Tag
This research explores how large language models develop new social biases through adaptive exploration, highlighting implications for AI fairness and ethical considerations.