Friday, September 08, 2017

Artificial Intelligence Backs 'Gay Face' Theory


You and your friends have been saying it for years, but now scientists are putting it to the test. The Independent reports that artificial intelligence can figure out a person’s sexual orientation by analyzing a picture of their face.

According to the authors of a controversial new study, who say they were "really disturbed" by their findings, the accuracy of an AI system can reach 91 percent for homosexual men and 83 percent for homosexual women.


I'm guessing the accuracy in women is lower because of this

The study also concludes that homosexual men and women tend to have “gender-atypical facial morphology, expression and grooming styles.”

The paper, titled "Deep Neural Networks Are More Accurate Than Humans at Detecting Sexual Orientation From Facial Images," was co-written by Stanford University’s Yilun Wang and Michal Kosinski, and first spotted by the Economist.


In it, they claim to “show that faces contain much more information about sexual orientation than can be perceived and interpreted by the human brain."

To find out more about why so many gay men look like they just sat on a huge c**k, keep reading HERE.


2 comments:

Unknown said...

I'm looking forward to Grinder adding "Gacial Recognition" to their next version.

Edgar_Carpenter said...

Several gay organizations have issued statements pointing out the serious methodological flaws with the study - it's not likely to be replicable when studying different or more culturally reflective LGBT cohorts.

And, although it's probably not going to stand up to further research, they pointed out that anti-gay governments may attempt to use it to ID gay people, leading to bad results for both gay and non-gay people.

This has been the kind of science "journalism" that makes journalists look bad, in most outlets.