Model disagreement helps spot risky inputs by measuring uncertainty with...

https://sierra-wiki.win/index.php/Loan_Approval_Models_Disagree_on_Borderline_Applicants:_What_Now%3F

Model disagreement helps spot risky inputs by measuring uncertainty with metrics like entropy or ensemble variance. When models disagree sharply, those top 1-2% cases get flagged for human review or targeted labeling