Support NeoGAF

[badboy78660]

http://www.bbc.com/news/technology-41188560

A facial recognition experiment that claims to be able to distinguish between gay and heterosexual people has sparked a row between its creators and two leading LGBT rights groups.

The Stanford University study claims its software recognises facial features relating to sexual orientation that are not perceived by human observers.

The work has been accused of being "dangerous" and "junk science".

But the scientists involved say these are "knee-jerk" reactions.

Details of the peer-reviewed project are due to be published in the Journal of Personality and Social Psychology.

Narrow jaws

For their study, the researchers trained an algorithm using the photos of more than 14,000 white Americans taken from a dating website.

They used between one and five of each person's pictures and took people's sexuality as self-reported on the dating site.

The researchers said the resulting software appeared to be able to distinguish between gay and heterosexual men and women.

In one test, when the algorithm was presented with two photos where one picture was definitely of a gay man and the other heterosexual, it was able to determine which was which 81% of the time.

With women, the figure was 71%.

"Gay faces tended to be gender atypical," the researchers said. "Gay men had narrower jaws and longer noses, while lesbians had larger jaws."

But their software did not perform as well in other situations, including a test in which it was given photos of 70 gay men and 930 heterosexual men.

When asked to pick 100 men "most likely to be gay" it missed 23 of them.

In its summary of the study, the Economist - which was first to report the research - pointed to several "limitations" including a concentration on white Americans and the use of dating site pictures, which were "likely to be particularly revealing of sexual orientation".

'Reckless findings'

On Friday, two US-based LGBT-focused civil rights groups issued a joint press release attacking the study in harsh terms.

"This research isn't science or news, but it's a description of beauty standards on dating sites that ignores huge segments of the LGBTQ (lesbian, gay, bisexual, transgender and queer/questioning) community, including people of colour, transgender people, older individuals, and other LGBTQ people who don't want to post photos on dating sites," said Jim Halloran, chief digital officer of Glaad, a media-monitoring body.

"These reckless findings could serve as a weapon to harm both heterosexuals who are inaccurately outed, as well as gay and lesbian people who are in situations where coming out is dangerous."

"Stanford should distance itself from such junk science rather than lending its name and credibility to research that is dangerously flawed and leaves the world - and this case, millions of people's lives - worse and less safe than before," said its director of research, Ashland Johnson.

The two researchers involved - Prof Michael Kosinski and Yilun Wang - have since responded in turn, accusing their critics of "premature judgement".

"Our findings could be wrong... however, scientific findings can only be debunked by scientific data and replication, not by well-meaning lawyers and communication officers lacking scientific training," they wrote.

"However, if our results are correct, Glaad and HRC representatives' knee-jerk dismissal of the scientific findings puts at risk the very people for whom their organisations strive to advocate."

'Treat cautiously'

Previous research that linked facial features to personality traits has become unstuck when follow-up studies failed to replicate the findings. This includes the claim that a face's shape could be linked to aggression.

One independent expert, who spoke to the BBC, said he had added concerns about the claim that the software involved in the latest study picked up on "subtle" features shaped by hormones the subjects had been exposed to in the womb.

"These 'subtle' differences could be a consequence of gay and straight people choosing to portray themselves in systematically different ways, rather than differences in facial appearance itself," said Prof Benedict Jones, who runs the Face Research Lab at the University of Glasgow.

It was also important, he said, for the technical details of the analysis algorithm to be published to see if they stood up to informed criticism.

"New discoveries need to be treated cautiously until the wider scientific community - and public - have had an opportunity to assess and digest their strengths and weaknesses," he said.

 

[Linkstrikesback]

"Our findings could be wrong... however, scientific findings can only be debunked by scientific data and replication, not by well-meaning lawyers and communication officers lacking scientific training," they wrote.

That's where I stand too.

By all means, they can do a study/hire someone else to do a study to prove the work the other scientists did cannot be replicated or has some other flaw, but feelings don't disprove the scientific method.

 

[ZOONAMI]

That's where I stand too.

By all means, they can do a study/hire someone else to do a study to prove the work the other scientists did cannot be replicated or has some other flaw, but feelings don't disprove the scientific method.

I think these LGBT groups are attacking the science knowing they really shouldn't be, it's just they reeeeaaallly don't like the potential use cases for this AI. It screams slippery slope into identifying and persecuting gays.

 

[TheLostBigBoss]

I think these LGBT groups are attacking the science knowing they really shouldn't be, it's just they reeeeaaallly don't like the potential use cases for this AI. It screams slippery slope into identifying and persecuting gays.

this

I respect the science, but I can easily see the dangers of such tech if public opinion shifts, oppressive governments want to mass target homosexuals, or even smaller groups of people who want to make the lives of people a living hell.

 

[CrocMother]

I think these LGBT groups are attacking the science knowing they really shouldn't be, it's just they reeeeaaallly don't like the potential use cases for this AI. It screams slippery slope into identifying and persecuting gays.

I mean going by the current thinking, should such a thing even be possible? It sounds like eugenics.

 

[UncleSporky]

I think these LGBT groups are attacking the science knowing they really shouldn't be, it's just they reeeeaaallly don't like the potential use cases for this AI. It screams slippery slope into identifying and persecuting gays.

I still don't understand why it isn't a positive overall by proving people are born gay and it's not a choice or something that can be burned out of a person somehow.

 

[Aaronrules380]

I still don't understand why it isn't a positive overall by proving people are born gay and it's not a choice or something that can be burned out of a person somehow.

because people by and large aren't rational and don't think like that

 

[OG Shaka Zulu]

This machine will be more trouble than its worth. Ruling class all over will ramp up using gay people as scapegoats for their society's ills.

 

[Linkstrikesback]

I respect the science, but I can easily see the dangers of such tech if public opinion shifts, oppressive governments want to mass target homosexuals, or even smaller groups of people who want to make the lives of people a living hell.

Oh, absolutely. If this is accurate and gets in to the hands of, for instance, one of the middle eastern countries where homosexuality is a crime, it would be disastrous. That doesn't make the work inaccurate or wrong though.

 

[Dandy Crocodile]

It also only used white people so it's trash all around. 💁

 

[Kayhan]

If true this would add weight to the argument that gays are born that way or at the very least that there is a clear biological component. I thought that was the view promoted by the gay community as well.

"We are born this way. It is not a lifestyle choice"

 

[Nightwing123]

Didn't we already have a thread for this? Either way this is interesting.

 

[UncleSporky]

Didn't we already have a thread for this? Either way this is interesting.

Not specifically about the press release/concerns just raised here.

 

[EGM1966]

That's where I stand too.

By all means, they can do a study/hire someone else to do a study to prove the work the other scientists did cannot be replicated or has some other flaw, but feelings don't disprove the scientific method.

Yah agree. The biggest slippery slope right now is anti science and while I get their concern on usage there's been too much politics in science already. If it's wrong it needs to be proven to be wrong scientifically.

 

[buttery fistful]

 

[BuggyDClown]

I wonder what it would do with my friends who are identical twins but one is straight the other is gay

 

[PdotMichael]

What a pointless action. An evil country could archive the same without much work.

 

[nekkid]

I still don't understand why it isn't a positive overall by proving people are born gay and it's not a choice or something that can be burned out of a person somehow.

I don't think it proves people are born gay, as it uses keys such as hair and dress, I believe.

I dunno, we had a pretty great time after Ian Malcom said that.

Spoiler

I agree with you in this context.

 

[Charles Brandon]

Well, I respect the science...I guess. But this surely can't be true.

 

[Laekon]

If true this would add weight to the argument that gays are born that way or at the very least that there is a clear biological component. I thought that was the view promoted by the gay community as well.

"We are born this way. It is not a lifestyle choice"

Unless it can detect across age ranges it does nothing to further that argument. If they were doing genetic research that would be 1 thing but creating face recognition software that is focused on finding a minority group isn't useful In my opinion.

 

[PillarEN]

I wonder what it would do with my friends who are identical twins but one is straight the other is gay

Huh. Never actually thought about identical twins. Didn't know that identical twins could have different sexualities. Would have thought they were identical in that respect too.

 

[Jonnax]

Well I imagine it's a matter of time where some people build a smartphone app and start attacking people based on their results.

I can easily imagine that happening in Russia.

I'm sure the researchers will be all "I just did it for science!"

 

[UncleSporky]

I don't think it proves people are born gay, as it uses keys such as hair and dress, I believe.

