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

Flagged first, then by citations

Show all 66 references

13 of the 66 references couldn’t be matched to a known paper.

Sorted by citations

Items 301–400 of 778

Examining Potential Harms of Large Language Models (LLMs) in Africa
Baguma, R., Namuwaya, H., Nabende, J. N., & Rashid, Q. M. (2024) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-031-56396-6_1 · Cited by 6
Language Agents and Malevolent Design
Yum, I. (2024) · Philosophy & Technology · doi:10.1007/s13347-024-00794-0 · Cited by 7
Proposal of Quality in Use in Software Quality
Fukuzumi, S., Hirasawa, N., Wada, N., et al. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-49059-1_31 · Cited by 5
Constructing AI: Examining how AI is shaped by data, models and people
Ingram, K. (2021) · The International Review of Information Ethics · doi:10.29173/irie415 · Cited by 5
“The authors of the study say they conducted this research in order to expose discrimination in existing facial recognition algorithms (Wang & Kosinski, 2018).”
Method
Human vs. Artificial Intelligence
Baeza-Yates, R. A., & Villoslada, P. (2022) · IEEE International Conference on Cognitive Machine Intelligence (CogMI) · doi:10.1109/cogmi56440.2022.00016 · Cited by 7
“Here we can mention as examples inferring personality traits from faces such as criminal propensity [72], [23], sexual [70] or political [35] orientation.”
Method
Assessment of Personal Values for Data-Driven Human Resource Management
Kimura, T. (2023) · Data Science Journal · doi:10.5334/dsj-2023-022 · Cited by 11
“Related to the above issue are controversies regarding the work by Wang and Kosinski (2018), who developed a deep neural network model to detect people’s sexual orientation based on facial images as input data.”
Background
A Unified Framework of Ethical Principles for AI
Floridi, L. (2023) · Oxford University Press eBooks · doi:10.1093/oso/9780198883098.003.0004 · Cited by 9
Discovering the hidden personality of lambs: Harnessing the power of Deep Convolutional Neural Networks (DCNNs) to predict temperament from facial images
Çakmakçı, C., Magalhães, D. R., Pacor, V. R., de Almeida, D. H. S., Çakmakçı, Y., Dalga, S., Szabó, C., María, G. A., & Titto, C. G. (2023) · Applied Animal Behaviour Science · doi:10.1016/j.applanim.2023.106060 · Cited by 6
Framing the Landscape of Technological Enhancements: Artificial Intelligence, Gender Issues, and Ethical Dilemmas
Dutta, S. (2023) · Signals and communication technology · doi:10.1007/978-3-031-45237-6_10 · Cited by 5
The Ethics of National Security Intelligence Institutions
Henschke, A., Miller, S., Alexandra, A., Walsh, P. F., & Bradbury, R. (2024) · Routledge eBooks · doi:10.4324/9781003106449 · Cited by 5
Artificial Intelligence and Human Rights at Work
Atkinson, J., & Collins, P. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0025 · Cited by 6
Distant supervision and knowledge transfer for domain-oriented text classification in online social networks
Khodorchenko, M. (2019) · Procedia Computer Science · doi:10.1016/j.procs.2019.08.192 · Cited by 5
La libertad de expresión en jaque, el panóptico del Siglo XXI. Big Data como amenaza para la democracia: a propósito del caso Cambridge Analytica
Mejía, A. F. (2020) · UNIVERSITAS Revista de Filosofía Derecho y Política · doi:10.20318/universitas.2020.5512 · Cited by 5
Facial Recognition: A cross-national Survey on Public Acceptance, Privacy, and Discrimination
Steinacker, L., Meckel, M., Kostka, G., & Borth, D. (2020) · Alexandria (UniSG) (University of St.Gallen) · doi:10.48550/arxiv.2008.07275 · Cited by 13
“Controversial research efforts to demonstrate a correlation between facial features and criminality (Wu & Zhang, 2016) or sexual orientation (Wang & Kosinski, 2018) signify the possibilities for misappropriating the technology, such as applying FRT for tracking ethnic minorities (Leibold, 2020).”
