Data & Innovation • Financial Services • AWS
AI applied to glossary analysis and optimization of bank statements
A solution that combines human logic and artificial intelligence to interpret and enrich glossaries more accurately and efficiently.
About the client
The only state-owned commercial banking institution in Chile, legally established in 1953 through the merger of several public financial entities.
Today, it offers a wide range of financial services to individuals and businesses, including savings accounts, mortgage loans, and financing for small and medium-sized enterprises.
It also manages the Unified Fiscal Account, facilitating the Chilean government’s financial operations.
Needs
The bank sought to implement a Proof of Concept (PoC) with Generative Artificial Intelligence (GenAI) as an alternative to its current system for enriching transaction log glossaries.
The goal was to improve the clarity of the language used in bank statements (cartolas) and make transaction movements easier for users to interpret.
Solution
An orchestrator was designed to simulate the process carried out by customer service advisors to identify and contextualize incomplete glossaries.
Using Generative AI, the solution explores transactional logs based on the keywords present in each entry, generating enriched versions that are clearer and more useful for the end customer.
This enabled a significant improvement in the construction of bank statements.
