Central Bank of the Philippines issues non-binding AI governance principles for supervised financial institutions

The Central Bank of the Philippines has issued a guidance paper on Governance Principles for Artificial Intelligence in Financial Services. The paper sets out non-binding minimum supervisory expectations for AI adoption and recommends that BSP-supervised financial institutions formally develop AI governance frameworks proportionate to the nature, extent, scale, complexity and materiality of their AI systems, as well as the institution’s overall risk profile and operational complexity. The principles apply to financial institutions supervised by or registered with the central bank and extend to vendors and outsourced service providers involved in AI-related activities under a shared-responsibility model. The framework is built around five STARS principles: sustainability, transparency, accountability, responsibility or social fairness, and security. It ties those principles to controls across the AI lifecycle from planning and development to validation, deployment and monitoring. Key expectations include maintaining AI inventories and documentation, notifying users when AI output is used, ensuring auditability and defined mitigation procedures, embedding human oversight and clear role segregation, using qualified staff, protecting personal data and avoiding discriminatory outcomes, conducting independent validation and testing, and extending cybersecurity, incident response, business continuity and ongoing performance monitoring to AI systems. The guidance states that existing laws and regulations remain applicable, while the forthcoming Model Risk Management framework will address algorithmic fairness and model risks across the model lifecycle. The central bank said it will continue to monitor developments in AI and may issue further regulations or policies when necessary.