SoftBank Corp. Builds Multi-AI Agent Platform for
its Large Telecom Model, a Generative AI for Telecom,
and Commences Verification of Autonomous Operations

SoftBank Corp. (President & CEO: Junichi Miyakawa, "SoftBank") announced it built a multi-AI agent platform within its Large Telecom Model (LTM)^*1, a generative AI foundation model for the telecommunications industry, enabling multiple AI agents^*2 to work in coordination to autonomously carry out processes ranging from analysis and decision-making to execution. SoftBank also commenced verification of the multi-AI agent platform in base station integration operations as part of its efforts to automate communication network operations.

SoftBank will continue to promote greater efficiency and sophistication in its communication network operations through the LTM while steadily advancing the transition to an AI-driven autonomous operational model.

[Notes]

1. *1
   For more details on the "Large Telecom Model" (LTM), please refer to the press release dated March 19, 2025, titled "SoftBank Corp. Develops a Foundational Large Telecom Model (LTM)."
2. *2
   An AI agent is a system that plans the steps required to achieve a given objective and autonomously executes multiple processes while utilizing external tools and data as needed.

Overview of the multi-AI agent platform

The newly developed multi-AI agent platform consists of multiple business-specific AI agents optimized for their respective roles that work in mutual coordination, and performs a series of operations ranging from analysis and decision-making to execution that were traditionally carried out separately by human operators. Each AI agent hands over the results of its detection and analysis to another AI agent, which then makes decisions and carries out appropriate responses. By exchanging information and collaborating with one another as necessary, the agents autonomously process the entire operation as a single continuous workflow.

Through coordination among AI agents specialized for individual tasks and roles, the platform aims not only to automate discrete tasks but also to achieve advanced automation across the entire network operations domain. This approach is expected to prevent the personalization of operational processes that previously required skilled personnel judgment and task handovers, and improve response speed through faster decision-making.

SoftBank will gradually advance the autonomy of operations across a wide range of areas beyond base station integration, including fault response, traffic optimization, quality improvement, and equipment maintenance, by proactively leveraging AI. Based on insights gained from real world operations, SoftBank will further enhance its LTM and AI agents to improve network stability and reliability, enabling faster and more flexible service delivery and contributing to the evolution of next generation networks.

Ryuji Wakikawa, Vice President, Head of the Research Institute of Advanced Technology at SoftBank Corp. said: "SoftBank has enhanced its LTM based on the expertise in network operations that it has cultivated over the years. With the establishment of the multi-AI agent platform, we will evolve network operations beyond analysis and recommendations to AI driven operations that carry out decision-making and execution. Going forward, we will continue to leverage cutting edge AI to pursue the provision of high quality and highly reliable network services."

• SoftBank, the SoftBank name and logo are registered trademarks or trademarks of SoftBank Group Corp. in Japan and other countries.
• Other company, product and service names in this press release are registered trademarks or trademarks of the respective companies.