The rapid expansion of Open Banking in Morocco, accelerated by Bank Al-Maghrib’s regulatory frameworks and the digital transformation of financial services, creates systemic challenges for credit risk assessment across distributed banking networks. This paper proposes a federated learning (FL) architecture enhanced with differential privacy (DP) mechanisms for collaborative credit risk modeling in the Moroccan banking sector. The framework enables financial institutions to jointly train predictive models on distributed client data without compromising individual privacy or violating data sovereignty constraints. Drawing on empirical data from a 500-client survey and 25 expert interviews conducted across the Rabat-Sale-Kenitra (RSK) region, we design a privacy-preserving gradient aggregation protocol tailored to the heterogeneous data structures of Moroccan retail banking portfolios. Our federated model achieves a 14.3% improvement in default prediction accuracy over centralized baselines while maintaining a privacy budget (epsilon) of 1.2 under the Gaussian mechanism, meeting the confidentiality thresholds mandated by Moroccan data protection law (Loi n 09-08). The results validate the feasibility of federated credit scoring as a secure, interoperable foundation for Open Banking ecosystems and contribute a blueprint for AI-driven financial inclusion across the African banking sector.