Cited by 778Face Recognition and PerceptionFace recognition and analysisEvolutionary Psychology and Human Behavior

The DOI matches its Crossref record; 53 of the 66 references it cites were found in a registry or an index.

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Open access

Wang, Yilun, Kosinski, Michal·2018·Journal of Personality and Social Psychology

Published version of a preprint

Open this paper in CoChat — free account

Abstract

We show that faces contain much more information about sexual orientation than can be perceived or interpreted by the human brain. We used deep neural networks to extract features from 35,326 facial images. These features were entered into a logistic regression aimed at classifying sexual orientation. Given a single facial image, a classifier could correctly distinguish between gay and heterosexual men in 81% of cases, and in 71% of cases for women. Human judges achieved much lower accuracy: 61% for men and 54% for women. The accuracy of the algorithm increased to 91% and 83%, respectively, given five facial images per person. Facial features employed by the classifier included both fixed (e.g., nose shape) and transient facial features (e.g., grooming style). Consistent with the prenatal hormone theory of sexual orientation, gay men and women tended to have gender-atypical facial morphology, expression, and grooming styles. Prediction models aimed at gender alone allowed for detecting gay males with 57% accuracy and gay females with 58% accuracy. Those findings advance our understanding of the origins of sexual orientation and the limits of human perception. Additionally, 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. (PsycINFO Database Record

Integrity signals

Registry

Crossref · 10.1037/pspa0000098

Cites

53 of 53 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Deep learning
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Deep Face Recognition
Parkhi, O. M., Vedaldi, A., & Zisserman, A. (2015) · Procedings of the British Machine Vision Conference 2015 · doi:10.5244/c.29.41 · Cited by 5,479
Private traits and attributes are predictable from digital records of human behavior
Kosinski, M., Stillwell, D., & Graepel, T. (2013) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1218772110 · Cited by 2,829
Self-citation
The Evolutionary Psychology of Facial Beauty
Rhodes, G. (2005) · Annual Review of Psychology · doi:10.1146/annurev.psych.57.102904.190208 · Cited by 1,821
Inferences of competence from faces predict election outcomes
Todorov, A., Mandisodza, A. N., Goren, A., & Hall, C. C. (2005) · Science · doi:10.1126/science.1110589 · Cited by 1,508
Social Cognition: Thinking Categorically about Others
Macrae, C. N., & Bodenhausen, G. V. (2000) · Annual Review of Psychology · doi:10.1146/annurev.psych.51.1.93 · Cited by 1,432
The evolution of human skin coloration
Jablonski, N. G., & Chaplin, G. (2000) · Journal of Human Evolution · doi:10.1006/jhev.2000.0403 · Cited by 1,212
Facebook as a research tool for the social sciences: Opportunities, challenges, ethical considerations, and practical guidelines.
Kosiński, M., Matz, S., Gosling, S. D., Popov, V., & Stillwell, D. J. (2015) · American Psychologist · doi:10.1037/a0039210 · Cited by 1,031
Self-citation

