{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T01:14:10Z","timestamp":1790730850268,"version":"4.1.0"},"reference-count":22,"publisher":"American Medical Association (AMA)","issue":"1","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JAMA Surg"],"published-print":{"date-parts":[[2024,1,1]]},"abstract":"<jats:sec id=\"ab-soi230083-4\"><jats:title>Importance<\/jats:title><jats:p>The progression of artificial intelligence (AI) text-to-image generators raises concerns of perpetuating societal biases, including profession-based stereotypes.<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-5\"><jats:title>Objective<\/jats:title><jats:p>To gauge the demographic accuracy of surgeon representation by 3 prominent AI text-to-image models compared to real-world attending surgeons and trainees.<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-6\"><jats:title>Design, Setting, and Participants<\/jats:title><jats:p>The study used a cross-sectional design, assessing the latest release of 3 leading publicly available AI text-to-image generators. Seven independent reviewers categorized AI-produced images. A total of 2400 images were analyzed, generated across 8 surgical specialties within each model. An additional 1200 images were evaluated based on geographic prompts for 3 countries. The study was conducted in May 2023. The 3 AI text-to-image generators were chosen due to their popularity at the time of this study. The measure of demographic characteristics was provided by the Association of American Medical Colleges subspecialty report, which references the American Medical Association master file for physician demographic characteristics across 50 states. Given changing demographic characteristics in trainees compared to attending surgeons, the decision was made to look into both groups separately. Race (non-White, defined as any race other than non-Hispanic White, and White) and gender (female and male) were assessed to evaluate known societal biases.<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-7\"><jats:title>Exposures<\/jats:title><jats:p>Images were generated using a prompt template, \u201ca photo of the face of a [blank]\u201d, with the blank replaced by a surgical specialty. Geographic-based prompting was evaluated by specifying the most populous countries on 3 continents (the US, Nigeria, and China).<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-8\"><jats:title>Main Outcomes and Measures<\/jats:title><jats:p>The study compared representation of female and non-White surgeons in each model with real demographic data using \u03c7<jats:sup>2<\/jats:sup>, Fisher exact, and proportion tests.<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-9\"><jats:title>Results<\/jats:title><jats:p>There was a significantly higher mean representation of female (35.8% vs 14.7%; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) and non-White (37.4% vs 22.8%; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) surgeons among trainees than attending surgeons. DALL-E 2 reflected attending surgeons\u2019 true demographic data for female surgeons (15.9% vs 14.7%; <jats:italic>P<\/jats:italic>\u2009=\u2009.39) and non-White surgeons (22.6% vs 22.8%; <jats:italic>P<\/jats:italic>\u2009=\u2009.92) but underestimated trainees\u2019 representation for both female (15.9% vs 35.8%; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) and non-White (22.6% vs 37.4%; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) surgeons. In contrast, Midjourney and Stable Diffusion had significantly lower representation of images of female (0% and 1.8%, respectively; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) and non-White (0.5% and 0.6%, respectively; <jats:italic>P<\/jats:italic>\u2009&amp;amp;lt;\u2009.001) surgeons than DALL-E 2 or true demographic data. Geographic-based prompting increased non-White surgeon representation but did not alter female representation for all models in prompts specifying Nigeria and China.<\/jats:p><\/jats:sec><jats:sec id=\"ab-soi230083-10\"><jats:title>Conclusion and Relevance<\/jats:title><jats:p>In this study, 2 leading publicly available text-to-image generators amplified societal biases, depicting over 98% surgeons as White and male. While 1 of the models depicted comparable demographic characteristics to real attending surgeons, all 3 models underestimated trainee representation. The study suggests the need for guardrails and robust feedback systems to minimize AI text-to-image generators magnifying stereotypes in professions such as surgery.