How might the risk of algorithmic bias—particularly from historical, non-diverse datasets—impact the fairness of AI-driven agricultural advice?