Show all 66 references

Sorted by citations

Items 701–778 of 778

Human Rights, Legal Personality, and Artificial Intelligence
Heine, K. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0030 · Cited by 1
Artificial Intelligence Screening and the Right to Asylum
Dhruv, S. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0022
Artificial Intelligence and Racial Discrimination
Koen, L., & Kgomotso, M. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0013
Putting Private Sector Responsibility in the Mix
Ebert, I. L., & Lisa, H. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0034
Artificial Intelligence, the Public Space, and the Right to Be Ignored
Pin, A. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0012
The Limits of AI Decision-Making
Gamper, F. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0032
Artificial Intelligence and the Right to an Effective Remedy
de Heer, S. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0020
Artificial Intelligence and Freedom of Assembly
Margaret, W. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0006
The Production of and Control over Data in the AI-Era
van der Sloot, B. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0011
Artificial Intelligence and the Right to Property
Tomada, L., & Xenidis, R. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0007
Artificial Intelligence and Disability Rights
Zarra, A., Favalli, S., & Matilde, C. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0017
Artificial Intelligence and LGBTQ+ Rights
Shahid, M. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0015 · Cited by 2
PICLAnony: Controlled Anonymization of Face Attributes Based on Parametric Imitation Comparison LearningPreprint
Liu, H., Ji, H., & Liang, J. (2024) · Research Square · doi:10.21203/rs.3.rs-4603713/v1
Machine Learning, Infrastructures, and Their Sociomaterial Possibilities
Pardo‐Guerra, J. P. (2024) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780197653609.013.22
Table of Domestic Cases
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0009
Table of International Cases
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0008
Table of Domestic Law
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0007
Table of International Law
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0006
Abbreviations
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0010
Copyright Page
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.002.0003
Ultimate Fantasy. Subwersyjne strategie queerowania Sztucznej Inteligencji w projektach transliterackich
Jędrusiak, D. (2024) · Przegląd Kulturoznawczy · doi:10.4467/20843860pk.24.004.20070
Board Diversity Appearance and Firm Performance: An Image-Based Deep Learning ApproachPreprint
Greger, L., Scholz, H., Stiller, A., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4947640
AI and Entrepreneurship: Facial Recognition Technology Detects Entrepreneurs, Outperforming Human ExpertsPreprint
Obschonka, M., Fisch, C. O., Fernando, T., & Fookes, C. (2024) · Research Square · doi:10.21203/rs.3.rs-4926308/v1 · Cited by 1
Algorithms
Franses, P. H. (2024) · Cambridge University Press eBooks · doi:10.1017/9781009428033.014
Collecting Digital Footprints in the Wild
Kosiński, M. (2024) · Cambridge University Press eBooks · doi:10.1017/9781009170123.016
Self-citation
Introduction
Kovač, M. (2024) · Springer eBooks · doi:10.1007/978-3-031-65514-2_1
Humans identification in various locations using spatial pyramid pooling algorithm with maximum mean average precision of locating objects over spatial pyramid pooling algorithm
Dinesh, C. S., & Thinakaran, K. (2025) · AIP conference proceedings · doi:10.1063/5.0258660
Hot Nudges on Hazy LandscapesPreprint
Devereaux, A. N. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5146881
Exploring the nexus between persuasive principles and regulatory focus: an online experiment
Ayaz, S., Shah, S. A., & Anwar, J. (2025) · Cogent Business & Management · doi:10.1080/23311975.2025.2501489
“…(Winter et al., 2021). negative effects of personalized ad messages can be avoided by focusing on less obvious personalization methods, for example, trait or motivational orientation-based personalization focusing on indistinguishable characteristics (hinds & Joinson, 2019; Wang & kosinski, 2018).”
Facial Recognition Technology and the Challenges to Democratic Systems: An Introduction
Menéndez, N., & Mobilio, G. (2025) · Law, governance and technology series · doi:10.1007/978-3-031-89794-8_1
Enhancing logical reasoning in language models: An investigation of the Capybara dataset
Guerrero, L. E. M., Ceballos, Y. F., & Rojas, L. D. T. (2025) · Contemporary Educational Technology · doi:10.30935/cedtech/16425
Character and Personality Perception and Social Trait Judgment After Facial Surgical and Medical Aesthetic Interventions
Atiyeh, B. S., Emsieh, S., Chrabieh, E. E., Hachem, T. E., Greige, G., Chalhoub, R. S., & Beaineh, P. T. (2025) · Plastic & Reconstructive Surgery Global Open · doi:10.1097/gox.0000000000006831
Diversifying Social Cognition Through Participatory Action Research Methods for Stimulus Development and Validation
Oswald, F., Paganini, G. A., & Winegar, A. (2025) · Social Cognition · doi:10.1521/soco.2025.43.3.276
EmoDetect: ML Based Facial Assessment of Anxiety & Depression
