Document Processing · Enterprise Knowledge Search
Polimill needed to reduce administrative burden across fragmented Japanese municipal systems so public employees could find prior decisions, prepare assembly responses, and deliver services without each municipality building separate AI tools.
Reported impact
Polimill created a common public-sector AI platform used across Japan, sped up its own product development, improved knowledge reuse across municipalities, and showed that less-experienced staff can draft stronger policy work with AI support.
Polimill built QommonsAI as a shared generative AI foundation for municipal work in Japan. The platform started by standardizing assembly minutes and other administrative records that were previously scattered across municipalities, then applied AI to add metadata and create a searchable cross-municipality knowledge base. QommonsAI now supports workflows such as assembly response, public services, social welfare, and legal search for about 1,050 municipalities and roughly 550,000 public employees. OpenAI GPT models provide the reasoning layer, while QommonsAI adds public-sector controls so administrators can inspect feature-usage history and restrict model availability according to policy. Polimill also uses Codex across its own development lifecycle—from requirements and code-consistency checks to implementation and testing—which accelerated validation cycles with municipal users and helped the company evolve the system toward a broader public OS and future super-agent marketplace called Qommons ONE.
Assembly minutes and administrative records from municipalities are standardized, enriched with metadata, and made searchable through QommonsAI. Public-sector workers query that shared knowledge base through GPT-powered interfaces, while Polimill’s internal team uses Codex to rapidly build and refine the platform from municipal feedback.
"About 1,050 municipalities and about 550,000 public employees across Japan now use it."
"Polimill collected and standardized assembly minutes from across Japan, then used AI to add metadata and build a high-precision search foundation"
"QommonsAI includes operational controls that let administrators review feature usage history and limit which models are available according to organizational policy."
"The company adopted Codex across the workflow... implementation, and testing."
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
OpenAI Sample Oct 8, 2026 Cooley Capital markets involve huge amounts of information that need to be synthesized and analyzed. "For our lawyers, it’s always very busy," explains David Wang, Chief Innovation Officer at Cooley. Cooley is an international law firm with a reputation for helping companies navigate capital markets and… ChatGPT Work · GO Public OpenAI Sample Oct 8, 2026 Gilbert + Tobin Gilbert + Tobin needed to improve research, drafting, analysis, and operational workflows across a law firm without compromising client confidentiality, governance requirements, or human accountability. 87% active usage among enabled ChatGPT users ChatGPT Enterprise · ChatGPT Work · Codex OpenAI Sample Aug 20, 2026 Stampli Stampli needed to launch complex new products and keep product knowledge current without waiting on scarce design, contractor, and operations bandwidth. 243 modeled active role-hours compressed to about 77 for the Deep Finance launch ChatGPT Work · Codex · Deep Finance 