When Your AI Tutor Knows Your Weaknesses But Not Your Name

Episode 60 of AI Ethics with Fexingo explores the hidden world of AI-powered personalized tutoring. Lucas and Luna unpack a 2025 Stanford study revealing that adaptive learning platforms like Khan Academy's Khanmigo and Carnegie Learning's MATHia collect granular student data—keystroke patterns, time per question, even emotional tone—to predict performance. But when these systems fail to explain why a student is struggling, they can reinforce stereotypes about 'slow learners' without considering external factors like home environment or learning disabilities. The hosts discuss how one school district in Texas found that its AI tutor flagged Black students as 'low engagement' at twice the rate of white peers, even after controlling for test scores. They also examine the European Union's proposed AI liability directive, which would require educational AI to provide human-readable explanations for their predictions. The episode closes with a call for transparency: if an AI knows your child's weaknesses, shouldn't it at least get their name right? #AIEthics #EdTech #AdaptiveLearning #StudentPrivacy #AlgorithmicBias #Khanmigo #CarnegieLearning #MATHia #StanfordStudy #TexasSchoolDistrict #EULiabilityDirective #EducationalAI #PersonalizedLearning #DataCollection #LearningAnalytics #Technology #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo