The common occurrences of recent large-scale disasters (Great East Japan Earthquake, Kumamoto Earthquake, Noto Peninsula Earthquake, etc.) are unexpected locations, scales, and tsunami damage, and there is an urgent need to continue to build social systems that can respond to unexpected events.
The AI that realizes the autonomous distributed collaboration that is the subject of this research is based on large-scale language models (LLMs), so it has the potential to flexibly respond to the above challenges, and an image of this is shown in Figure 1. The outcomes aimed at by this study are as follows.
・Learn the hazard maps and disaster prevention plans of all local governments, regions, ministries, and infrastructure companies
・In addition to expected events, it is possible to generate unexpected events and conduct training using them as a background.
・Because it is possible to communicate in natural language, information sharing with residents and government can be carried out in real time.
Kato Y.(2020) A novel platform for efficiently collecting real-world data in cooperation with multi-vendor electronic medical records: CyberOncology, Precision Medicine No.9, p.34-38.
Kato Y.(2022) Effects to accumulate real-world data in the deployment of many medical facilities and future prospects., Precision Medicine No.9, p.48-51.
Kato Y.(2023) Efforts to utilize AI in CyberOncology and future prospects., Precision Medicine No.7, p.60-53.
Kato Y.(2024) et al. Collecting real-world data using generative AI, PHARMSTAGE Vol.23, No.11.
Kato Y.(2024) et al. Utilization of large-scale language model LLM in Cyber Oncology, Precision Medicine No.2, p.35-39.