What is the launch date of Cabosia's SEISAKU DB product?

The SEISAKU DB was launched by Cabosia on June 9, 2026.

Which company released the SEISAKU DB for analyzing policy and spending data?

Cabosia Co., Ltd. released the SEISAKU DB on June 9, 2026.

What specific data connections does Cabosia's SEISAKU DB provide?

SEISAKU DB links parliamentary debates, programs, grants, and procurement to companies.

How can AI systems access data from Cabosia's SEISAKU DB?

AI systems can access SEISAKU DB via APIs and MCP connectors directly.

Cabosia Launches 'SEISAKU DB' (Policy DB) for AI-Driven Analysis of Corporate Policy Connections and Public Spending

Q: What functionality does the SEISAKU DB offer regarding policy and budget analysis?

SEISAKU DB enables visualization of gaps between policy discussions and budget allocations.

Published Jun 9, 2026 6:00 PM ・ Updated Jun 14, 2026 2:50 PM ・ 4 min read ・ Source: PR TIMES

On June 9, 2026, Cabosia Co., Ltd. launched 'SEISAKU DB,' a policy data infrastructure that links Japanese parliamentary debates, administrative programs, grants, and public procurement to companies. It provides APIs and MCP connectors for direct AI access, enabling the visualization of gaps between policy discussions and budget allocations.

In June, when the 'Basic Policies for Economic and Fiscal Management' (often called the blueprint for the budget) are finalized, Cabosia Co., Ltd. (HQ: Tokyo, CEO: Riku Koike) released 'SEISAKU DB' (Policy DB) on June 9, 2026. This infrastructure links Japanese parliamentary records, administrative programs, subsidies, public procurement, and economic security certifications to listed companies.

The main feature is the 'visualization of the gap between discussion and budget.' Since January 2024, AI has analyzed 269 parliamentary comments related to AI (linked to 295.8 billion yen in budget) and 2,667 comments related to economic security (linked to 2.309 trillion yen). Commercial databases that quantify the difference between discussion volume and budget allocation per theme are limited.

Simply by connecting to MCP connectors for ChatGPT or Claude, AI can directly reference policy data. 23 APIs and 31 MCP tools are already available. The pricing is on par with the sibling services 'EDINET DB' and 'Real Estate DB' (Pro plan from 4,980 yen/month; free during beta).

Public Spending of 24 Trillion Yen – Tracking 'Which Company Receives How Much Tax Money' by Corporate ID We have aggregated 24 trillion yen in public spending for 1,702 companies over 5 years. By mapping administrative program review expenditures to the master corporate ID list (Corporate Number), we enable users to track funds by company.

Policy Themes × Involved Companies – Drilling down from theme to recipient For 20 policy themes (e.g., semiconductors, defense, GX, space, storage batteries, quantum, medical, economic security, AI), we list companies that received public funds with the corresponding amounts.

Consolidating 4 Public Funding Sources under one Corporate ID Users can view integrated data on one screen: subsidy selection (gBizINFO), government contracts/bidding (Procurement Portal), government procurement/commissioning (NEDO, etc.), and competitive research funds (AMED/KAKEN).

FACT BOX

Source: PR TIMES
Category: New Product
Products / services: SEISAKU DB / EDINET DB

Cabosia Launches 'SEISAKU DB' (Policy DB) for AI-Driven Analysis of Corporate Policy Connections and Public Spending

⚡ Key Points

Policy Themes × Involved Companies – Drilling down from theme to recipient For 20 policy themes (e.g., semiconductors, defense, GX, space, storage batteries, quantum, medical, economic security, AI), we list companies that received public funds with the corresponding amounts.

Consolidating 4 Public Funding Sources under one Corporate ID Users can view integrated data on one screen: subsidy selection (gBizINFO), government contracts/bidding (Procurement Portal), government procurement/commissioning (NEDO, etc.), and competitive research funds (AMED/KAKEN).

FACT BOX

Editorial & Verification Standards

FAQ

Cite this article — HOW TO CITE

AI CRAWLER ACTIVITY