From AI Pilot to Real-World Impact in Credit Assessment

At an AI Association of Hong Kong panel, Louis Lee shared Propcap's experience deploying AI in credit risk assessment—and the lessons learned along the way.

Mortgage risk assessment presents real challenges, including fragmented data, regulatory scrutiny and the need to earn human trust. Propcap's Agentic AI system now automates 95% of its credit assessment process, reducing approval time by 70% and supporting the expansion of ESG lending. These gains must go hand in hand with explainability and bias checks to help ensure decisions are transparent and fair.

For organisations moving from prototype to production, Louis emphasised starting small, demonstrating impact early and working with the right partners. Phased implementation and cross-functional alignment can help teams make progress, even with limited resources.

His advice: treat AI as a business transformation, not simply a technology project. Define measurable goals, invest in reliable data foundations and build compliance into the process from the outset. With time, trust and iteration, AI can help financial institutions make better decisions, serve customers and scale their impact.