Background
Predicting Twitter Users' Political Orientation: An Application to the Italian Political Scenario
Cardaioli, M., Kaliyar, P., Capuozzo, P., et al. (2020) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam49781.2020.9381470 · Cited by 5
“For instance, recent evidence shows that data gathered from OSNs can be successfully employed for predicting users’ personal information like personality traits [5], sexual orientation [6] and ethnicity [7].”
Background
Using Facial Recognition Tools for Health Assessment
Boczar, D., Ávila, F. R., Carter, R. E., Moore, P. A., Giardi, D., Guliyeva, G., Bruce, C. J., McLeod, C. J., & Forte, A. J. (2021) · Plastic and Aesthetic Nursing · doi:10.1097/psn.0000000000000357 · Cited by 5
“Controversial tools have also been developed that can use facial morphology to distinguish an individual’s sexual orientation (Wang & Kosinski, 2018).”
Background
To What Extent are Prenatal Androgens Involved in the Development of Male Homosexuality in Humans?
Tasos, E. (2021) · Journal of Homosexuality · doi:10.1080/00918369.2021.1933792 · Cited by 4
“…networks (DNNs) have been able to accurately determine sexual orientation from facial structure than human judges based on facial structure alone (Wang & Kosinski, 2018), though confounders such as lifestyle and presentation may have a great effect on such experiments, while there are many…”
Background
Facial structure and perception of sexual orientation: Research with face models based on photographs of real people
González, J., & Sos‐Peña, R. (2021) · International Journal of Psychology · doi:10.1002/ijop.12811 · Cited by 5
Facial analysis: automated surveillance and the attempt to quantify emotion
Black, D. (2021) · Information Communication & Society · doi:10.1080/1369118x.2021.2011948 · Cited by 5
“(Wang & Kosinski, 2018) While critics pointed out that these posed and self-curated photographs contained information about far more than facial morphology, and the accuracy of the system was not terribly high in any event, the researchers ’ appeal to a biological essence highlights a broader…”
Background
Simulation vs. Understanding: A Tension, in Quantum Chemistry and Beyond. Part B. The March of Simulation, for Better or Worse
Hoffmann, R., & Malrieu, J. (2019) · Angewandte Chemie · doi:10.1002/ange.201910283 · Cited by 5
Queer Reflections on AI
Klipphahn-Karge, M., Koster, A.-K., & Bruss, S. M. D. S. (2023) · Routledge eBooks · doi:10.4324/9781003357957 · Cited by 7
On Quantifying and Understanding the Role of Ethics in AI Research: A Historical Account of Flagship Conferences and Journals
Prates, M. O. R., Avelar, P. H. C., & Lamb, L. (2018) · Global Conference on Artificial Intelligence · doi:10.29007/74gj · Cited by 12
“Researchers have also revived debates concerning the controversial field of physiognomy, with many people asking whether artificial intelligence even should try and classify people’s sexual orientation according to their facial features [32].”
Background
Streaking
McGlotten, S. (2019) · TDR/The Drama Review · doi:10.1162/dram_a_00881 · Cited by 5
Server, server in the cloud. Who is the fairest in the crowd?Preprint
Böhlen, M., Chandola, V., & Salunkhe, A. S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.08801 · Cited by 5
The AgeGuess database: an open online resource on chronological and perceived ages of people aged 3-100Preprint
Jones, J., Nash, U. W., VIEILLEFONT, J., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.10063 · Cited by 5
“controversy, being studied by computer scientists [21].”
Examining the Presence of Gender Bias in Customer Reviews Using Word EmbeddingPreprint
Mishra, A., Mishra, H., & Rathee, S. (2019) · arXiv (Cornell University) · doi:10.13140/rg.2.2.15426.02240 · Cited by 5
“Reviews are considered the voice of customers by firms (Griffin and Hauser 1993) because they inform firms about consumer opinions, sentiments, needs, (Humphreys and Wang 2017) and can be used as a voluntary source of market research data (Lee and Bradlow 2011).”
