IQSS affiliate Mahzarin Banaji finds LLMs amplify human biases rather than mitigate them. A recent experiment saw an LLM advise girls to pursue stereotypical careers and to ask for lower salaries than what it advised boys.
news.harvard.edu/gazette/stor...
#AI #ImplicitBias #MachineLearning #Harvard
news.harvard.edu
One effort to safeguard humanity: AI “constitutions” — Harvard Gazette
Berkman Klein Center heard calls to keep models “honest, helpful, harmless.” It’s a work in progress.