Athena Technologies, a private AI startup based in Tokyo, and Kiyo Bank, headquartered in Wakayama, have launched a proof of concept for a sales support AI that runs in an environment isolated from the external internet. The AI uses past transaction data and external information to generate hypothetical client needs and discussion points for meetings. The initiative aims to support customer understanding and sales preparation while testing effectiveness and practicality under strict financial security requirements.
Responding to More Complex Client Needs
Corporate clients now present increasingly diverse and sophisticated management challenges, requiring relationship managers to assess situations accurately and propose tailored solutions. At the same time, preparation for client meetings demands substantial effort to review past transactions, external information, and potential areas of concern. These conditions create a need for tools that reduce preparation time without lowering accuracy or security.
How the Sales Support AI Works
The AI organizes hypothetical needs for each corporate client using Kiyo Bank's existing transaction records and relevant external information. It then presents key perspectives and specific questions that staff should confirm during face-to-face meetings as hearing points. By supporting information collection and organization, the system helps relationship managers deepen customer understanding and improve the quality of their proposals.
Security Measures for Confidential Banking Data
Banking operations involve highly confidential customer information, so generative AI deployments must meet strict security standards. The pilot operates the AI within an environment cut off from the external internet, which helps protect sensitive data while still enabling practical use. This design allows Kiyo Bank and Athena to verify that convenience and safety can be balanced in a real banking context without exposing data to outside networks.
Expected Outcomes and Efficiency Gains
By assisting with pre-meeting preparation, the AI is expected to help relationship managers work more efficiently and with greater confidence. The time saved on information collection and organization can be redirected toward direct dialogue with customers, which is central to relationship-based banking. Kiyo Bank and Athena aim to deepen customer understanding, enhance proposal quality, and strengthen overall sales activities as a result.
Company Roles and Collaboration
Kiyo Bank was established in 1895 and is headquartered in Wakayama Prefecture, Japan, where it serves a broad regional customer base. Athena Technologies was founded in December 2023 and operates from Bunkyo Ward, Tokyo, as a provider of private AI solutions. The company develops the Athena Platform, an AI security platform, and delivers research, development, and consulting services.
Future Development and Broader Plans
During the proof of concept, Athena will evaluate the AI's effectiveness, practicality, and operational challenges in Kiyo Bank's actual business environment. The company will use these findings to improve and refine features that fit the bank's workflows and address any issues identified during testing. Athena also plans to continue developing private AI solutions for financial institutions and other organizations with high security requirements.
This collaboration highlights a practical route for adopting generative AI in banking without compromising data security. By combining Kiyo Bank's customer information with Athena's private AI technology, the pilot addresses both sales productivity and strict confidentiality needs. The results could serve as a reference for secure AI implementation across the broader financial services industry and other highly regulated sectors.