Background
Statistical Significance RevisitedPreprint
Tormählen, M., Klinkova, G., & Grabinski, M. (2021) · Preprints.org · doi:10.20944/preprints202103.0398.v1 · Cited by 5
“The authors of article [9] conclude that humans recognize homosexuality in men with 61% certainty only by looking at their faces.”
Background
#BigTech @Minors: social media algorithms have actionable knowledge about child users and at-risk teens.Preprint
Hilbert, M., Cingel, D. P., Zhang, J., Vigil, S. L., Shawcroft, J. E., Xue, H., Thakur, A., & Shafiq, Z. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4674573 · Cited by 7
ToonerGAN: Reinforcing GANs for Obfuscating Automated Facial Indexing
Thakral, K., Prasad, S., Aswani, S., et al. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01034 · Cited by 4
“Face De-identification/Anonymization The face de-identification/anonymization techniques range from attribute anonymization [3, 28] to identity anonymization [21, 34].”
Background
A New Look at Person Memory
Horgan, T. G. (2020) · Springer eBooks · doi:10.1007/978-3-030-34964-6_7 · Cited by 4
AI’s Fast and Furtive Spread by Infusion into Technologies That Are Already in Use—A Critical Assessment
Engström, E. (2022) · Social and cultural studies of robots and AI · doi:10.1007/978-3-030-88615-8_4 · Cited by 4
The Banality of (Automated) Evil: Critical Reflections on the Concept of Forbidden Knowledge in Machine Learning Research
Julián, R. M. S., & Bueso, D. (2022) · Recerca Revista de pensament i anàlisi · doi:10.6035/recerca.6147 · Cited by 5
Athena Unbound
Baldwin, P. A. (2023) · The MIT Press eBooks · doi:10.7551/mitpress/14887.001.0001 · Cited by 5
Physiognomy in the Age of AI
y Arcas, B. A., Mitchell, M., & Todorov, A. T. (2023) · doi:10.1093/oso/9780192889898.003.0013 · Cited by 4
A megastudy on the predictability of personal information from facial images: Disentangling demographic and non-demographic signals
Tkachenko, Y., & Jedidi, K. (2023) · Scientific Reports · doi:10.1038/s41598-023-42054-9 · Cited by 4
Can machine learning make naturalism about health truly naturalistic? A reflection on a data-driven concept of health
Guersenzvaig, A. (2023) · Ethics and Information Technology · doi:10.1007/s10676-023-09734-6 · Cited by 2
“The first example: two Stanford researchers (Wang & Kosinski, 2018) used machine learning (deep neural networks performing logistic regression) to analyze more than 35,000 facial images to classify these according to sexual orientation; i.e., to distinguish between gay and heterosexual persons.”
