all better right titan why

00:50:42.320 is that what do you want to be the patient or do we need a stuffed animal to be the patient 0.51

Based Camp - August 11, 2025

Phrenology is Back, Baby! AI is VERY Good at Making Predictions From Face Scans

Episode Stats

Summary

Transcript