The COSMIQ Blog
Articles tagged “language models” · Clear filter
AI Models Show Bias Against Certain Accents in Scoring Interviews and Tests, UW Researchers Find
A University of Washington-led research team has uncovered that leading AI models exhibit bias against specific accents when evaluating spoken English in critical contexts like job interviews, academic presentations, and language proficiency tests. This finding raises important questions about fairness and equity in AI-driven assessment tools.
Learn anything, free.
COSMIQ is a free, voice-driven AI tutor for every learner. No credit card, ever.
Start learning free →