New AI n gus whether you're gay or straight om a photograph | TexAgs

Contents:

• STANFORD UNIVERSY'S AI N TELL IF YOU'RE GAY OR STRAIGHT OM A PHOTO
• THIS AI CLAIMS TO KNOW WHETHER YOU’RE GAY OR STRAIGHT OM A SGLE PHOTO
• AI CAN TELL IF YOU'RE GAY: ARTIFICIAL INTELLIGENCE PREDICTS SEXUALY FROM ONE PHOTO WH STARTLG ACCURACY
• THREAD: NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH
• NEW AI N WORK OUT WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH
• NEW AI N GUS WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH
• NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH
• NEW AI CAN GUS WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH
• NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH, 91% OF THE TIME

STANFORD UNIVERSY'S AI N TELL IF YOU'RE GAY OR STRAIGHT OM A PHOTO

Michal Kosski and Yilun Wang's algorhm n accurately gus if a person is straight or gay based on a photo of their face.

This was achieved g a sample of over 35, 000 imag that had been publicly posted on a US datg se, wh an even reprentatn of gay and straight (termed through their profil) and male and female subjects.

THIS AI CLAIMS TO KNOW WHETHER YOU’RE GAY OR STRAIGHT OM A SGLE PHOTO

Gay men tend to have "genr-atypil" featur, exprsns and groomg styl, meang they appeared more feme. The oppose was te for gay women, who generally had larger jaws and smaller foreheads. The system was eventually able to distguish between gay and straight men 81 percent of the time, and 74 percent of the time for women, jt by reviewg a photo.

AI CAN TELL IF YOU'RE GAY: ARTIFICIAL INTELLIGENCE PREDICTS SEXUALY FROM ONE PHOTO WH STARTLG ACCURACY

The fdgs strongly support the theory that the same hormon affectg a fet' velopg bone stcture play a part termg sexualy, meang that people are born gay - 's not a choice. Google rults show the phrase "is my hband gay?

There's also the ncern that the software uld be ed untri such as Iran, Sudan, Sdi Arabia and Yemen, where homosexualy is punishable by ath.

A new artificial telligence system has been created that claims to be pable of accurately gusg whether a person is gay or straight, based purely on a photograph of their face. Rearchers om Stanford Universy veloped a ep nral work – ed to analyse visual patterns om large datasets – and extracted featur om 35, 000 imag pulled om a US datg creatg a predictive mol om patterns the data, they showed the system pictur of gay and straight men and women. They found the work was able to rrectly distguish sexual orientatn 81% of the time for men, and 74% for rults raise a number of sensive qutns about the blogil origs of sexual orientatn, as well as the potential for facial regnn systems to be abed as a tool for LGBT study, published the Journal of Personaly and Social Psychology and first picked up by the Enomist, found that the fac of gay men and women had “genr-atypil” featur and “groomg styl” rearchers – Dr Michal Kosski and Dr Yilun Wang – say this provis “strong support” prenatal hormone theory (PHT) of sexual orientatn, which poss that exposure to certa hormon before birth n terme sexualy.

THREAD: NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH

Their fdgs supported PHT nclns that gay men tend to have narrower jaws, longer nos and larger foreheads pared to heterosexual men, while gay women tend to have smaller foreheads pared to heterosexual study also asked humans to gus the sexualy of the subjects.

On the other hand, the nral work was found to improve s accuracy when shown five imag of each person, judgg rrectly 91% and 83% of the time for men and women the paper offers a shockg glimpse to how AI uld be ed to profile cizens, s thors fend their rearch as a necsary statement about the technology’s potential the start of their paper, Kosski and Wang note that the predictabily of sexual orientatn “uld have ser and even life-threateng implitns to gay men and women and the society as a whole”.

