A novel extreme learning machine (ELM) based grid search method with cutoff selection is proposed in this paper for credit risk assessment.Different from existing models using a fixed cutoff (0.0 or 0.5), the proposed novel model especially considers the cutoff as one important evaluation parameter in credit risk modeling, to enhance the assessment accuracy.In particular, with the powerful artificial intelligence (AI) tool of ELM as the basic classification, the simple but efficient optimization algorithm of grid search method is performed to select the optimal cutoff.Three main steps are included: (1) ELM training using the training dataset, (2) cutoff optimization via the grid search method using the training and validation datasets, and (3) classification generation based on the optimal ELM and cutoff using the testing dataset.For illustration and verification, the experimental study with two publicly available credit datasets as the study samples confirms the superiority of the proposed ELM-based grid search method over other some popular classification techniques without cutoff selection.