<\/jats:p><\/jats:sec>","DOI":"10.1001\/jamasurg.2023.5695","type":"journal-article","created":{"date-parts":[[2023,11,15]],"date-time":"2023-11-15T16:01:39Z","timestamp":1700064099000},"page":"87","source":"Crossref","is-referenced-by-count":90,"title":["Demographic Representation in 3 Leading Artificial Intelligence Text-to-Image Generators"],"prefix":"10.1001","volume":"159","author":[{"given":"Rohaid","family":"Ali","sequence":"first","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Oliver Y.","family":"Tang","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"},{"name":"Department of Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ian D.","family":"Connolly","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Massachusetts General Hospital, Boston"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hael F.","family":"Abdulrazeq","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fatima N.","family":"Mirza","sequence":"additional","affiliation":[{"name":"Department of Dermatology, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rachel K.","family":"Lim","sequence":"additional","affiliation":[{"name":"Department of Dermatology, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin R.","family":"Johnston","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Brigham and Women\u2019s Hospital, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael W.","family":"Groff","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Brigham and Women\u2019s Hospital, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Theresa","family":"Williamson","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Massachusetts General Hospital, Boston"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Konstantina","family":"Svokos","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tiffany J.","family":"Libby","sequence":"additional","affiliation":[{"name":"Department of Dermatology, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John H.","family":"Shin","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Massachusetts General Hospital, Boston"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziya L.","family":"Gokaslan","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Curtis E.","family":"Doberstein","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James","family":"Zou","sequence":"additional","affiliation":[{"name":"Department of Biomedical Data Science and, by courtesy, Computer Science and Electrical Engineering, Stanford University, Stanford, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wael F.","family":"Asaad","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island"},{"name":"Department of Neuroscience, Norman Prince Neurosciences Institute, Rhode Island Hospital, Providence"},{"name":"Department of Neuroscience, Brown University, Providence, Rhode Island"},{"name":"Department of Neuroscience, Carney Institute for Brain Science, Brown University, Providence, Rhode Island"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10","reference":[{"key":"soi230083r2","doi-asserted-by":"publisher","DOI":"10.2196\/43110","article-title":"What does DALL-E 2 know about radiology?","volume":"25","author":"Adams","year":"2023","journal-title":"J Med Internet Res"},{"key":"soi230083r3","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1016\/j.bjps.2023.04.055","article-title":"What could we make of AI in plastic surgery education.","volume":"81","author":"Koljonen","year":"2023","journal-title":"J Plast Reconstr Aesthet Surg"},{"key":"soi230083r4","first-page":"1","article-title":"Gender shades: intersectional accuracy disparities in commercial gender classification.","volume":"81","author":"Buolamwini","year":"2018","journal-title":"Proc Mach Learn Res"},{"issue":"6464","key":"soi230083r6","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations.","volume":"366","author":"Obermeyer","year":"2019","journal-title":"Science"},{"issue":"3","key":"soi230083r7","doi-asserted-by":"publisher","first-page":"e69","DOI":"10.1227\/neu.0000000000002576","article-title":"Letter: the urgency of neurosurgical leadership in the era of artificial intelligence.","volume":"93","author":"Tang","year":"2023","journal-title":"Neurosurgery"},{"issue":"4","key":"soi230083r9","doi-asserted-by":"publisher","first-page":"1084","DOI":"10.1007\/s11606-020-05646-z","article-title":"Impact of physician race on patient decision-making and ratings of physicians: a randomized experiment using video vignettes.","volume":"35","author":"Saha","year":"2020","journal-title":"J Gen Intern Med"},{"issue":"11","key":"soi230083r10","doi-asserted-by":"publisher","first-page":"1086","DOI":"10.1001\/archsurg.141.11.1086","article-title":"Influences on medical student career choice: gender or generation?","volume":"141","author":"Sanfey","year":"2006","journal-title":"Arch Surg"},{"issue":"1","key":"soi230083r11","doi-asserted-by":"publisher","first-page":"702","DOI":"10.1186\/s12909-022-03765-1","article-title":"Trainee doctors\u2019 perceptions of the surgeon stereotype