AI Can Tell If You're Gay or Straight?

[+ quoththeraven]

I'm a little skeptical of the underlying claim, but it's worth a discussion, especially about the supposed physical markers of gayness.

 

https://www.theguardian.com/technology/2017/sep/07/new-artificial-intelligence-can-tell-whether-youre-gay-or-straight-from-a-photograph

[Chuckball]

I like men but not all men. In fact most of the guys I pass on the street never twig my curiosity. Then there is the minority that I must consciously avoid staring at. I have always thought that with a sixth sense gays could identify gays. The article suggests there may be some science behind my speculation.

On the other hand a quiz came up on Facebook the other day showing pictures with various images hidden and the reader is asked to check the image seen first. I completed the test and scored 100% straight - never had a homosexual experience and likely never will. I almost choked on the cock I was sucking.

Edited September 8, 2017 by Chuckball

[longtime lurker]

Oh I took that quiz too (or one similar to it) and was also 100% heterosexual. The whole joke involves whether or not you see naked ladies in images. You are a closeted gay if you make sure they are found at all times, just to "prove" you find them attractive. Either that or they disturb you enough to stick out. Lol!

 

As for this article line: "The data also identified certain trends, including that gay men had narrower jaws, longer noses and larger foreheads than straight men..." Sounds suspiciously like the way Jews were classified in old Nazi propaganda. On the plus side, a larger forehead may equal a bigger brain.

[LaffingBear]

Interesting study described. Except, its a study of people who post pictures on dating sites. Thats not exactly a representative sample.

 

Do gay men on dating sites groom themselves differently? And are there some features for gay men openly acknowledging their orientation in hopes of attracting another gay man that trigger facial recognition?

 

Facial recognition and profiling make for fascinating ethical debate.

Edited September 8, 2017 by LaffingBear

[+ Funguy]

Oh I took that quiz too (or one similar to it) and was also 100% heterosexual. The whole joke involves whether or not you see naked ladies in images. You are a closeted gay if you make sure they are found at all times, just to "prove" you find them attractive. Either that or they disturb you enough to stick out. Lol!

 

As for this article line: "The data also identified certain trends, including that gay men had narrower jaws, longer noses and larger foreheads than straight men..." Sounds suspiciously like the way Jews were classified in old Nazi propaganda. On the plus side, a larger forehead may equal a bigger brain.

 

What about the horn in our forehead?

[+ quoththeraven]

Looks like my skepticism puts me in good company. If they're measuring noise, the ethical dilemmas are more hypothetical than real.

Edited September 9, 2017 by quoththeraven

[+ sync]

I have also taken such a test and was graded as 100% straight. Terrific, more self-doubt.

[+ sniper]

90% accuracy isn't really as good as it sounds when you are searching for something with a frequency that is 10%. Think about it -100 people, 90 of them gay, 10 of them straight. The AI will catch 9 of the actual 10 gay people and also mislabel 9 of the straight people as gay. So it will identify 18 people as gay, and only half of them will be. It gets worse when the frequency of gayness is lower.

This is the reason lie detectors aren't useful to catch a wrongdoer in a company, say.

[MikeyGMin]

90% accuracy isn't really as good as it sounds when you are searching for something with a frequency that is 10%. Think about it -100 people, 90 of them gay, 10 of them straight. The AI will catch 9 of the actual 10 gay people and also mislabel 9 of the straight people as gay. So it will identify 18 people as gay, and only half of them will be. It gets worse when the frequency of gayness is lower.

This is the reason lie detectors aren't useful to catch a wrongdoer in a company, say.

I'm not sure about your math, but I was never that good at statistics in the first place.

 

I'm more interested in the concept that gay men's facial features have been somewhat feminized due to genetics or biology. I will admit that it feeds into a fundamental belief that I have, not priven by science (yet), that gay men are born gay due to genetics and or hormones while in the womb. Then they are socialized in different ways that determine whether they act out on their natural orientation.

 

We have all seen men that you can tell are gay just by looking at them. And I'm talking just about their features not because of mannerisms or having the current trendy gay haircut, etc. I think it's easier to see in teenagers before they have had a chance to hyper-masculinize with muscles, steroids and facial hair.

 

However, I don't think that 81% of the gay men I know fall into this category. So the fact that AI can detect something better than the naked eye is pretty impressive.

[OCClient]

I'm not sure about your math, but I was never that good at statistics in the first place.

 

I'm more interested in the concept that gay men's facial features have been somewhat feminized due to genetics or biology. I will admit that it feeds into a fundamental belief that I have, not priven by science (yet), that gay men are born gay due to genetics and or hormones while in the womb. Then they are socialized in different ways that determine whether they act out on their natural orientation.

 

We have all seen men that you can tell are gay just by looking at them. And I'm talking just about their features not because of mannerisms or having the current trendy gay haircut, etc. I think it's easier to see in teenagers before they have had a chance to hyper-masculinize with muscles, steroids and facial hair.

 

However, I don't think that 81% of the gay men I know fall into this category. So the fact that AI can detect something better than the naked eye is pretty impressive.