Then it couldn't be used for persecution either, as people could just say "I'm not gay, honest, let me change my hairstyle and clothes and scan me again."

But IIRC from the original thread, their research may have resulted in something deeper than that. People said it doesn't just assume any big guy with a beard is a bear, somehow it can still figure it out with a high degree of accuracy.

 

[benicillin]

I predict this will probably fall apart under more scrutiny. Lots of "science" like this does.

 

[Fuchsdh]

I think these LGBT groups are attacking the science knowing they really shouldn't be, it's just they reeeeaaallly don't like the potential use cases for this AI. It screams slippery slope into identifying and persecuting gays.

Well that much is obvious. They can't make up their minds—is it "junk science" that doesn't work, or something that legitimately threatens to out people who aren't ready to be outed?

Until a peer-reviewed paper is out to be scrutinized it's all just whistling dixie.

 

[Maledict]

Huh. Never actually thought about identical twins. Didn't know that identical twins could have different sexualities. Would have thought they were identical in that respect too.

Nope it's well documented that you can have identical twins with different sexualities. Whatever makes us gay isn't exclusively genetic - it's believed to be a combination of genetic disposition plus environmental factors, and even that doesn't guarantee a gay person.

 

[TrutaS]

I don't believe this will ever reach a level of reliability to be used - and I'm glad for that. Simply because from personal experience these "narrow jaw" kind of evaluations are highly variable and fallible. This to me seems as reliable as my gaydar - pretty good but I wouldn't put my money on it.

With that in mind, I think the slippery slope of even attempting feels risky and I don't blame anyone for being concerned.

 

[Maledict]

I still don't understand why it isn't a positive overall by proving people are born gay and it's not a choice or something that can be burned out of a person somehow.

Because people all over the world are killed for being gay on a daily basis. Because Chechnya has concentration camps for gay people. Because even today, with gay marriage passed and everything else, LGBT kids try to kill themselves far more than other kids, and unfortunately often succeed.

We've come a long way in a short time but there's still a way to go, and allowing you to identify someone as gay or not via software is terrifying to a large number of gay people. Coming out is still an extremely traumatic experience for many people - imagine what it's like if someone does it for you via software?

 

[entremet]

Didn't we already have a thread for this? Either way this is interesting.

This is about the reaction to the tech. The original was about the tech itself.

 

[5taquitos]

Who knew resting gay face was a thing.

 

[LakeEarth]

But their software did not perform as well in other situations, including a test in which it was given photos of 70 gay men and 930 heterosexual men.

When asked to pick 100 men "most likely to be gay" it missed 23 of them.

Makes sense. Despite good accuracy scores, the fact that there are way more straight men than gay men means the software spits out a lot of false positives.

 

[xandaca]

Huh. Never actually thought about identical twins. Didn't know that identical twins could have different sexualities. Would have thought they were identical in that respect too.

If I remember correctly, the research suggests a much higher chance of both twins being gay if one is than any individual child, but it's by no means a sure thing.

 

[Kill Your Masters]

Then it couldn't be used for persecution either, as people could just say "I'm not gay, honest, let me change my hairstyle and clothes and scan me again."

But IIRC from the original thread, their research may have resulted in something deeper than that. People said it doesn't just assume any big guy with a beard is a bear, somehow it can still figure it out with a high degree of accuracy.

I'm sure the Saudis for example stone first, question later.

 

[Sai-kun]

If true this would add weight to the argument that gays are born that way or at the very least that there is a clear biological component. I thought that was the view promoted by the gay community as well.

"We are born this way. It is not a lifestyle choice"

unfortunately, the people who actually make the argument of it being a 'lifestyle' wouldn't really change their minds even if it was 100% scientifically proven that not being straight was genetic. they'd still hate us and view us as less than them.

 

[StarCreator]

My first reaction to hearing this a few days ago was that if this is possible, then it proves sexual preference at least has a genetic component. The consequences of that? All the people who think being gay is a choice are scientifically proven wrong.

 

[pantsmith]

I still don't understand why it isn't a positive overall by proving people are born gay and it's not a choice or something that can be burned out of a person somehow.

In an ideal world, yeah, this is what it proves.

In our current world the Mike Pences out there (basically irrational, emotional conservatives with a say in public policy) could use this to create a scientifically verifiable secondary class of citizen. Which is way bad.

