A company uses an AI tool to screen CVs. Fewer female candidates are shortlisted. What is the most likely cause of this bias?, The training data contained historical hiring biases , The AI prefers shorter resumes , Female candidates apply less often , The software is broken, A city uses facial recognition, but it misidentifies people with darker skin tones. What is the main issue here?, Cameras are low quality , Poor lighting conditions , Lack of diversity in training data , Slow processing speed, AI used in a bank denies more loans from certain neighbourhoods. What is the most likely reason?, The bank has limited funds , Random system errors , Incomplete applications , The algorithm uses biased historical data, Men are shown higher-paying job ads more often. Why might this happen?, The algorithm learns from biased engagement data , Ads are randomly assigned , Women don’t search for these jobs , System malfunction, AI underestimates needs of certain ethnic groups. What is a key reason?, The system runs too slowly , Incomplete or unrepresentative training data , Doctors ignore AI suggestions , Patients provide wrong data, An algorithm keeps sending police to the same areas based on past data. What issue does this highlight?, Hardware failure, Lack of staff, Bias in the algorithm reinforced by its own past decisions, Random selection, Low-income applicants to a private school are disadvantaged. What may be causing this?, System crashes , Students apply late , C. Teachers input wrong grades , Algorithm is biased, A voice assistant struggles with certain accents. What is the main issue?, The device is outdated , Lack of diverse voice data , Users speak too quietly , Internet connection issues, When renewing home insurance, certain groups are charged higher premiums. What is the ethical concern?, Faster processing, Customer demand, Discrimination caused by biased data, Company profit goals, An algorithm used to grade student essays gives lower marks to students who write in certain dialects or styles. What could explain this?, The essays are shorter, Limited storage, Teachers manually change the grades, Misunderstanding language due to biased training data.

Algorithmic bias

par
Imprimer
Incorporer

Classement

Style visuel

Options

Changer de modèle

)
Restauration auto-sauvegardé :  ?