Critical Review of Stack Ensemble Classifier for the Prediction of Young Adults’ Voting Patterns Based on Parents’ Political Affiliations
Elo, G., Ghansah, B., & Kwaa-Aidoo, E. K. (2024) · Informing Science The International Journal of an Emerging Transdiscipline · doi:10.28945/5254 · Cited by 4
Predicting personality or prejudice? Facial inference in the age of artificial intelligence
Madan, S., & Park, G. (2024) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2024.101815 · Cited by 4
The five generations of facial recognition usage and the Australian privacy law
Conde, J., & Svantesson, D. J. B. (2024) · International Data Privacy Law · doi:10.1093/idpl/ipae007 · Cited by 4
Artificial Intelligence Human Rights Impact Assessment
Ortalda, A., & Hert, P. D. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0035 · Cited by 3
Queering AI as a Speculative Practice: An Analysis of the Artistic Explorations of New Paradigms for Developing Inclusive AI
Jędrusiak, D. (2024) · ACM Conference on Hypertext and Social Media (HT) · doi:10.1145/3648188.3675157 · Cited by 4
Leveraging Artificial Intelligence in Healthcare
Mustafa, S. M. N. (2024) · Advances in healthcare information systems and administration book series · doi:10.4018/979-8-3693-7051-3.ch012 · Cited by 5
Technologies of the Future
Kabir, M. N. (2019) · Palgrave Macmillan US eBooks · doi:10.1057/978-1-137-34809-8_4 · Cited by 4
Artificial Intelligence Legal Policy
Дремлюга, Р., & Natalia, P. (2019) · International Conference on Software and Computer Applications (ICSCA) · doi:10.1145/3316615.3316627 · Cited by 4
Creepy Technologies and the Privacy Issues of Invasive Technologies
McWhorter, R. R., & Bennett, E. E. (2020) · Advances in business strategy and competitive advantage book series · doi:10.4018/978-1-7998-2914-0.ch010 · Cited by 4
Fair Facial Attribute Classification via Causal Graph-Based Attribute Translation
Kang, S., Kim, G., & Yoo, C. D. (2022) · Sensors · doi:10.3390/s22145271 · Cited by 5
I know it when I see it: the heteronormativity of Google's SafeSearch
Monea, A. (2022) · Porn Studies · doi:10.1080/23268743.2022.2086163 · Cited by 5
The End of Vagueness: Technological Epistemicism, Surveillance Capitalism, and Explainable Artificial Intelligence
Kerr, A. D., & Scharp, K. A. (2022) · Minds and Machines · doi:10.1007/s11023-022-09609-7 · Cited by 4
Consent, Background Justice and Patterned Privacy Principles
Powell, M. (2023) · Political Studies · doi:10.1177/00323217231167074 · Cited by 5
“8 In the Wang and Kosinski (2018) case, we can see that numerous individuals consenting to the use of the data they uploaded to their dating profiles by third parties has enabled the creation of a technology that can determine similar information about other individuals who have not consented.”
Present
Floridi, L. (2023) · Oxford University Press eBooks · doi:10.1093/oso/9780198883098.003.0002 · Cited by 6
Artificial Intelligence and the Right to a Healthy Environment
Quintavalla, A. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0028 · Cited by 4
Le deep learning au service de la prédiction de l’orientation sexuelle dans l’espace public
Baya-Laffite, N., Beaude, B., & Garrigues, J. (2018) · Réseaux · doi:10.3917/res.211.0137 · Cited by 4
The Uncanny Return of Physiognomy.
Bendel, O. (2018) · National Conference on Artificial Intelligence · Cited by 4
A Requirement Analysis for Privacy Preserving Biometrics in View of Universal Human Rights and Data Protection Regulation
Whiskerd, N., Dittmann, J., & Vielhauer, C. (2018) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553045 · Cited by 4
“Sexual Orientation can be divulged in reactions of pupil dilation in the eye, and exploratory works exist in identifying orientation from face images [17].”
Background
Privacy in Research Ethics
Macnish, K. (2019) · Springer eBooks · doi:10.1007/978-3-030-16759-2_9 · Cited by 4
Un appel à un cadre éthique lors de l’utilisation des données des médias sociaux pour des applications d’intelligence artificielle dans la recherche en santé publique
Gilbert, J.-P., Ng, V., Niu, J., & Rees, E. E. (2020) · Relevé des maladies transmissibles au Canada · doi:10.14745/ccdr.v46i06a03f · Cited by 4
“(22) ont déclaré être en mesure de déterminer l’orientation sexuelle des personnes en traitant des photos de personnes tirées d’un site de rencontres.”
Background
Ethische Aspekte der Künstlichen Intelligenz
Buxmann, P., & Schmidt, H. (2021) · Künstliche Intelligenz · doi:10.1007/978-3-662-61794-6_13 · Cited by 4
Psychological targeting in the age of Big Data
Appel, R. E., & Matz, S. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-819200-9.00015-6 · Cited by 4
Some Examples of Discrimination
Arthur Charpentier, A. (2024) · Springer Actuarial · doi:10.1007/978-3-031-49783-4_6 · Cited by 4
State of the Art in Fair ML: From Moral Philosophy and Legislation to Fair ClassifiersPreprint
Baumann, E., & Rumberger, J. L. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.09539 · Cited by 4
“– Kosinski and Wang [55] apply computer vision algorithms on facial images from an online dating platform to predict the sexual orientation of individuals with higher accuracy than humans.”
Background
What makes a market transaction morally repugnant?Preprint
Leuker, C., Samartzidis, L., & Hertwig, R. (2020) · doi:10.31234/osf.io/dgz4s · Cited by 4
Computational PsychologyPreprint
Kosiński, M. (2023) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/tkhg4 · Cited by 3
Self-citation
Personality Assessment
Sherman, R. A., & Hogan, R. T. (2023) · doi:10.1093/oso/9780197611050.003.0006 · Cited by 4
AI Life Cycle and Human Rights
Martina, Š., Luboš, K., & Drchal, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0002 · Cited by 4
The Dark Side of Gamification
Kapoor, U. (2025) · IGI Global eBooks · doi:10.4018/979-8-3693-8886-0.ch009 · Cited by 3
Invisible-Face: Rethinking Facial Attribute Privacy in Social Media Photo Sharing
Yang, Y., Li, C., Jiang, Y., Li, J., Zhang, X., Wang, Z., Ji, S., & Chen, W. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3579592 · Cited by 3
A new epoch of face analytics: technological evolution through ethical and legal challenges
Conde, S., Speelman, C. P., & Johnstone, M. (2025) · AI and Ethics · doi:10.1007/s43681-025-00678-9 · Cited by 6
“Recent research studies exact that AI is now capable of using data-mined facial features to predict such features as gender [3], rule consciousness [4 ], sexual orientation [ 5], mental health disorder [6], neuroticism [7], and political orientation [8].”
#BigTech @Minors: social media algorithms have actionable knowledge about child users and at-risk teens
Hilbert, M., Cingel, D. P., Zhang, J., Vigil, S. L., Shawcroft, J. E., Xue, H., Thakur, A., & Shafiq, Z. (2025) · Telematics and Informatics · doi:10.1016/j.tele.2025.102341 · Cited by 3
Avrupa Birliği Yapay Zekâ Tüzüğü'ne Göre Yasaklanmış Yapay Zekâ Uygulamaları
Akbulut, B., & AÇIL, M. (2026) · Ankara Hacı Bayram Veli Üniversitesi Hukuk Fakültesi Dergisi · doi:10.34246/ahbvuhfd.1704398 · Cited by 3
Big Data Value Chain: Making Sense of the Challenges
Wirén, M., Mäntymäki, M., & Islam, A. K. M. N. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-29374-1_11 · Cited by 4
Capital City Cruising: Surveillance, Pleasure, and Discursive Practices of Queer Communities in Ottawa
Kenyon, M. (2024) · Surveillance & Society · doi:10.24908/ss.v22i1.16521 · Cited by 3
“For example, attempts have been made to train deep neural networks to identify the faces of queer individuals (Wang and Kosinski 2018), which would allow for sexual identification from sight alone.”
Background
Artificial Intelligence and Data Analytics
Laukyte, M. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0019 · Cited by 5
Smile variation leaks personality and increases the accuracy of interpersonal judgments
Witkower, Z., Tian, L., Tracy, J. L., & Rule, N. O. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae343 · Cited by 3
“The current research also builds on past work demonstrating that neutral faces leak personality and dispositions (e.g. 2, 57) and permit modestly accurate judgments of social traits (3, 58).”