I agree, hormones play a critical role where they are a factor in the outcome of an expected chain of events, and that outcome clearly varies, in a number of categories you could say. The outcome of a certain gay look can only be a sort of covariate to same sex attraction IMO. It's all very interesting. How we are socialized affects behavior, of course, but I doubt it is a significant factor in driving ones attraction to the same sex. Sometimes one doesn't need a Stanford Gaydar to tell, when a certain kind of fellow creates such an amazing gay construct, as if he's on stage 24/7.

[Courage]

Wouldn't the Turing Test suggest proper AI generate the responses indistinguishable from humans? That said, I suppose I know humans who claim "gaydar" is a thing...

[MikeyGMin]

Wouldn't the Turing Test suggest proper AI generate the responses indistinguishable from humans? That said, I suppose I know humans who claim "gaydar" is a thing...

You don't believe in gaydar?

[Courage]

You don't believe in gaydar?

 

I believe it is something some people think they have. I suspect it picks up on some gays with some traits fairly reliably. I suspect it misses people who lack those traits but are gay--i.e., men attracted to men but who are butch, closeted, and/or in denial.

[+ Gar1eth]

Here's the article on LBQTNation. I was just about to post a thread when I saw @quoththeraven

 

 

The interesting things I thought

 

1. Before reading the article I thought the guy on the right might be gay-and this is the composite gay face.

 

2. Maybe it's being in Seattle or knowing a lot of Bear types. But most of the gay men I know have some sort of facial scruff when the article mentions we are more likely to be clean shaven. Possibly the results would be skewed by a dating site having younger people on it and more older people have scruff on their face?

 

3. I don't think I really had an inkling towards the composite female faces on which was the gay and which was the straight one.

 

Computers can now predict if white people are gay or straight

Alex BollingerSaturday, September 9, 2017

 

Wang and Kosinski

Two researchers published a study that showed that computers can see whether someone is gay from their faces with surprising accuracy.

 

Advertisement

 

Michal Kosinski and Yilun Wang used about 35,000 images of people from an American dating website (they don’t name the site). About half were men and half were women. Half the men and women identified as gay, the other half identified as straight.

 

Also, the images were all from white people between the ages of 18 and 40, and people who did not look like the gender in their profiles (according to workers hired by the researchers) were excluded. Bisexual people were not included in the study. The pictures were cropped to focus on the face, and the faces were fully visible.

 

They plugged those images into Face++, a program that analyzes and recognizes faces, in order to develop a model that could assign a probability that someone is gay or straight, depending on hundreds of traits that the program found in people’s faces.

 

The researchers then used that model to assign probabilities to another set of face pics. When comparing a gay man’s face to a straight man’s face, the model assigned a higher probability of being gay to the gay man 81% of the time. It gave the gay man a higher probability 91% of the time when five pics were available. For women, the model was correct 71% of the time with one image and 83% of the time when there were five pics available. (They tested the model on pics from Facebook too, just in case dating sites are weird, and they got similar results.)

 

They also used the program to determine what parts of the face are more informative about people’s sexual orientation. Red areas provided the most information about sexual orientation.

 

Wang and Kosinski

Using the 100 images most likely to be gay and straight for both men and women, they created composite face images.

 

Wang and Kosinski

The most informative traits were both things that can be easily controlled by people and things that can’t. Lesbian women had larger jaws and narrower foreheads than straight women, and they were also more likely to wear hats and not wear make-up in their pictures. Gay men had narrower jaws and longer noses than straight men, but they also were less likely to have facial hair.

 

So can computers tell who is gay now?

Not really.

 

The study found that a gay person’s picture would usually be assigned a higher probability of being gay than a straight person’s picture.

 

So, for example, if two pictures were compared and one had a 54% chance of being gay and the other had a 43% chance of being gay, most of the time the first one will be the one from a gay person. But a 54% chance of a picture being from a gay person is not enough for anyone to figure out that someone is gay.

 

Still, with more images to train on, better quality images, and better computer programs that will be developed in the next few decades, this could change.

 

Advertisement

 

Does this mean people are born gay?

Wang and Kosinski mention this in relation to the theory that sexual orientation is set in the womb based on hormones that fetuses get exposed to. If a male fetus gets exposed to fewer androgens (hormones that include testosterone) than average, then it might turn out to be a gay boy, and the reverse for female fetuses.

 

Considering how some of the identifying features are clearly not the result of fashion choices – like the width of the forehead, the length of the nose, and the shape of the jaw – it certainly supports the idea that sexual orientation is linked to early development. The study also found that gay men’s faces were more like women’s faces than straight men’s faces, and lesbians’ faces were more like men’s faces than straight women’s faces, which also suggests that sexual orientation is related to the development of biological sex.

 

What about people of color?

An interesting little piece of data in this study is that gay men had lighter skin than straight men. Wang and Kosinski suggest this could be related to exposure to the sun (gay boys are the “indoor recess” crowd) or testosterone (which is linked to melanin production), but consider a third possibility: gay men are less likely to identify as white compared to straight men.

