Trust in AI deep dive: Explainability
When a company questions “how did the AI reach that conclusion?” about a critical AI-driven decision, the answer must be clear. Without it, the company cannot accept the decision or take responsibility for it. “Explainability” turns AI from something we simply trust into a partner we can confidently entrust with tasks.
Featured Content: A Journey Through History—Learning from the Past
Events that marked turning points in history raise questions that still resonate with today’s business leaders: the heliocentric theory, the Renaissance, the Titanic, the Silk Road, and the Age of Discovery. This series uses stories from the past as clues to understand the meaning of “trust” in the age of AI.
“Causal AI” answering the “why” behind the data
Technology introduction
Causal AI analyzes causal relationships in data and recommends concrete actions that help solve business challenges. By integrating multiple datasets and existing causal knowledge, it identifies factors that influence business outcomes and presents them through explainable evidence and visualized causal relationships.
Why it matters
The value of this technology lies in helping organizations understand which factors are influencing outcomes and determine what actions should be taken. Because it considers not only potential benefits but also possible side effects, it supports more practical and actionable decision-making. Rather than stopping at analysis, it helps organizations quickly translate insights into business results.
Example use case
Health improvement initiatives often fail to deliver consistent results because genetic factors and lifestyle habits are interconnected in complex ways, making it difficult to identify true causes through correlation analysis alone. By using Causal AI, genetic and lifestyle data from approximately 4,000 individuals can be analyzed to identify hidden causal relationships among genetic traits, dietary habits, and health conditions. As a result, organizations can discover previously unseen causal mechanisms—such as the impact of drinking habits on dietary behavior or the influence of taste preferences on BMI—and use these insights to support more personalized health improvement strategies.
“Knowledge Graph-Enhanced RAG” connecting knowledge at scale
Technology introduction
Knowledge Graph-Enhanced RAG enables AI to understand and utilize large volumes of enterprise information across organizational silos. By structuring regulations, manuals, logs, source code, and other enterprise data as a knowledge graph, it helps AI understand not only information itself but also the relationships between pieces of information, enabling more grounded answers and logical reasoning.
Why it matters
Beyond simply finding information, this technology helps organizations connect and interpret information to support decision-making. It can be applied to advanced enterprise Q&A, root-cause analysis using logs and incident records, software understanding, and automated generation of design documents. As a result, it reduces the effort required for investigation and analysis while enabling faster, evidence-based decisions.
Example use case
In modernization projects, understanding legacy systems with incomplete or outdated design documentation often requires significant time and expert effort. By applying Knowledge Graph-Enhanced RAG, organizations can automatically generate high-quality design documentation from source code and remaining design assets, even when source code contains limited comments. This helps improve productivity and supports digital transformation initiatives by making legacy systems easier to understand and modernize.