Controversial Study Links Facial Features to Sexual Orientation

Stanford University researchers are responding to national LGBTQ groups that have called their study on using artificial intelligence to determine sexual orientation “dangerous and flawed.”

In their report released this month, “Deep neural networks can detect sexual orientation from faces,” Michal Kosinski, an assistant professor of organizational behavior at Stanford, and Yilun Wang, who studies computer science at the school, extracted features from more than 35,000 facial images that had been posted to a dating website and entered them into a logistic regression designed to classify sexual orientation. (The pair established people's orientation by looking at the gender of the partners they were seeking.)

“Given a single facial image, a classifier could correctly distinguish between gay and heterosexual men in 81 percent of cases, and in 74 percent of cases for women,” Kosinski and Wang said in a summary of their findings.

In a Sept. 8 statement, the national groups GLAAD and Human Rights Campaign blasted the research, adding that it “could cause harm to LGBTQ people around the world.”

Kosinski and Wang responded through a statement of their own, saying the groups “do a great disservice to the LGBTQ community by dismissing our results outright without properly assessing the science behind it, and hurt the mission of the great organizations that they represent.”

In their research, the pair said, “Human judges achieved much lower accuracy” of accuracy than the algorithms, with "61 percent for men and 54 percent for women.”

When five facial images for each person were used, “the accuracy of the algorithm increased to 91 percent and 83 percent, respectively,” they said.

They added that among other findings, “Composite faces” that were examined “suggest that gay men had larger foreheads than heterosexual men, while lesbians had smaller foreheads than heterosexual women.”

Kosinski and Wang weren’t available for comment for this story, but in their statement, they acknowledged: “Given that companies and governments are increasingly using computer vision algorithms to detect people’s intimate traits, our findings expose a threat to the privacy and safety of gay men and women.”

Among other concerns, GLAAD and HRC said that the research hadn’t been peer reviewed and that information from the online profiles hadn’t been independently verified.

“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 community, including people of color, transgender people, older individuals, and other LGBTQ people who don’t want to post photos on dating sites,” Jim Halloran, GLAAD’s chief digital officer, said in a statement. “At a time where minority groups are being targeted, these reckless findings could serve as 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.”

In response to such criticism, Kosinski and Wang wrote that the study “was peer reviewed and accepted for publication in the Journal of Personality and Social Psychology, the leading academic journal in psychology. … In addition, before it was sent for a formal peer review, the manuscript was reviewed by over a dozen experts in the fields of sexuality, psychology, and artificial intelligence.”

They also addressed GLAAD and HRC's complaint about information not being independently verified.

“We put much effort into ascertaining that our data was as valid as possible, and there are no reasons to believe that there are gross inaccuracies,” the researchers said. “Our approach was no different than in other similar studies. More than a dozen scholars who have reviewed this work did not see any issues in how we handled those variables.”

In a call with local media outlet Bay Area Reporter, Jessica Stern, executive director of OutRight Action International, said she wasn’t aware of any governments using artificial intelligence algorithms like the ones in the study, but “governments around the world use arbitrary standards to target LGBTI people already” and use social media and dating sites “to trap people.”

Stern, who hadn’t read the Stanford study, said, “I don’t want to see any more tools put in the hands of governments that could use so-called irrefutable evidence to target LGBTI people.”

Kosinski and Wang indicated they’re as concerned as anyone else that “widely available tools can be used to detect sexual orientation from images of people’s faces,” which “may already” be taking place.

“Our findings could be wrong,” they said. “In fact, despite evidence to the contrary, we hope that we are wrong. … If our findings are wrong, we merely raised a false alarm. 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 organizations strive to advocate.”

This article was originally published Sept. 19, 2017, in the Bay Area Reporter.  


