Selected for ministry's public-private R&D commercialization project

[Shinhan Card]
[Shinhan Card]

Shinhan Card is developing an AI engine that will recommend financial products tailored to individual customers in real time and support their decision-making.

The company said Wednesday that it will begin developing the technology after being selected for the "2026 Public-Private Joint Technology Commercialization R&D" project's third round, overseen by the Ministry of SMEs and Startups.

The project is aimed at preparing for the era of "agent commerce," in which AI agents search for financial products, compare conditions and make selections on behalf of customers. The goal is to build a next-generation AI infrastructure that systematically reflects complex product terms and related policies and regulations to find the optimal product for each customer.

Across the financial industry, the use of generative AI is expanding beyond product recommendations based on spending patterns and asset analysis to supporting specialized financial operations. Shinhan Financial Group has also been widening the scope of its AI agent applications — from customer consultations to product recommendations and internal task support — while investing in related technology.

However, current AI-based decision-making systems, which rely on probabilistic judgment, have limitations such as inconsistent answers or the omission of complex conditions. This can lead to distortions, and changes in policy or regulations are not always reflected in time.

To address this, Shinhan Card plans to structure the complex conditions embedded in financial products, services, policies and regulations, and to automate the decision-making process. The system will be designed so that when a product or related system changes, only the relevant condition needs to be updated, allowing the change to be reflected in the AI's judgment process in real time.

For card products, factors such as the previous month's usage record, merchant categories, and discount or point accumulation limits must be considered. These must also be weighed alongside exception conditions — such as excluded industries or transactions — as well as card type and status. Shinhan Card plans to systematically explore the complex relationships among these conditions and incorporate internal policies, operating standards and financial regulations to recommend products suited to each customer.

Shinhan Card will first apply the technology to financial product recommendations before gradually expanding its use to risk management, marketing and customer consultation. The company is developing and testing the technology jointly with AI infrastructure company Connectionary.

"We will secure the accuracy and reliability of financial decision-making in an environment where AI judges and acts on its own, strengthening our competitiveness," a Shinhan Card official said. "We will continue to upgrade the customer experience through AI-driven innovation in financial services."

-


forest@heraldcorp.com