AI Bias
Responsible AI, Safety & SecurityChapter 11: Safe, Kind and FairTerm 189 of 253
Systematically unfair outcomes due to skewed data or design.
At Sunrise School
Gopal asksBit replied faster and better in English than in Telugu.
Bit answersThat's AI bias from skewed training data. We test every language and fix it, so every family gets the same quality.
In the real worldHiring model favouring one gender.
Watch the short
Short coming soonSeason 3 · Day 19Follow @haigopala
Short coming soonSeason 3 · Day 19Follow @haigopala.telugu
Short coming soonSeason 3 · Day 19Follow @haigopala.hindi
Quick check
Quick check. Is it fine if one language gets worse answers? Yes or no?
No! Every family deserves the same quality.
Practise it
Try this term and all 253 others in the free Model Training Arena.
Open the Arena →Follow Bit daily: YouTube English · తెలుగు · हिंदी | Instagram English · తెలుగు · हिंदी