and its impact on professional identification: a qualitative study.","volume":"22","author":"Wainwright","year":"2022","journal-title":"BMC Med Educ"},{"issue":"3","key":"soi230083r12","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1016\/j.jamcollsurg.2018.12.011","article-title":"Reducing implicit bias: association of women surgeons #heforshe task force best practice recommendations.","volume":"228","author":"DiBrito","year":"2019","journal-title":"J Am Coll Surg"},{"issue":"8","key":"soi230083r13","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2019.10228","article-title":"Defining barriers and facilitators to advancement for women in academic surgery.","volume":"2","author":"Thompson-Burdine","year":"2019","journal-title":"JAMA Netw Open"},{"issue":"11","key":"soi230083r14","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1001\/jama.2022.13081","article-title":"Graduate medical education, 2021-2022.","volume":"328","author":"Brotherton","year":"2022","journal-title":"JAMA"},{"issue":"1","key":"soi230083r16","doi-asserted-by":"publisher","first-page":"378","DOI":"10.1186\/s12909-022-03453-0","article-title":"Surgical career choices of medical students in China: does gender bias play a role?","volume":"22","author":"Zhou","year":"2022","journal-title":"BMC Med Educ"},{"issue":"4","key":"soi230083r17","doi-asserted-by":"publisher","first-page":"1122","DOI":"10.3758\/s13428-014-0532-5","article-title":"The Chicago face database: a free stimulus set of faces and norming data.","volume":"47","author":"Ma","year":"2015","journal-title":"Behav Res Methods"},{"issue":"3","key":"soi230083r18","doi-asserted-by":"publisher","DOI":"10.2196\/34144","article-title":"Inequities in health care services caused by the adoption of digital health technologies: scoping review.","volume":"24","author":"Yao","year":"2022","journal-title":"J Med Internet Res"},{"issue":"1","key":"soi230083r19","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1007\/s11060-020-03503-x","article-title":"The influence of race and socioeconomic status on therapeutic clinical trial screening and enrollment.","volume":"148","author":"Morshed","year":"2020","journal-title":"J Neurooncol"},{"issue":"3","key":"soi230083r20","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1007\/s11060-022-04020-9","article-title":"Predicting access to postoperative treatment after glioblastoma resection: an analysis of neighborhood-level disadvantage using the Area Deprivation Index (ADI).","volume":"158","author":"Rivera Perla","year":"2022","journal-title":"J Neurooncol"},{"key":"soi230083r22","doi-asserted-by":"publisher","DOI":"10.1227\/neu.0000000000002551","article-title":"Performance of ChatGPT, GPT-4, and Google Bard on a neurosurgery oral boards preparation question bank.","author":"Ali","year":"2023","journal-title":"Neurosurgery"},{"key":"soi230083r8","first-page":"1493","article-title":"Easily accessible text-to-image generation amplifies demographic stereotypes at large scale.","author":"Bianchi","year":"2023"},{"key":"soi230083r1","unstructured":"Open AI. DALL\u00b7E now available without waitlist. Published 2022. Accessed October 13, 2023. https:\/\/openai.com\/blog\/dall-e-now-available-without-waitlist"},{"key":"soi230083r5","doi-asserted-by":"crossref","unstructured":"Dastin? J. Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. Published 2018. Accessed October 13, 2023. https:\/\/www.reuters.com\/article\/us-amazon-com-jobs-automation-insight\/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G","DOI":"10.1201\/9781003278290-44"},{"key":"soi230083r15","unstructured":"American Association of Medical Colleges. Physician specialty data report. Published 2022. Accessed October 13, 2023. https:\/\/www.aamc.org\/data-reports\/workforce\/report\/physician-specialty-data-report"},{"key":"soi230083r21","unstructured":"Open AI. Reducing bias and improving safety in DALL-E 2. Published 2023. Accessed October 13, 2023. https:\/\/openai.com\/blog\/reducing-bias-and-improving-safety-in-dall-e-2"}],"container-title":["JAMA Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/jamanetwork.com\/journals\/jamasurgery\/articlepdf\/2811920\/jamasurgery_ali_2023_oi_230083_1703714449.54725.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,10]],"date-time":"2024-01-10T16:02:06Z","timestamp":1704902526000},"score":1,"resource":{"primary":{"URL":"https:\/\/jamanetwork.com\/journals\/jamasurgery\/fullarticle\/2811920"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,1,1]]},"references-count":22,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2024,1,1]]}},"URL":"https:\/\/doi.org\/10.1001\/jamasurg.2023.5695","relation":{},"ISSN":["2168-6254"],"issn-type":[{"value":"2168-6254","type":"print"}],"subject":[],"published":{"date-parts":[[2024,1,1]]}}}