They pot to reports of ernments and rporatns velopg face-based predictn tools to judge the likelihood of a person mtg a crime, and also note that several untri nsir homosexualy to be a crimal act:“It is th cril to rm policymakers, technology pani and, most importantly, the gay muny, of how accurate face-based predictns might be. Artificial telligence n now tell whether you are gay or straight simply by analyzg a picture of your Stanford Universy rearchers have reported startlg accuracy predictg sexual orientatn g puter Michal Kosski and Yilun Wang, whose rearch will be published by the Journal of Personaly and Social Psychology, say that AI n distguish between the face of a heterosexual man and a homosexual man 81 percent of s. For women, the predictive accuracy is 71 parison, the scientists say, the average human is ls apt at intifyg between straight and gay people purely based on an image: We are only able to gus rrectly 61 percent of s for men and 54 percent for scientists prented the algorhm wh five facial imag of a sgle person, the accuracy creased to 91 percent for men and 83 percent for women.

NEW AI N WORK OUT WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH

Advanced puter analysis pared different facial characteristics and found that gay men and women tend to have "genr-atypil" featur: fixed featur such as the shape of one's nose or jaw, and transient featur such as hairstyle and facial scientists then tted this ia to measure the "facial femy" of each image, or the probabily of that face beg female.

"The rults show that the fac of gay men were more feme and the fac of lbians were more mascule than those of their rpective heterosexual unterparts, " the study said. "Among men, the data-driven measure of facial femy posively rrelated wh the probabily of beg gay.

However, Kosski tells The Enomist, the rearch is not tend to be ed to profile or "out" homosexual men and women.

NEW AI N GUS WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH

Artificial telligence n accurately gus whether people are gay or straight based on photos of their fac, acrdg to new rearch that suggts mach n have signifintly better “gaydar” than humans.

The study om Stanford Universy – which found that a puter algorhm uld rrectly distguish between gay and straight men 81% of the time, and 74% for women – has raised qutns about the blogil origs of sexual orientatn, the ethics of facial-tectn technology, and the potential for this kd of software to vlate people’s privacy or be abed for anti-LGBT purpos.

NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH

The rearch found that gay men and women tend to have “genr-atypil” featur, exprsns and “groomg styl”, sentially meang gay men appeared more feme and vice versa. The data also intified certa trends, cludg that gay men had narrower jaws, longer nos and larger foreheads than straight men, and that gay women had larger jaws and smaller foreheads pared to straight women. The paper suggted that the fdgs provi “strong support” for the theory that sexual orientatn stems om exposure to certa hormon before birth, meang people are born gay and beg queer is not a choice.

NEW AI CAN GUS WHETHER YOU’RE GAY OR STRAIGHT OM A PHOTOGRAPH

Like any new tool, if gets to the wrong hands, n be ed for ill purpos, ” said Nick Rule, an associate profsor of psychology at the Universy of Toronto, who has published rearch on the science of gaydar.

NEW AI N GUS WHETHER YOU'RE GAY OR STRAIGHT OM A PHOTOGRAPH, 91% OF THE TIME

The om Stanford Universy – which found that a puter algorhm uld rrectly distguish between gay and straight men 81% of the time, and 74% for women – has raised qutns about the blogil origs of sexual orientatn, the ethics of facial-tectn technology, and the potential for this kd of software to vlate people’s privacy or be abed for anti-LGBT purpos.

3 Steps to figure out your Sexualy3 Steps to figure out your SexualyArtificial telligence n accurately gus whether people are gay or straight based on photos of their fac, acrdg to new rearch that suggts mach n have signifintly better “gaydar” than study om Stanford Universy – which found that a puter algorhm uld rrectly distguish between gay and straight men 81% of the time, and 74% for women – has raised qutns about the blogil origs of sexual orientatn, the ethics of facial-tectn technology, and the potential for this kd of software to vlate people’s privacy or be abed for anti-LGBT mache telligence tted the rearch, which was published the Journal of Personaly and Social Psychology and first reported the Enomist, was based on a sample of more than 35, 000 facial imag that men and women publicly posted on a US datg webse. The rearchers, Michal Kosski and Yilun Wang, extracted featur om the imag g “ep nral works”, meang a sophistited mathematil system that learns to analyze visuals based on a large rearch found that gay men and women tend to have “genr-atypil” featur, exprsns and “groomg styl”, sentially meang gay men appeared more feme and vice versa. The data also intified certa trends, cludg that gay men had narrower jaws, longer nos and larger foreheads than straight men, and that gay women had larger jaws and smaller foreheads pared to straight judg performed much worse than the algorhm, accurately intifyg orientatn only 61% of the time for men and 54% for women.