[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81937-en":3,"doc-seo-81937-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},81937,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Whose Fairness Structural Concentration in AI Bias Research","AI increasingly mediates consequential decisions across domains such as healthcare, law, and public services, while research has developed methods to measure and mitigate bias. The field treats fairness definitions, benchmarks, and debiasing frameworks as universal, even though the composing research community has never been systematically characterized. Analysis of 692 publications across five thematic domains shows structural concentration by country, institution, and author—especially in the foundational general fairness and mitigation area. This concentration raises concerns about limited generalization to populations and settings where AI is deployed, motivating an interactive atlas for continuous monitoring.","arXiv :2607 .05574v 1 [ cs .CY] 6 Jul 2026  \nWhose fairness? Structural concentration in AI  \nbias research  \nAbhash Shrestha 1*, Subigya Gautam 1 , Anu Sapkota 1 , Sanju Tiwari2,3 , Tek Raj Chhetri 1,4*  \n1* Center for Artificial Intelligence (AI) Research Nepal,  \nSundarharaincha-09, Koshi, Nepal.  \n2 Sharda University, Delhi-NCR,India.  \n3 Shodhguru Innovation and Research Labs, India.  \n4 McGovern Institute for Brain Research, Massachusetts Institute of Technology, 43 Vassar Street, Cambridge, 02139, MA, United States.  \n*Corresponding author(s). E-mail(s): [abhash.shrestha@cair-nepal.org](abhash.shrestha@cair-nepal.org) ;  \n[tekraj.chhetri@cair-nepal.org](tekraj.chhetri@cair-nepal.org) ;[tekraj@mit.edu](tekraj@mit.edu) ; Contributing authors: [subigya.gautam@cair-nepal.org](subigya.gautam@cair-nepal.org) ; [anu.sapkota@cair-nepal.org](anu.sapkota@cair-nepal.org) ; [tiwarisanju18@ieee.org](tiwarisanju18@ieee.org) ;  \nArtificial intelligence increasingly mediates consequential decisions in healthcare, law, and public services, and the field has responded with an extensive methodology for measuring and mitigating bias. Yet the fairness definitions, benchmarks, and debiasing frameworks on which this methodology rests are treated as universal while being produced by a research community whose composition has never been characterized. We show that the AI bias research are structurally concentrated, and that this concentration is greatest, geographically, in precisely the domain the rest of the field inherits from. Analyzing 692 publications spanning five thematic domains, combining bibliometric analysis with semantic clustering, we find that research activity is dominated by a small set of countries, institutions, and authors, with the United States leading publication output and collaboration networks across every domain and most strongly in general fairness and bias mitigation, the largest, most-cited domain with meaningful representation across all four semantic clusters. Low-and middle-income countries remain largely absent from the community and its collaboration networks, and citation influence is highly skewed (median = 9; mean =93.5 ), indicating that a  \n1  \nsmall fraction of publications disproportionately shapes the field. Because the generalfairness domain supplies the definitions and benchmarks that application areas apply, concentration of research effort in this foundational domain propagates across AI bias research as a whole-raising the concern that mitigation methods developed and validated within a narrow set of contexts may not generalize to all populations and settings where AI is deployed. We provide an interactive atlas for continuous monitoring of the field’s structure.  \n1 Introduction  \nArtificial intelligence (AI) is increasingly reshaping how knowledge is produced, services are delivered, and decisions are made across society. In this article, we use the term AI broadly to refer to computational paradigms such as machine learning (ML), deep learning (DL), and large language models (LLMs) . The rapid adoption of these systems across domains such as healthcare [1, 2] and education to governance and public services [3] has intensified concerns about AI bias [4–9] . We define AI bias as systematic error or variation in the behavior or outputs of an AI system that produces distorted or harmful outcomes for particular individuals or groups, often reflecting biases in data, model design, human decision-making, or broader sociotechnical structures. Such bias can reproduce and amplify existing inequalities, with direct consequences for people’s lives when AI systems are embedded in highstakes sociotechnical contexts. The stakes are well documented: medical AI systems overassociate Hispanic and Asian patients with tuberculosis and under-recommend computed tomography (CT) and magnetic resonance imaging (MRI) imaging for Black patients [7], while risk-assessment tools such as COMPAS (Correctional Offender Manage","cbCaitwDYdwTUrla","https://ap.wps.com/l/cbCaitwDYdwTUrla","pdf",4739688,6,1,27,"English","en",105,"# Introduction\n## Defining AI bias and stakes\n## Fairness frameworks as “universal” assumptions\n## Motivation: concentration and generalization risk","[{\"question\":\"What does the paper mean by “AI bias”?\",\"answer\":\"AI bias is defined as systematic error or variation in an AI system’s behavior or outputs that leads to distorted or harmful outcomes for specific individuals or groups. It can originate from data, model design, human decision-making, and broader sociotechnical structures.\"},{\"question\":\"What does the study find about the structure of AI bias research?\",\"answer\":\"The paper finds that AI bias research is structurally concentrated across countries, institutions, and authors. The concentration is strongest in the general fairness and bias mitigation domain, which supplies definitions and benchmarks used by other application areas.\"},{\"question\":\"Why might concentrated research lead to biased real-world AI deployments?\",\"answer\":\"Because foundational definitions and benchmarks are developed and validated within a narrow set of contexts, mitigation methods may not generalize to all populations and settings. The paper also notes that low- and middle-income countries are largely absent from the community and its collaboration networks.\"}]","Whose Fairness Structural Concentration in AI Bias Research | PDF",1784177149,68,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"whose-fairness-structural-concentration-in-ai-bias-research","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/whose-fairness-structural-concentration-in-ai-bias-research/81937/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-07-30","2026-07-16",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What does the paper mean by “AI bias”?","Question",{"text":77,"@type":78},"AI bias is defined as systematic error or variation in an AI system’s behavior or outputs that leads to distorted or harmful outcomes for specific individuals or groups. It can originate from data, model design, human decision-making, and broader sociotechnical structures.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"What does the study find about the structure of AI bias research?",{"text":82,"@type":78},"The paper finds that AI bias research is structurally concentrated across countries, institutions, and authors. The concentration is strongest in the general fairness and bias mitigation domain, which supplies definitions and benchmarks used by other application areas.",{"name":84,"@type":75,"acceptedAnswer":85},"Why might concentrated research lead to biased real-world AI deployments?",{"text":86,"@type":78},"Because foundational definitions and benchmarks are developed and validated within a narrow set of contexts, mitigation methods may not generalize to all populations and settings. The paper also notes that low- and middle-income countries are largely absent from the community and its collaboration networks.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]