Hallucinario Inc. (Yokohama City, Representative Director: Masayuki Anekawa) will launch "Hallucinario," a trust-building service that adds M365/Entra ID authentication, access restrictions, secret information management, monitoring, data handling, and responsibility demarcation to small-scale business apps created with AI tools and low-code/no-code development, on July 6, 2026.
The service aims to prepare apps that function well in the field but are difficult for companies to approve for use, making them suitable for internal deployment within a limited scope.
Providing a trust-building service for on-site apps that cannot be deployed internally for various reasons
Background: "It works," but companies can't use it
The spread of generative AI and low-code/no-code development has enabled field staff and small teams to create business apps in a short period. However, even if these apps function effectively for business operations, those lacking authentication, access restrictions, secret information management, monitoring, and clear responsibility demarcation prevent IT and management departments from having the necessary information to approve their use, leading to a situation where they cannot be deployed internally.
An app working and a company being able to approve its continuous use are separate matters. Without organized information on users, data, secret information, monitoring, responsibility scope, and change rules, the field staff will feel "it works but can't be used," and IT departments will judge, "we cannot accept it because the conditions are unclear."
What is Hallucinario?
"Hallucinario" is a service that adds the necessary trust-building layer for company use to business apps created in the field.
Its purpose is not simply to host the app in the cloud. It is to provide explainable information on usage conditions, technical status, operational responsibilities, and residual risks, thereby preparing the necessary materials for companies to make acceptance decisions.
AI/Scanners assist in the confirmation process, but acceptance eligibility, scope, and estimates are not automatically determined. Final decisions are made by humans after representative review.
Trust-building elements to be added
Service Details
Overview
M365/Entra ID Authentication
Single tenant login, access restrictions by specified user/group
Secret Information Management
Migration of API keys, DB connection strings, OpenAI keys, etc., to external management
Monitoring and Incident Response
Setup of health endpoints, external monitoring, and basic alerts
Data Handling
Classification and documentation of users, purposes of use, and data handled
Responsibility Demarcation
Clarification of our and the customer's responsibility scope as acceptance evidence
30-Day Coverage
Monitoring, incident response, and inquiry support for 30 consecutive days from the acceptance date
Deliverables
The results of the trust-building process are organized into the following three deliverables:
Acceptance Evidence: A set of documents for acceptance decision, organizing information on users, data classification, secret management policy, monitoring configuration, responsibility demarcation, excluded items, and residual risks.
Accepted Version: A snapshot recording the technical configuration and state at the time of acceptance (Git SHA, container image digest, ACA revision, authentication method, monitoring settings, etc.).
30-Day Coverage: A period of monitoring, incident response, and inquiry support for 30 consecutive days after deployment.
Target Apps
Category
Target
Framework
Next.js (TypeScript), FastAPI (Python). Other stacks available upon consultation.
Configuration
Solo (single app) or Duo (front-end + back-end, up to 2 services)
Execution Environment
Azure Container Apps (single region, Public HTTP)
Authentication
Azure OpenAI (limited to simple inference processing, excluding RAG, etc.)
Acceptance Unit
1 app / 1 accepted version / 30 consecutive days
Target users include solo developers, field departments (business divisions, sales planning, CS, marketing, etc.), production companies, and individuals whose internal deployment is stalled due to non-functional requirements.
Standard Exclusions
The following conditions are outside the scope of the standard v1. Feasibility will be determined on a case-by-case basis with a custom quote.
Complex network/distributed configurations (VNet, private endpoint, 3+ tiers, microservices, jobs/worker/queue, WebSocket)
Advanced AI configurations (RAG, agent, tool calling, memory)
Complex permission management (RBAC/ABAC, approval workflows)
Large-scale infrastructure (WAF, CDN, Front Door, multi-region)
Audited/highly regulated data (company-wide SOC2/ISMS, My Number, PCI, payment cards, large volumes of medical/biometric data)
Please note that Hallucinario's suitability assessment is for determining fit, coverage, estimates, and plan compatibility, and is not a diagnostic or audit evaluating security safety.
Pricing
Pricing is based on a cumulative calculation method combining add-ons such as authentication, AI, and DNS with the core process (suitability diagnosis to 30-day coverage, ¥225,000) common to all projects.
Plan Name
Price (excl. tax)
Contents
Quick Check
¥29,500
Suitability diagnosis and trust-building gap report. Implementation work not included.
Small Start
¥295,000~
Core process + M365 authentication. Minimal trust-building for small-scale internal use.
Regular Trust
¥925,000~
All features of the Small Start plan + extended authentication + Duo + Azure OpenAI + AI Safety + Storage.
Enterprise
¥1,455,000~
All features of the Regular Trust plan + customer Azure environment support + DNS + security checklist compliance.
Azure (cloud) usage fees are separate and borne by the customer. If there are constraints such as Azure Policy in the customer's Azure environment, an environment surcharge (¥80,000–¥200,000 per item) will be diagnosed and presented in advance.
Free Online Seminar
A free online seminar will be held to explain the service's scope, pricing, and the process from application to trust-building completion in 60 minutes.
Item
Content
Title
Why Companies Cannot Accept Business Apps Created with Generative AI
Date and Time
July 13, 2026 (Mon) 14:00–15:00
Format
Online
Participation Fee
Free
Speaker
Masayuki Anekawa (Representative Director, Hallucinario Inc.)
Application
https://hallucinar.io/#seminar
Application Method
Applications will be accepted via email after attending the seminar.
Representative's Comment
The rapid advancement of generative AI has enabled the field to create business applications without relying on AI vendors or system integrators. However, an app working and a company being able to deploy and continuously operate it as an asset are different matters. Without clear authentication, access restrictions, secret management, monitoring, and responsibility demarcation, neither the field staff nor the IT department can move forward.
Hallucinario is a service designed to prepare business value generated in the field into a state that companies can accept, without discarding that value.
Masayuki Anekawa, Representative Director, Hallucinario Inc.
Company Overview
Item
Content
Company Name
Hallucinario Inc.
Representative Director
Masayuki ANEKAWA
Location
Yokohama City, Kanagawa Prefecture
Business Description
Adding a trust-building layer to on-site created apps
Company URL
https://corp.hallucinar.io
Contact
Contact for Inquiries Regarding This Matter
Hallucinario Inc.
The representative, with over 10 years of engineering experience, founded the company independently in Yokohama City, Kanagawa Prefecture in February 2025. The company operates the trust-building layer addition service for on-site created apps, "Hallucinario."
* Information contained in this release is current as of the date of announcement. Service details and prices are subject to change without notice.
FACT BOX
- Source: PR TIMES
- Category: サービス開始
- Organizations: Microsoft (M365/Entra ID) / Azure / Azure OpenAI