BSP sets AI governance framework for banks
The Bangko Sentral ng Pilipinas (BSP) has released voluntary governance principles for the use of artificial intelligence (AI) in financial services, laying out its expectations for how banks should manage the risks of the fast-growing technology without stifling innovation.
The guidance, released through Memorandum No. M-2026-031 is intended to help banks and other financial institutions develop their own policies for overseeing AI and managing the risks associated with its use.
The central bank recommends that financial institutions establish AI governance frameworks tailored to the nature, scale and complexity of their AI systems, as well as their overall operations and risk profiles.
The BSP stressed that the principles are voluntary and nonbinding, saying they are meant to promote the ethical and responsible use of AI rather than limit technological advancement. At the same time, it said it would continue to monitor developments in the field and issue regulations when necessary to foster innovation while safeguarding the stability and competitiveness of the financial system.
“The BSP does not endorse or prohibit specific technologies or tools,” the memo read. “The adoption of these governance principles must align with full compliance with existing laws, rules, and regulations.”
The guidance is anchored on five governance principles, summarized by the acronym STARS: sustainability, transparency, accountability, responsibility and security.
Long-term value
Under the framework, banks are encouraged to ensure AI delivers long-term value while minimizing environmental and social risks, clearly disclose when the technology is used and maintain sufficient documentation so its outputs can be understood and audited.
The BSP also said institutions should keep humans ultimately responsible for decisions made with AI, with boards and senior management providing oversight.
The guidance also calls on banks to guard against bias and discriminatory outcomes by using high-quality data and respecting customers’ privacy rights, while strengthening cybersecurity defenses against AI-specific threats such as data poisoning and adversarial attacks. Institutions are also urged to continuously monitor AI systems for errors, bias and other vulnerabilities after deployment.
“The absence of a comprehensive AI governance framework introduces multiple layers of risk, including operational risk, information technology, model risks, market conduct risk, reputational risk, strategic risk and legal risk,” the BSP said.
“Moreover, without clear guidelines and safeguards, AI systems may perpetuate biases, leading to unfair practices and the exclusion of individuals from access to financial products and services,” it added.
