AI forecasting improves with info asymmetry
Giving AI agents different evidence to discuss—rather than the same—makes group forecasting significantly more accurate.
Giving AI agents different evidence to discuss—rather than the same—makes group forecasting significantly more accurate.