Mirae Asset Launches AWS AI-Powered "Financial Product Knowledge Search" Service
Mirae Asset Securities announced on November 20 that it will officially launch its “Financial Product Knowledge AI Search” service, built using generative AI technology from Amazon Web Services (AWS).
The service enables users to search across financial product information, company information, disclosures, research materials, and other data scattered across multiple systems. When users ask questions in everyday language, AI finds relevant information and provides answers along with the sources and applicable dates. The service was designated an innovative financial service in March and received a passing grade in the Financial Security Institute’s generative AI security assessment in July.
Previously, whenever employees explained financial products to customers, they had to check multiple systems and documents. With the new service, they can find the information they need all at once simply by asking AI, which is expected to improve work efficiency and the accuracy of customer support.
The service was designed with a focus on improving the accuracy and reliability of the information provided by AI. Using AWS, the company designed an ontology, or framework for classifying and linking financial product and company information, and applied knowledge graph technology to connect approximately 13,000 documents with reference data for financial products. It also designed the system so that multiple AI models divide the tasks of retrieving information and drafting answers. If information is unavailable, the system indicates that it does not have the information rather than generating an answer, reducing the risk of providing incorrect information.
The company has also put safeguards in place to enhance security and stability. It uses Amazon Bedrock Guardrails to screen for malicious external instructions and harmful content in advance, and Amazon Neptune to systematically store and retrieve relationships among products, companies, and documents. The service also includes security reviews, AI red teaming (testing to identify weaknesses in AI from an attacker’s perspective), and performance revalidation procedures to ensure answer quality is maintained when models are changed.
In addition, the service is built on Amazon Bedrock, which supports the use of a variety of foundation models, or general-purpose AI models pretrained on vast amounts of data. This allows the company to use frontier models approved for innovative financial services simultaneously and to switch to better models as soon as they become available, following the Financial Security Institute’s procedures.
The service was launched for employees in August, and by October 20, more than 450 employees had used it for over 2,000 complex queries. Mirae Asset Securities plans to expand the service, which currently primarily serves employees, to customers in the first half of 2027. The service is expected to be an example of how the company is putting into practice the core aims of “Mirae Asset 3.0”: supporting AI-based investment decisions and differentiating the customer experience. The company also plans to introduce an “AI Pension Work Agent” in 2027 in collaboration with AWS. The plan is to use AI to search for pension-related information and support pension-related work, improving the accuracy and consistency of customer service while strengthening links to product information.
Hot Picks Today
"Two Spend 500,000 Won in a Flash, Yet Wait Two Hours?" ... 'Dreamland' Shunned by Kids and Foreign Tourists
- The Era of Getting 200 Million Won After 30 Years Is Over... A Japanese Bank's "Perpetual Department Head" Who Just Clocked In Cries Out
- Spain Hit by Its Worst Tragedy... Lee Kangin, Moreno and Real Madrid Join Soccer World in Mourning
- They Said, “Make It Your Home for Life” - Then Suddenly Kicked 200 Passengers Off... What Happened to the Around-the-World Cruise?
- "130 Rice Bowls Sold Out in a Day"... 'Unfiltered Face' Adds 8 Million Won to Night Market Sales: China's Bungeunwoo
“The starting point of this project was the understanding that data must be ready before models if AI is to be properly used in our work,” said Joo Semin, head of the Tech&AI Division at Mirae Asset Securities. “Now that we’ve created a structure in which people and AI can work with the product knowledge that was previously scattered across different systems, it’s time to deliver those benefits to our customers as well.”
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.