 

We tend to think of race as an immutable characteristic assigned at birth based on someone’s ancestry. But a lot of people exist in complicated positions in America’s racial classification system and they make choices about their race, mainly because they live in a culture that tells them they have to choose one word to describe where hundreds of ancestors came from, how they understand that ancestry, and how others see them. When people change their racial identity between Censuses, it’s pretty clear that there is some choice in the matter even if people aren’t going full Rachel Dolezal.

 

So it could be that gay men who are at the edges of whiteness – Hispanic gay men, biracial gay men, etc. – are more likely to check the other box than their straight siblings because they already feel like they’re outsiders.

 

That said, race is a social construct and it would have been interesting if the study had included all races. While Wang and Kosinski say they didn’t have enough pictures of gay people of color to do separate studies on them, a broader sample that includes all races would have avoided differences in racial identity.

 

Privacy concerns

The most obvious issue raised here is how this technology can be used to violate people’s privacy. As Wang and Kosinski point out, they used pretty common software and techniques that are known to people who work with this kind of data. They didn’t develop anything new – they just showed what could be done with stuff that’s already out there.

 

Considering how many pictures people put online, even in professional contexts like on LinkedIn, a better version of this software could be used to out people.

 

While that might be terrifying to some people, it’s also important to remember that there are other oppressed minorities out there who are easily recognizable, so being easily seen as gay won’t put us in new territory. Also, if straight people could see how many gay people they already know, it would do a lot to change attitudes.

 

Either way, the technology isn’t going to go away because we’re afraid of it, so the most we can do is to reduce homophobia as much as possible.

 

 

 

Gman

[+ quoththeraven]

A vigorous rebuttal to the study:

 

https://scatter.wordpress.com/2017/09/10/guest-post-artificial-intelligence-discovers-gayface-sigh/

[Kevin Slater]

The study from Stanford University... found that a computer algorithm could correctly distinguish between gay and straight men 81% of the time, and 74% for women.

 

I'll put my Stanford degree to use and develop a better algorithm. And here it is. I worked on it for about 46 seconds:

Since reportedly 3.8% of us are gay, my model is correct 96.2% of the time. (My model borrowed the previous study's assumption of binary sexuality.)

 

Kevin Slater

[+ quoththeraven]

The study from Stanford University... found that a computer algorithm could correctly distinguish between gay and straight men 81% of the time, and 74% for women.

 

I'll put my Stanford degree to use and develop a better algorithm. And here it is. I worked on it for about 46 seconds:

Since reportedly 3.8% of us are gay, my model is correct 96.2% of the time. (My model borrowed the previous study's assumption of binary sexuality.)

 

Kevin Slater

Which is better than theirs.

[+ sniper]

Not exactly, their methodology was to put up two pictures, one of whom was straight and one of whom was gay, and pick the straight one. Saying straight to both would be wrong 50 percent of the time....

[longtime lurker]

A vigorous rebuttal to the study:

 

https://scatter.wordpress.com/2017/09/10/guest-post-artificial-intelligence-discovers-gayface-sigh/

 

I should have been more straight-to-the-point earlier. I agree that the AI business is very bogus because you have to utilize certain stereotypes in order for it to work. It is like saying blacks are supposed to have bigger lips so they can eat watermelon better. In the 1940s, the Kinsey reports claimed that the majority tends to be "in between" homosexual and heterosexual somewhere on a grid. Later in the post-McCarthy '50s, Masters & Johnson had statistics claiming the number of homosexuals and bisexuals in America was far fewer than the Kinsey studies a decade earlier. Some think it is because Alfred Kinsey himself was bisexual and probably wanted the majority to be like him, but I think it is due to that post-WW2 period when study of sex was brand new and the researchers were great with subjects' privacy contrasting to a far less tolerant and more repressive period when the subjects were less trusting and less honest (i.e. better to be heterosexual in all your reporting). What this study here might confirm is whether or not a person would be more open about being gay with others and accepting certain cultural aspects in appearance to fit in with a still segregated against group.

[+ VeryHappyCustomer]

People are repeating his experiment already. I predict that within six months we will all know whether they replicated his results.

[+ friendofsheila]

I'm a little skeptical of the underlying claim, but it's worth a discussion, especially about the supposed physical markers of gayness.

 

https://www.theguardian.com/technology/2017/sep/07/new-artificial-intelligence-can-tell-whether-youre-gay-or-straight-from-a-photograph

 

I barely read this post, and certainly did not click on the link and read the article, so I'm going to reply by saying:

 

How do you suck someone's dick by way of a photograph? If they've figured that one out, they can sell the technology in membership for to gay men and earn enough money to build themselves a Death Star.

 

[samhexum]

[samhexum]

 

[+ Gar1eth]

 

????

 

There's no there here.

 

Gman

[samhexum]

???? There's no there here. Gman

 

 

Sometimes posted comic strips stay for a long time, sometimes they disappear fairly quickly. It's the same thing in the 'good pun' thread.