Face recognition and biometric research are contributing to rapid growthinnewbiometricsurveillancetechnologies.Butmanyofthe datasets used for these technologies rely on media collected from non-consensual,nonregulatedsources.Researchersrefertothismediaasbeing"inthewild."Thisanalysisexaminesthewidespreadand largely unregulated use of images "in the wild" that were captured fromcampuses,CCTVcamerafeeds,socialmedia,celebritydatabases,andbyscrapinginternetsearchengines.Thefindingspresented here show that millions of individuals have unknowingly been used fortrainingfacerecognitionandotherbiometricanalysisalgorithms inbothacademicandcommercialapplications.Datacompiledforthis project,alongwithmorein-depthanalysesforeachdataset,isavailableontheresearchprojectwebsitehttps://megapixels.cc[https:// exposing.ai/].

No abstract