This is to say nothing of the countries out there where it is illegal to be gay and people are regularly killed for it.

My first reaction to hearing this a few days ago was that if this is possible, then it proves sexual preference at least has a genetic component. The consequences of that? All the people who think being gay is a choice are scientifically proven wrong.

Most of the people who feel this way aren't basing their feelings on science. It wont win them over.

 

[Always-honest]

Pretty amazing if true (since it only solidifies the being born this way). But why the fuck would you want to know? WHo cares?
Well, i know who cares, but i hope they will never get their hands on software like this.

 

[Interframe]

So doesn't this whole AI project fall apart when you take into account masculine or "Straight-passing" (hate using both those terms) gay people, or even bisexual people?

Makes this whole thing sound ridiculous.

This AI seems to be identifying and grouping people into categories based on physical traits and characteristics, not human sexuality.

 

[shaneo632]

Exactly what I thought upon reading the OP.

 

[valeu]

a common plastic surgery for hollywood men is the "mandibular glove".....which gives you a square "man jaw"

 

[nekkid]

Then it couldn't be used for persecution either, as people could just say "I'm not gay, honest, let me change my hairstyle and clothes and scan me again."

But IIRC from the original thread, their research may have resulted in something deeper than that. People said it doesn't just assume any big guy with a beard is a bear, somehow it can still figure it out with a high degree of accuracy.

Oh yeah, I meant that those were some of the things it was looking for in the overall algorithm, not the only things.

But while those have to part of the equation it can't possibly prove this is nature over nurture, that's all I'm saying.

 

[UncleSporky]

So doesn't this whole AI project fall apart when you take into account masculine or "Straight-passing" (hate using both those terms) gay people, or even bisexual people?

Makes this whole thing sound ridiculous.

Deep learning algorithms take all that into account. It's not superficial when it's analyzing thousands/millions of photographs. Humans aren't inputting details like "if has feminine features then 70% gay," it's looking impartially at tons of individuals who identify as gay and building a profile. It might not be based on "feminine" features at all.

 

[Linkstrikesback]

Until a peer-reviewed paper is out to be scrutinized it's all just whistling dixie.

The paper is submitted and due to published. The pre-print can be viewed online already, so it's out to be scrutinized.

 

[Interframe]

Deep learning algorithms take all that into account. It's not superficial when it's analyzing thousands/millions of photographs. Humans aren't inputting details like "if has feminine features then 70% gay," it's looking impartially at tons of individuals who identify as gay and building a profile. It might not be based on "feminine" features at all.

The original study/trial didn't take into account bisexuals or people of color. That will make things that much more complicated for this algorithm to figure anything out.

Way too much ambiguity and variation for any mathematical pattern to be found when you bring in all that, especially when this is human beings and human sexuality we're talking about, a topic that we don't even fully understand yet.

 

[Keasar]

Reading about the device, I am reminded of an item called the G-Reader in the comic Transmetropolitan that scanned and gave back information if people had specific genes within them, which was used by groups of people to target and murder folks in a plot of the comic.

 

[PK Gaming]

Called it

This gif is no good 99% of the time, but it's pretty accurate here

 

[Beth Cyra]

If true this would add weight to the argument that gays are born that way or at the very least that there is a clear biological component. I thought that was the view promoted by the gay community as well.

"We are born this way. It is not a lifestyle choice"

In an ideal world it would be.

However as a lesbian/Trans person this tech is scary as fuck to me.

So many LBGT are attacked and even killed just for being who they are that this could lead to even worse treatment as people could easily target those like me even easier.

Let's be honest we have enough doctors and scientist already explaining how it's genetic that the vast majority who are willing to learn are going to with the info we already have, this tech would just be weaponizwd in many, many places.

Let's not forget Death camps for gays in places like Russia going on RIGHT NOW, what if the Russian government got this tech?

 

[A Fish Aficionado]

I really doubt it's replicability.
This is like an fmri case study all over again.

 

[hodgy100]

there are two ways this can go.

1. Its more proof that people are born a certain sexuality and that it isn't through choice and may further foster acceptability of LGBT's

2. it can be used to identify and persecute "the gays" and may lead to modern day witch hunts if used incorrectly

 

[Clefargle]

Imagine if Chechnya got a hold of this