Core-Pilot-Knowledge

System Overview

Core-Pilot Knowledge is a security‑focused add-on for the PrivateAI-Platform that provides a RAG foundation and access control mechanisms for safely leveraging internal knowledge. It strictly manages which information can be referenced by each department, role, or project, ensuring that the system generates responses only within the user’s authorized scope—minimizing risks of unauthorized access or information leakage. Designed in alignment with the AI risk‑management principles required by ISO 42001, it enables enterprises to clearly define who can access which information through which pathways, making it suitable for organizations aiming to operate generative AI safely in production environments.

Key Features

• Manage sensitive data within your own AWS environment, operating securely via API integration with LLMs.
• Enables access control that meets the AI risk management requirements of ISO/IEC 42001.
• Prevents unauthorized reference to confidential information through permission settings by department and position.
• Securely leverage the latest technologies through integration with multiple LLMs such as OpenAI and Gemini.
• Operates high-precision internal AI search (RAG) both safely and effectively.

Implementation Cost (Reference)

Core-Pilot-RAG is offered as a high-end solution for companies with particularly high AI security requirements. The following are standard implementation cost estimates.

• Deployment to AWS Environment: 1,200,000 JPY (Excl. Tax)
• Annual Support: 240,000 JPY (Excl. Tax)