Gondhalekar, B., Mahato, S., Salian, S., et al. (2025) · doi:10.38124/ijisrt/24apr1755
Smart City(ies) and Urban Living: The Paradox of the Good Life
Roberts, A. (2025) · Sustainable development · doi:10.5772/intechopen.1011336
Privacy: Concept, Right, and Underlying Values
Powell, M. (2025) · Philosophy & Technology · doi:10.1007/s13347-025-00931-3
“For instance, genetic databases of small proportions of populations can make possible the identification of those who have not consented to their identity being linked to their genetic information (Wang & Kosinski, 2018).”
Human Rights, Datafication, and the Democratic Governance of AI
Land, M. K., & Wong, W. H. (2025) · Edward Elgar Publishing eBooks · doi:10.2139/ssrn.5371289
Generative AI and the Fundamental Limitations of US Privacy Law
Zeide, E. (2025) · Cambridge University Press eBooks · doi:10.1017/9781009492553.021
Enhancing Social Media Privacy With Apriori
Alan, A. (2025) · Security and Privacy · doi:10.1002/spy2.70096
“Simultaneously, concerns about the privacy implications of AI inference are underscored by Wang and Kosinski [15], who showed that AI can infer sensitive traits such as sexual orientation from facial images at a level surpassing human judgment.”
First Year CS Students Exploring Implications of Social Media Recommendation Algorithms
Apiola, M. (2025) · International Convention on Information and Communication Technology, Electronics and Microelectronics · doi:10.1109/mipro65660.2025.11131972
“Second, students’ perceptions of proﬁling and related data were rich, but more silent on inferences e.g. related to gender & sexual orientation [42], [43] or personality [44]–[47].”
Aplikasi Kencan Online Sebagai Platform Pembentukan Identitas Sosial: Studi Kasus di Kalangan Remaja dan Dewasa Muda Kota Surakarta
Mohammad, Z., & Mesra, R. (2025) · Jurnal Masyarakat Digital · doi:10.64924/ja7gvd31
A Taxonomy of AI
Williams, A. (2025) · Springer eBooks · doi:10.1007/978-3-032-02665-1_2
Inhabiting the Face: From Fake to Deepfake
Ravetto‐Biagioli, K. (2025) · Theory Culture & Society · doi:10.1177/02632764251370095
“…company Faception (see Figure 3) claims to be able to detect whether someone is a paedophile, terrorist or of high intelligence by scanning their face, while Stanford’s Business School researchers claim that AI is better at detecting one’s sexual orientation than humans (Wang and Kosinski, 2018).”
Big Social Interactions
Bainbridge, W. A. (2025) · Cambridge University Press eBooks · doi:10.1017/9781009343602.013
From pixels to privacy: A comprehensive review of visual privacy detection technologies and challenges
Li, Y., Li, X., Jiang, X., Chen, T., & Ji, B. (2026) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2026.104887
Why Do People Trust Physiognomic AI?
Gell, C., & Lewis, P. R. (2025) · IEEE Technology and Society Magazine · doi:10.1109/mts.2025.3627873
“…caution is often countered with a focus on mathematics to validate the model’s results as in Wu and Zhang [29], or by apparently exercising heavy caution before drawing inference from findings, and with the research being presented as strongly analytical or “objective” in nature [17], [26], [27].”
AI’s Impact on Marketing Success
Sampath, L., Neelam, H., & Vasundhara, S. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-0269-1_64
The Hidden Dangers of “Objective” Technologies: An Analysis Through Time
Sarraf, I. (2025) · IEEE History of Electrotechnology Conference (HISTELCON) · doi:10.1109/histelcon64051.2025.11286061
“Researchers at Stanford, Yilun Wang and Michal Kosinski, built an algorithm to detect a person’s sexuality by employing the “prenatal hormone theory” that hypothesizes that a person’s facial shape is influenced by the prenatal sex hormones they were exposed to [11].”
Additional file 2 of Current cancer driver variant predictors learn to recognize driver genes instead of functional variants
Raimondi, D., Passemiers, A., Fariselli, P., et al. (2021) · Figshare · doi:10.6084/m9.figshare.13572103
Self-Esteem Assessment Based on Self-Introduction: A Multimodal Approach to Personality Computing
Zang, X., & Yang, J. (2025) · Assessment · doi:10.1177/10731911251403907
Intelligence artificielle et marketing : défis éthiques et horizons de recherche
Goglin, C., Serhal, R., & Medic, D. (2026) · Question(s) de management · doi:10.3917/qdm.237.0031
Erkennungsaufgaben
Bläsius, K. H. (2026) · Springer eBooks · doi:10.1007/978-3-662-72526-9_2
Integrating LLM in Privacy-Sensitive Age Estimation: A Tongue-Based Biometric Framework
Ang, J. H., Li, A., Wang, J., & Zhang, B. (2025) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb65343.2025.11410665
“Several studies have indicated that the leaked facial images could be potentially used for the prediction of sexual orientation [31], personality [17], and political orientation [18].”
Category trouble: AI’s queer problem
Gáti, D. (2026) · AI & Society · doi:10.1007/s00146-025-02818-5
“…determine the type of animal on the image, to decide a person’s gender, or, as in a controversial example, to classify whether the depicted person is straight or gay (Wang and Kosinski 2018)—the model must necessarily have a finite and discrete list of options to choose from for classification.”
