Mobilus Launches New Generative AI-Powered 'RAG Chatbot' Function
Mobilus Corporation (HQ: Shinagawa-ku, Tokyo; President: Tomohiro Ishii), a developer of CX solutions for contact centers, announced the release of 'MOBI BOT RAG Answer Management' on May 26, 2026. This new feature for the 'MOBI BOT' chatbot utilizes Retrieval-Augmented Generation (RAG) to enhance response accuracy, promote user self-service, and reduce the workload on human operators.
'MOBI BOT RAG Answer Management' automatically generates answers based on multiple knowledge sources such as FAQs. It includes anti-hallucination measures to prevent the generative AI from creating unreliable answers, while also supporting cases where generative AI responses are not applicable. This streamlines the user experience by eliminating the need to compare multiple knowledge sources, achieving safe and rapid self-resolution.
Mobilus aims to support next-generation contact center operations that improve customer satisfaction and experience (CX) through enhanced operational efficiency and response quality.
*1 RAG (Retrieval-Augmented Generation): A technique that improves response accuracy by combining external information retrieval with text generation by Large Language Models (LLMs).
Background
According to an announcement by the Ministry of Internal Affairs and Communications in April 2026, the population aged 15-64 in Japan continues to decline, reaching 73.54 million (a decrease of 200,000 year-on-year). Labor shortages due to aging demographics are critical in the contact center industry, creating challenges in recruitment, training, and meeting increasingly diverse customer needs. Consequently, there is an urgent demand to develop and implement generative AI services to maintain high service quality with limited resources.
To address these issues, Mobilus developed 'MOBI BOT RAG Answer Management' to automatically generate candidate responses based on various knowledge bases such as FAQs.
New Function Overview
- RAG Chatbot 'MOBI BOT RAG Answer Management' Unlike traditional machine learning chatbots that present candidates from similar FAQs, requiring users to search for the desired information themselves, this new function instantly creates and presents the exact response a user is looking for by synthesizing multiple knowledge sources. This promotes user self-resolution and significantly reduces the need for human intervention.
<Feature Details> - Hallucination Suppression If there is no knowledge base to support a response, the system does not use the AI's output and instead presents a pre-defined 'correct guidance message' to the user. This mitigates the risk of conveying false information caused by AI hallucinations.
FACT BOX
- Source: PR TIMES
- Category: New Product
- Products / services: MOBI BOT / MOBI BOT RAG Answer Management