ART framework sets governance foundations for banking AI
Responsible AI adoption in banking is the focus of the Hong Kong Monetary Authority (HKMA)'s ART framework, announced on October 15, 2026. It combines supervisory guidance, training data sharing and further testing of agentic AI.
Three routes to responsible deployment
ART centers on Accountability, Resilience and Trust.
The HKMA issued updated Supervisory Guidance for Responsible Adoption of Artificial Intelligence, drawing on the second Generative AI (GenAI) Sandbox cohort.
A training data sharing platform with Hong Kong Cyberport Management Company Limited (Cyberport) and the University of Hong Kong makes data quality assurance a core design principle.
The framework also includes further agentic AI testing through GenAI Sandbox++ and an upcoming discussion paper proposing a supervisory framework for agentic AI governance.
That paper will explore intent-anchored accountability, a registry of material AI agents and effective intervention mechanisms.
The measures support adoption while maintaining governance and operational resilience.
Agents dominate sandbox trials
More than 60 percent of second-cohort sandbox use cases explored agentic AI, including automated planning and execution of multi-step tasks within defined boundaries and controls.
The HKMA reported positive technical trial results.
It is co-hosting the two-day Finance + AI Forum with Cyberport on October 15 and 16, 2026, marking completion of those trials.
Deputy Chief Executive Arthur Yuen placed ART under Fintech 2030 and emphasized clear accountability, resilient operating models and effective human oversight.
Autonomy needs enforceable boundaries
The HKMA links growing AI autonomy to governance and operational resilience as systems gain access to data, tools and workflows.
Its proposed agent registry and intervention mechanisms remain subjects for a discussion paper.
Yuen ties responsible deployment to trust and financial stability.