Artificial Intelligence in Video Game Advertising
Çeber, B., & Yanmaz, Ş. (2026) · IGI Global eBooks · doi:10.4018/979-8-3373-6309-7.ch005
The Right to Privacy as a Right to Obscurity: On Big Data, Inferences, and Attention
Hurshman, C. E. (2026) · Philosophy & Technology · doi:10.1007/s13347-026-01060-1
“Algorithms have been created that can purportedly infer such things as one’s sexual orientation based on a picture of one’s face (Wang & Kosinski, 2018) and future mental illness based on social media posts (Thorstad & Wolff, 2019), and generative AI may soon be used to create content tailored to…”
Droit et intelligence artificielle : essai pour la reconnaissance du droit computationnel
Papineau, C. (2019) · theses.fr (ABES) · doi:10.70675/400ff8d8zc4bcz445dz9236z6001ef28508d
La prédiction judiciaire par les algorithmes
Coletta, A. (2021) · theses.fr (ABES) · doi:10.70675/ae54bcccz0309z4c28zabc4zf71e24d718eb
La vie en rose : petite histoire d'une couleur aux prises avec le genre.
Bideaux, R. (2021) · doi:10.70675/1de48733z300fz4b85z91bdze6c2068aacd1
La philosophie computationnelle : pour une reprise en main épistémique du monde numérique
Etienne, H. (2022) · doi:10.70675/d2cbaf0dz1dcdz44daz9c9cz45bb5d3c64fd
La régulation juridique de l'intelligence artificielle au sein de la justice : une étude comparative entre le droit de l'Union européenne et les lois fédérales des Etats-Unis
Shahrjerdi, S. (2022) · theses.fr (ABES) · doi:10.70675/18755d07z72b0z486az9b48z697c3efb7df2
Modélisation de la beauté et étude de son influence sur les comportements sociaux d'un primate non-humain
Tieo, S. (2023) · theses.fr (ABES) · doi:10.70675/165819bazd172z4b4czad51z1ff5e85c73a0
Enhancing Access to Health Information for Adolescent Girls in Bangladesh Through a Culturally Responsive AI Chatbot
Ahamed, S., Zaman, M., Mashud, M. A. A., Khan, N., Sarker, F., & Mamun, K. A. (2025) · Gender and Technology Conference (GTC) · doi:10.1109/gtc64325.2025.11478191
“Risks linked to high-risk AI applications, including deepfakes and algorithmic biases, could reinforce existing biases or lead to automated job losses [3], [38], [47].”
Implications of Artificial Intelligence for Workplace Monitoring and Privacy
Blachly, B. D., Kyeremateng, T. K., Shafiei, M., & Ravid, D. M. (2026) · doi:10.1093/9780197807309.003.0017
Personality Assessment and Artificial Intelligence
Sherman, R. A. (2026) · Oxford University Press eBooks · doi:10.1093/9780197807309.003.0009
Leveraging AI and Blockchain technology in 6G Network
Pattanayak, A., Mitra, A. A., Anand, A. G. M., et al. (2026) · doi:10.1109/iccece69715.2026.11493080
“…to gather or buy data, including sensitive personal information, potentially infringing on individuals' privacy rights. b) Ethical concerns related to human vulnerabilities involve the misuse of AI: Technologies like facial recognition and image generation can surpass human capabilities [22].”
From Identification to Recognition: A Lesson for Biometric Technology From the History of Personal Identity
Mordini, E., & Wayman, J. L. (2026) · IET Biometrics · doi:10.1049/bme2/3937059
Visual artificial intelligence as a methodological frontier in entrepreneurship research: a roadmap and empirical demonstration
Obschonka, M., Fisch, C. O., Beliaeva, T., Jabeur, S. B., Fernando, T., & Fookes, C. (2026) · Small Business Economics · doi:10.1007/s11187-026-01221-8
“Beyond entrepreneurship, numerous impactful studies show that AI can infer intimate personal attributes with accuracy from facial images (e.g., Kosinski et al., 2024; Wang & Kosinski, 2018).”
Diversity in the Digital Age
Melson-Silimon, A. T., Harmata, R., LeFevre-Levy, R., Behrend, T. S., & Carter, N. T. (2022) · Emerald Group Publishing Limited eBooks · doi:10.1108/978-1-64802-826-720251007
Gender-Aware or Gender-Blind Fashion Recommendation Systems: A State-of-the-Art Review and Future Agenda
Santos, R., Landim, A. R. D. B., Lima, D., de Araujo, R., Moura, A., Costa, E., Wanick, V., & Bazaki, E. (2026) · Springer eBooks · doi:10.1007/978-3-032-15780-5_8
Between Efficiency and Fairness
Ertürk, M., & Ayhan, E. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-2303-7.ch005
Improving Student Behavior within Educational Units Based on Profiling Similarity
Gryparis, P. V., Ntalianis, K., & Mastorakis, N. E. (2026) · International Journal of Education and Information Technologies · doi:10.46300/9109.2026.20.12
A Queer Critique of Surveillance Capitalism
Ellis, J. R. (2026) · doi:10.1093/9780198960256.003.0040
Ethical Concerns in the Field of Artificial Intelligence and Data Analytics in Insurance Industry
Pant, P., Mishra, K. K., Srinivas, B., et al. (2026) · Management for professionals · doi:10.1007/978-981-95-7809-2_11
Technology
Weigel, C. (2026) · Springer eBooks · doi:10.1007/978-3-642-27828-0_674-1
Image and Video Data Analysis
Fan, L. (2025) · Emerald Group Publishing Limited eBooks · doi:10.1561/978-1-63828-477-220251015
Ethicality Analysis Framework for Active OSINT Research
Byers, J., Driscoll, E., Lu, A., & Michaels, A. J. (2026) · Electronics · doi:10.3390/electronics15184164
Governing the Financialization of Human Intent in the Shift from Privacy to Cognitive Sovereignty
Ruaro, R. L., & de Lucena, P. A. C. (2026) · Beijing Law Review · doi:10.4236/blr.2026.173051

