Elith List of Products and Services

"GENFLUX Security" is a security service that visualizes shadow AI and enables safe AI utilization. This service is not intended to prohibit AI, but rather to allow companies to use AI with confidence by visualizing actual usage, applying necessary rules, and leaving explainable traces. It can detect the transmission of personal information, Japan-specific information, payment and authentication information, and corporate secrets before they are sent, allowing for appropriate responses and corrective suggestions based on the situation. 【Features】 ■ Visualization of AI usage: Understanding employee usage, detecting unauthorized AI, checking the dashboard ■ Control of AI usage: Blocking prohibited AI services, real-time control of confidential information transmission ■ Recording of AI usage: Saving usage logs, recording who sent what and when *For more details, please download the PDF or feel free to contact us.

"GENFLUX Security" is an AI security platform that visualizes usage patterns, applies necessary rules, and leaves explainable evidence, enabling companies to utilize AI with peace of mind. It features three key points: visualization of employee AI usage and detection of unauthorized AI services, real-time control to block prohibited AI services and prevent the transmission of confidential information, and the recording of usage logs as evidence. [Features] ■ Visualization of usage patterns ■ Real-time usage control ■ Recording and storage of usage logs *For more details, please download the PDF or feel free to contact us.

"GENFLUX" is an AI safety platform that supports everything from the introduction of AI to operation, evaluation, and security. It provides powerful assistance to safely and reliably deploy AI in areas where failure is not an option. It offers a comprehensive solution from shadow AI countermeasures and quality management of AI and agents to verification, correction, and the construction of custom AI, thereby supporting the safe utilization and operation of AI. 【Features】 ■ Comprehensive support from AI introduction to operation, evaluation, and security ■ Implementation of shadow AI countermeasures and quality management of AI and agents ■ Support from verification to correction and custom AI construction *For more details, please download the PDF or feel free to contact us.

This document is an explanatory material that summarizes the standards and know-how for companies to practically and gradually build and demonstrate AI governance. The importance of objectively proving compliance with public guidelines in their AI utilization and operational systems, and ensuring trust with customers, is increasing. It explains the outline of the new standards established by the Japan Deep Learning Association (JDLA) and key points for establishing a system to obtain certification. [Contents] ■ Overview of AI Governance C Certification ■ Main requirements for certification ■ Support phases and comprehensive support menu *For more details, please download the PDF or feel free to contact us.

By reading this document, you will gain core know-how for security operations that supports everything from vulnerability assessment to patch application, combining the speed of AI processing with critical human judgment. In daily code changes and pre-release validation, there is a demand for a system that not only detects vulnerabilities but also rapidly generates and applies patches. This document introduces the background and methods for protection that leverage existing environments. 【Contents】 ■ A seamless support flow from detection to correction decision ■ Risk assessment and presentation of assessment materials ■ Integration configuration with GitHub and CI/CD environments ■ Expansion to cloud LLMs and local LLMs *For more details, please download the PDF or feel free to contact us.

The "GENFLUX Code" is an AI coding agent that allows for code investigation, editing, execution, and Git operations within an internal environment. It places AI models, code, and execution permissions on the internal network, managing the scope of use and operation history. This product generates change proposals and validation results, while design, quality, and release decisions are made by developers. Since it operates entirely within the internal LAN, input data is not learned externally, and multiple developers can connect to a common environment without the need to individually install the model on each PC. 【Features】 ■ Sizing and performance optimization tailored to the number of users and amount of code ■ Appropriate model selection based on real-world task benchmarks ■ Provision of specialized models for your company through additional training using your own code ■ Design and provision of guardrails and agent controls that detect and block dangerous deletions, overwrites, and unauthorized operations *For more details, please download the PDF or feel free to contact us.

"GENFLUX Evaluation" is an evaluation and diagnostic service that "measures, understands, and improves" AI and web services. It conducts cross-sectional evaluations of response quality, performance, safety, red teaming, and compliance, organizing risks, failure factors, and points for improvement. The evaluation and diagnosis clarify weaknesses and recommended responses, providing support until improvements are made. It enables quality checks and sharing of decision-making materials at various phases, such as before implementation and during operation. 【Services Offered】 ■ Quality and safety evaluation of AI (verification of responses and behaviors of generative AI, RAG, AI agents, etc.) ■ Web application and API vulnerability diagnosis (verification of authentication, authorization, input/error responses, data storage/caching, etc.) *For more details, please download the PDF or feel free to contact us.

"GENFLUX" is an AI agent platform for businesses that builds AI agents to a state where they can be safely operated in production. It is based on AI agents with proven operational performance created from previous AI development, enabling the implementation of "production-ready" solutions customized to meet the business requirements of each company. The standard Safety features visualize and control risks, allowing for easy management and updates after delivery. 【Features】 ■ Risk visualization and control through standard Safety features ■ Automatic quality assessment using quantitative metrics ■ Customized implementation according to business requirements ■ Default encryption during data transmission *For more details, please download the PDF or feel free to contact us.

This product is a local AI infrastructure that incorporates security, governance, and agent control. It operates in a completely closed network environment, ensuring that confidential data is not sent externally, achieving zero risk of data leakage and zero cost for LLM inference. It comes standard with RAG and agent functionalities, allowing for rapid deployment without the need for development. It integrates four security solutions and supports customization and fine-tuning according to business requirements. 【Features】 ■ Rapid deployment possible with standard RAG and agent functionalities ■ Risk visualization and control with standard Safety features ■ Multiple AI functionalities available within a closed network ■ Customization to meet business requirements *For more details, please download the PDF or feel free to contact us.

This service is a security infrastructure service for the AI era that visualizes usage conditions, applies necessary rules, and leaves explainable evidence, allowing companies to utilize AI with confidence. It supports understanding employees' AI usage, detecting unauthorized AI services, controlling real-time transmission of confidential information, and storing usage logs. It also addresses pre-transmission detection and control of personal information, Japan-specific information, payment and authentication information, and corporate secrets. 【Features】 ■ "Visualization" that makes usage conditions clear ■ "Control" that applies necessary rules ■ "Recording" that leaves explainable evidence *For more details, please download the PDF or feel free to contact us.