Artificial intelligence startup Enhans said on the 2nd that it has begun a pilot project with Industrial Bank of Korea (024110) to test ontology-based AI data analytics.
The pilot centers on structuring the bank's internal operational knowledge and data so that AI agents can use them, and on verifying whether frontline employees can retrieve and analyze the data they need using only plain-language questions. Until now, employees had to request materials from the IT department or handle database query commands themselves.
Enhans plans to install its proprietary AI agent platform, AgentOS (AOS), on IBK's internal network. The approach involves building a semantic layer that links business meaning and relationships to the bank's information-system data, and applying ontology technology to organize business concepts and their relationships.
On that basis, the project applies Text-to-SQL, which converts plain-language questions into SQL, the command language used to query databases. The company also plans to improve answer accuracy by reflecting the bank's own business terminology and decision criteria in retrieval-augmented generation (RAG). Beyond simple data retrieval, the pilot will test whether the system can analyze causes, draw out business implications, and present the reference data and reasoning path used in its answers.
The pilot runs through next month. The two sides plan to quantitatively assess the results using a "golden set," a benchmark dataset prepared in advance for evaluation.
"In finance, it is important to build a semantic framework for data so that AI can make use of domain knowledge such as business terminology and decision criteria," Enhans CEO Lee Seung-hyun said. "We will verify whether analysis results can be presented together with their supporting evidence, even on a bank's internal network."