Authors: Wang, Y., & Kosinski, M.
Affiliations: Stanford University (US)
ORCID: Kosinski, Michal
Venue: Journal of Personality and Social Psychology
Publisher: American Psychological Association (APA)
Published: 2018-02
Volume: 114
Issue: 2
Pages: 246-257
ISSN: 0022-3514 (ISSN-L)
1939-1315
DOI: 10.1037/pspa0000098
OpenAlex: W2781265533
Cited by: 766
References: 66
Citations by year: 2026: 29 · 2025: 57 · 2024: 57 · 2023: 146 · 2022: 98 · 2021: 92 · 2020: 155 · 2019: 75 · 2018: 51
FWCI: 55.54
Type: journal article
Language: en
Open access: green
doi.org
Licence: cc-by
Locations: Journal of Personality and Social Psychology (publishedVersion)
PubMed (publishedVersion)
PsyArXiv (OSF Preprints) (submittedVersion)
OSF Preprints (OSF Preprints) (submittedVersion)
PsyArXiv (OSF Preprints) (submittedVersion, cc-by)
Topics: Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
Social Sciences › Psychology › Experimental and Cognitive Psychology › Evolutionary Psychology and Human Behavior
Keywords: face images, deep neural network, gay men, gay women, facial morphology, privacy, computer vision, logistic regression
MeSH terms: Facial Recognition, Deep Learning, Adult, Face, Female, Humans, Male, Social Perception, Sexuality

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Abstract

Overview

Integrity signals

Verification

Claims the paper makes

Claims other papers make about this one

Citations

Cites

Cited by

Related

Metadata

Record

About these checks