  1. Add Kosinski and Wang to the long list of scientists who’ve discovered the Left’s regard for science begins and ends with its political agenda. From Darwin — whose natural selection has been bastardized to allow for equality, to Murray and Herrnstein — shunned for publishing disallowed truths in The Bell Curve, to J. Michael Bailey — targeted for ruin for publishing findings disliked by transgenders, Stanford’s new targets of Leftist persecution will discover that truth is no defense, reason takes second to emotion, and the scientific method is no more secure from toppling than is a Confederate statue. In other words, intellectually the Left is indistinguishable from chimpanzees: capable of anger and tantrums but otherwise incapable.

    Imagine how remarkable it would be were Kosinski and Wang incorrect, and that over the tens of millions of years of primate evolution nothing connected with sexual orientation — not an observable difference nor detection skill, ever developed despite the significant reproductive penalties of uninformed investing. With all the fertility and status-related signaling going on between heterosexuals, Mother Nature would’ve had to have pulled-off a miracle of politically-correct manipulation to keep concealed the identity of those with the least value in the reproduction market. Of course, given Mother Nature’s obsession with successful reproduction, that miracle never happened, rendering any statue erected to her in grave jeopardy.

  2. Trying to submit this again. This is old news, old theories on sexuality. A very popular page from way back in the early days of the internet proposed that you could tell if someone is a pedeophile simply by if they had a “Pedosmile”, a half smirk. (Disclaimer, pedophelia is considered a crime by some, a sexuality by others. I think it’s the former, and in no way does homosexuality between adults represent pedophilia, just trying to make the point this isn’t the first time the theory of telling a persons sexuality based on certain visual clues has been proposed)

    I think that was 1997 or so when that came out.

    In 2011, a scientific researcher put more thought and research into this hypothesis, and cites a few scientific articles and proposes:

    If the pedosmile is present in childhood and predictive of pedophile offenses, then intense research should be undertaken to understand the biological, social and environmental inputs in its creation.

    The only difference between now and then is the use of computational neural networks. Is it dangerous and flawed? I don’t see how the analysis can be flawed (this, after reading all 47 pages of the research) This is research with no connection to government or law enforcement here in the US. Certainly a country of Sharia law would abuse such a system, but we have export controls on certain technologies.

    Thing is, it’s very well documented that a persons default facial expression can relay some type of insight into a persons emotional state. We can all look at each other and tell if we’re happy, sad, etc (except for sociopaths, they’re great at faking it)

    If anything, the LGBT folks should praise this research, because it clearly cements their stances on “We don’t choose to be homosexual” Maybe the problem is as a society, we’ve repressed our ability to see these cues in homosexuals in the name of political correctness. Maybe we should be OK with people “Looking gay” because it’s no different than “looking sad” It’s just another facet on the spectrum of human emotion.

  3. I’ve heard gay inlaws describing their use of “gaydar” to determine whether someone is gay or not gay.

    I’ve never been given an explanation of how it works.

    • But more importantly, as I watch Monday Night Football, does it say something important about our society that suspected female abuser, the black, presumably oppressed and deserving of social justice discriminatee Ezekiel Elliot just scored a touchdown that benefited Jerry Jones, his white, presumably racist plantation owner style boss who was promptly kissed and hugged by a beautiful young white lady by his side who presumably hasn’t been given the memo regarding the glass ceiling, equality of the sexes and all that crap? Did Ezekiel correctly, respectfully, reverently kneel during the national anthem? Did Jerry Jones correctly kneel? Did his bimbo girlfriend? And did any of them vote for Donald Trump? These are the questions that need to be answered before our diverse society can come together and progress forward in a fashion that is acceptable to the fussier, the pickier, the more anal retentive, in other words, the more Democrat among us.

  4. The Stanford Gaydar Machine! How many tax dollars did we, (pardon the pun), blow on this toy?
    I wonder if someone could do something useful like spend $300 on a camera that could Identify gang members with face tattoos!

  5. Oddly enough, research at Durham University identified composite faces similar to Stanford’s gay ones as “Players” as opposed to relationship material. Stanford’s straight men and women are “Boyfriends” and “Girlfriends” in the Durham study

    Having been pleased for some time to have the “Player” look, now I feel like the Ajay Naidu character in “Bad Santa.”

Leave a Reply

Your email address will not be published. Required fields are marked *