BSP sets AI guardrails for financial institutions

The Bangko Sentral ng Pilipinas has issued governance principles for the adoption of artificial intelligence by all BSP-supervised financial institutions, aiming to improve financial services while protecting consumers and maintaining trust in the financial system.

Memorandum No. M-2026-031, issued on June 24, 2026, contains the “Governance Principles for Artificial Intelligence in Financial Services.”

The guidance is intended to help BSP-supervised financial institutions, or BSFIs, develop their own AI governance and risk management frameworks.

The central bank organized its approach around five principles: sustainability, transparency, accountability, responsibility and security, collectively abbreviated as STARS.

“AI is spreading across BSFI operations and STARS provides them with principles that can help them innovate while mitigating unintended consequences from the use of the technology,” said Deputy Governor Lyn I. Javier.

“We want BSFIs to take advantage of AI, especially to serve their customers, and do so while being guided by developing global standards.”

The memorandum applies to every BSFI, but implementation must be proportionate to the nature, extent, scale, complexity and materiality of each institution’s AI systems.

Implementation must also correspond to the institution’s overall operational complexity and risk profile.

The proportionate approach means that an institution’s safeguards should reflect the potential consequences of its AI use rather than follow a single set of controls regardless of risk.

The guidance also covers outsourced service providers supporting AI-related activities under a shared responsibility model.

This provision is significant because outsourcing a system does not remove the supervised institution’s responsibility for risks to customers, data and operations. Existing BSP standards similarly place accountability for outsourced technology services on a BSFI’s board and senior management, according to the central bank’s IT risk management guidelines.

Although the principles are voluntary and non-binding, they reflect the BSP’s minimum supervisory expectations for AI adoption.

BSFIs are encouraged to incorporate STARS into their governance frameworks, risk management strategies and operational processes.

Risks arising from emerging technologies remain covered by the BSP’s existing Information Technology Risk Management framework, which focuses on information security, outsourcing and project management.

The new principles are aligned with guidelines developed by the Organization for Economic Cooperation and Development, the Association of Southeast Asian Nations, the Financial Stability Institute and other organizations.

The OECD AI Principles, adopted in 2019 and updated in 2024, promote transparency, explainability, security, accountability and respect for human rights throughout an AI system’s life cycle.

The ASEAN Guide on AI Governance and Ethics likewise provides a voluntary framework that organizations can tailor to their industry, technological complexity and level of risk.

For the public, the BSP guidance matters because financial institutions use AI for activities such as credit underwriting, fraud detection, risk management, back-office automation and customer interaction, according to the Bank for International Settlements.

Weak governance could allow inaccurate, biased or opaque systems to affect access to credit, flag legitimate transactions as fraudulent or expose sensitive customer information.

Clear accountability and transparency can help customers understand AI-supported outcomes, challenge adverse decisions and identify which institution remains responsible when an outside provider operates the technology.

The issuance forms part of the BSP’s broader effort to use technology in advancing financial stability and inclusion.

The complete memorandum is available on the BSP website.