Result
Catch me if you can: Using machine learning and behavioral interventions to reduce unethical behavior
Hauser, O., Greene, M. A., & DeCelles, K. A. (2025) · Behavioural Public Policy · doi:10.1017/bpp.2024.50 · Cited by 3
“It is important to acknowledge that both the use of machine learning and nudges could potentially be used in nefarious ways, such as identifying individuals for the purposes of discrimination (Wang and Kosinski, 2018).”
Improving Workforce Management with Machine Learning: A Novel Approach to Employee Classification
M, V. T. R. P. K., Nagalakshmi, V., Sriram, B., Wasib, S. A., Sadiq, S., & Samhitha, M. D. (2025) · International Conference on Augmented Intelligence and Sustainable Systems (ICAISS) · doi:10.1109/icaiss61471.2025.11042033 · Cited by 3
“Wang and Kosinski [19] highlight AI’s advanced classification capabilities but raise concerns regarding bias and fairness in AI-driven hiring and decision-making.”
Residual permutation tests for feature importance in machine learning
Huang, P. (2025) · British Journal of Mathematical and Statistical Psychology · doi:10.1111/bmsp.70009 · Cited by 8
Datenschatten und die Gravitation fast richtiger Vorhersagen
Gransche, B. (2020) · Anthropologie, Technikphilosophie, Gesellschaft · doi:10.1007/978-3-658-27149-7_5 · Cited by 3
Inference of a universal social scale and segregation measures using social connectivity kernels
Hoffmann, T., & Jones, N. S. (2020) · Journal of The Royal Society Interface · doi:10.1098/rsif.2020.0638 · Cited by 4
“Online social networks and the widespread use of mobile phones provide us with detailed information about connections between individuals [40], and seemingly private traits such as socio-economic status [41,42], sexual orientation [43], age, gender and political ideology can be inferred [44].”
Background
Are We Slaves to Our Genes?
Alexander, D. R. (2020) · Perspectives on Science and Christian Faith · doi:10.1017/9781108566520 · Cited by 3
Machine Learning Algorithms to Detect Sex in Myocardial Perfusion Imaging
Filho, E. M. D. S., Fernandes, F. D. A., Portela, M. G. R., Newlands, P. H., de Carvalho, L. N. D., dos Santos, T. F., dos Santos, A. A. S. M. D., Mesquita, E. T., Seixas, F. L., Mesquita, C. T., et al. (2021) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2021.741679 · Cited by 3
“From the analysis of 5 images of each person, the algorithm obtained an accuracy of 91% for distinguishing heterosexual men from gay men and 83% for determining heterosexual women from lesbians (47).”
Method
Using Facial Recognition Tools for Health Assessment
Boczar, D., Avila, F. R., Carter, R. E., Moore, P. A., Giardi, D., Guliyeva, G., Bruce, C. J., McLeod, C. J., & Forte, A. J. (2021) · Plastic and Aesthetic Nursing · doi:10.1097/psn.0000000000000410 · Cited by 4
From Outside In: Profiling, Persuasion and Political Opinion in the Age of Big Data
Dhawan, S., & Hegelich, S. (2022) · Studies in neuroscience, psychology and behavioral economics · doi:10.1007/978-3-030-98546-2_10 · Cited by 6
Artifacting Identity. How Grillz, Ball Gags and Gas Masks Expand the Face
Voto, C., & Soro, E. (2022) · Topoi · doi:10.1007/s11245-022-09819-9 · Cited by 3
Vloga umetne inteligence v izobraževanju in za izobraževanje
Pesek, I., & Krašna, M. (2022) · doi:10.18690/um.ff.11.2022.13 · Cited by 3
Unconscious Other’s Impression Changer: A Method to Manipulate Cognitive Biases That Subtly Change Others’ Impressions Positively/Negatively by Making AI Bias in Emotion Estimation AI
Futami, K., Yanase, S., Murao, K., & Terada, T. (2022) · Sensors · doi:10.3390/s22249961 · Cited by 4
“, an AI that estimates an individual’s sexual orientation based on facial features [23,52].”
Background
Automatic analysis of face images for college degree verification
Zulkashev, R., & Polyak, M. D. (2023) · doi:10.1109/itnt57377.2023.10139216 · Cited by 3
“Later Kosinski has shown that machine learning algorithms are capable of extracting sensitive information such as sexual [5] and political orientation [6] from facial images.”
Background
Finding Traces of Music Education on Facial Images
Senichenkova, Y. O., & Polyak, M. D. (2023) · doi:10.1109/scm58628.2023.10159064 · Cited by 3
“There exist technologies to determine sexual orientation [4] and political views [5] by analyzing a person’s face only.”
Background
Conclusion
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0037 · Cited by 3
Artificial Intelligence and Women’s Rights
Gagliardi, M. P. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0016 · Cited by 3
Introduction
Quintavalla, A., & Temperman, J. (2023) · Oxford University Press eBooks · doi:10.1093/law/9780192882486.003.0001 · Cited by 4
The effect of sexual orientation on voice acoustic properties
Holmes, L., Rieger, G., & Paulmann, S. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1412372 · Cited by 3
Nonverbal Gendered Cues as Accurate Signals of Sexual Orientation
Henry, P. J. (2024) · Sex Roles · doi:10.1007/s11199-024-01527-5 · Cited by 4
“…aspects of the faces, including focusing only on the eyes, the hair, or the mouth (Rule et al., 2008), a point corroborated by later work on computer judgments of sexual orientation (described in detail below) that also found these areas as cues to sexual orientation (Wang & Kosinski, 2018).”
Influential
The Improvement of Digital Literacy to Secure Data and Privacy in the Digital Age
Alfiani, F. R. N., & Sara, R. (2024) · Interdiciplinary Journal and Hummanity (INJURITY) · doi:10.58631/injurity.v3i12.1374 · Cited by 4
“However, challenges such as bias, ethics, privacy and work disruption need to be addressed (Wang & Kosinski, 2018).”
Singularity und weitere kritische Debatten über Künstliche Intelligenz
Buxmann, P., & Schmidt, H. (2018) · Künstliche Intelligenz · doi:10.1007/978-3-662-57568-0_12 · Cited by 3
Digital Sensory Phenotyping for Psychiatric Disorders
Dai, R., Yu, C., Xia, C., Zhou, J., Liu, C., & Chen, C. (2020) · Journal of Psychiatry and Brain Science · doi:10.20900/jpbs.20200015 · Cited by 3
“discovered that deep neural network analysis of facial images may threaten the privacy and safety of gay men and women [153].”
Background
There and back again: how target market determination obligations for financial products may incentivise consumer data profiling
Bednarz, Z. (2022) · International Review of Law Computers & Technology · doi:10.1080/13600869.2022.2060469 · Cited by 3
Unit Roots in Macroeconomic Time Series: A Comparison of Classical, Bayesian and Machine Learning Approaches
Ahmad, Y. S., Check, A., & Lo, M. C. (2023) · Computational Economics · doi:10.1007/s10614-023-10397-0 · Cited by 3
“…congestion (Asencio-Cortés et al., 2016); health scientists predicting breast cancer (Macias-Garcia et al., 2017); improving upon facial recognition software (Goodfellow et al., 2015), even to the extent of psychologists being able to detect human sexuality from photos (Wang & Kosinski, 2018).”
Foundations of Data Protection Law
Ausloos, J. (2020) · doi:10.1093/oso/9780198847977.003.0002 · Cited by 3
Attacking Machine Learning Models for Social Good
Belavadi, V., Zhou, Y., Kantarcıoğlu, M., & Thuriasingham, B. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-64793-3_25 · Cited by 3
“Wang & Kosinski [30] applied Deep Learning to test whether the sexual orientation of a person can be accurately predicted better than a human predicting it.”
Background
Psicometría, perfiles y sesgos
Gómez, J. C. S. (2021) · InMediaciones de la Comunicación · doi:10.18861/ic.2021.16.2.3156 · Cited by 3

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