AI Briefing
KO

Polimill Builds Public AI Infrastructure Supporting 1,050 Japanese Municipalities with OpenAI Technology

·2026.08.31 16:00

Key point

Polimill has built the next-generation public OS 'QommonsAI' using OpenAI technology to support 1,050 municipalities in Japan.

Details

Polimill has developed QommonsAI, a generative AI platform based on OpenAI technology that supports approximately 1,050 municipalities and 550,000 civil servants in Japan. This platform provides AI features specialized for various public service areas, including council responses, administrative services, social welfare, and legal search.

Integration and Utilization of Distributed Administrative Data

The biggest obstacle to introducing generative AI in the public sector was the fragmentation of data due to differing workflows and document formats across municipalities. Polimill collected and standardized council meeting minutes from across the country, then added metadata via AI to build a high-precision search foundation that operates across municipalities and time periods.

This structure expanded beyond councils to other administrative areas such as welfare and law. QommonsAI transforms distributed administrative information into knowledge usable in daily tasks, serving as a common foundation that all municipalities can use equally.

Balancing Security and Usability, and Improving Development Efficiency

To ensure essential information management and audit response in the public sector, QommonsAI includes operational control features that allow administrators to review feature usage history and restrict available models according to organizational policies. At the same time, the high recognition and versatility of GPT models contributed to lowering the initial adoption barrier for civil servants.

During the development process, Codex was introduced to automate the entire workflow from requirements definition and code consistency checks to implementation and testing. This allowed engineers to review AI-generated plans and focus on high-level decision-making, improving development speed by 3 to 5 times.

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.