CAC Corporation and DeepSearch Co., Ltd., a South Korean financial AI company, have concluded a partnership agreement regarding agent-type AI business for financial institutions in Japan.
Through this partnership, the two companies will provide Japanese financial institutions with an “agent-type AI” platform in which AI agents perform financial operations themselves.
In recent years, Japanese financial institutions have shown increasing interest in using “agent-type AI,” which goes beyond traditional AI tools that assist human staff, to autonomously lead AI agents in tasks such as deal sourcing (identifying potential mergers, acquisitions, and investment opportunities that align with strategic objectives), corporate analysis, and document creation.
Also Read: Rebellions and ai& Partner to Bring Energy-Efficient AI Inference Infrastructure to Japan
On the other hand, in financial operations that handle customer and transaction information, it is necessary to build an environment that complies with the security requirements and financial regulations set by each financial institution. To achieve this, a financial data infrastructure, agent platform, and financially specialized LLM are required, as well as the technical expertise to build and operate these in accordance with the system environment and requirements of each financial institution.
CAC has been involved in the construction and operation of systems for major companies, including financial institutions, for a long time and has a wealth of experience and knowledge in the Japanese financial industry. DeepSearch, on the other hand, reliably shares approximately 300 million financial data points per month with major financial institutions in South Korea and has a track record of building AI agents that can handle various financial operations such as investment, research, M&A, and corporate valuation.
By concluding this agreement, both companies will combine their respective strengths and aim to provide an agent-type AI environment to Japanese financial institutions.
Source: PRTimes


