Sumitomo Mitsui Card and ELYZA launch GenAI for customer support

Sumitomo Mitsui Card and ELYZA, a company promoting the social implementation of Large Language Models (LLMs), have launched the production use of generative AI, powered by ELYZA’s Retrieval-Augmented Generation (RAG) technology, at Sumitomo Mitsui Card’s contact center.

The implementation of this generative AI will enhance the speed of operator responses, strengthen inquiry channels, and make it more convenient for customers to contact Sumitomo Mitsui Card.

Background of the initiative

In 2023, cashless payments accounted for a record high of 39.3% of total consumption. As cashless payments continue to expand, Sumitomo Mitsui Card has received over 5 million new applications in fiscal 2023 (including 3 million applications for proprietary cards), demonstrating its strong customer base. As a result, Sumitomo Mitsui Card’s contact center receives over 500,000 inquiries per month regarding usage and other matters, making it imperative to improve both the quality and quantity of responses.

Recent advancements in generative AI have contributed to the sophistication and efficiency of contact center operations. Leveraging generative AI can also enhance the quality of customer support. ELYZA’s generative AI solution, with its sophisticated utilization of RAG technology and robust architecture supported by its proprietary LLM infrastructure, provides a secure platform suitable for a high-security environment like a contact center, leading to its selection by Sumitomo Mitsui Card.

Details of the implemented generative AI

This generative AI automatically generates draft responses to inquiries received by Sumitomo Mitsui Card’s contact center using Retrieval-Augmented Generation (RAG) technology. This technology searches internal data (Exploration AI) and then generates draft responses (Response Generation AI). The AI has been in use for email response operations at the contact center since the end of June 2024. Within the year, Sumitomo Mitsui Card plans to expand the use of similar generative AI to chat inquiries, potentially integrating it into business applications via APIs. Ultimately, this implementation is expected to reduce the time required for operators to respond to customer inquiries by up to 60%, significantly improving productivity.

This generative AI is provided through “ELYZA App Platform,” ELYZA’s LLM implementation platform.

Please follow us to read more about Finance & FinTech in Japan, like hundreds of readers do every day. We invite you to also register for our short weekly digest, the “Japan FinTech Observer”, on Medium or on LinkedIn. Our global Finance & FinTech Podcast, “eXponential Finance” is also available through its own LinkedIn newsletter, or via our Podcast Page.

Should you live in Tokyo, or just pass through, please also join our meetup. In any case, our YouTube channel and LinkedIn page are there for you as well.