AI-Powered Microlearning Ecosystem for Hyper-Personalized Workforce Development Qorus-NTT DATA Innovation in Insurance Awards 2026

Concept and Objectives

The project introduces an AI-based microlearning system that delivers automatically generated, personalized learning units directly into employees’ daily work environments. Instead of relying on static training catalogs, the platform uses internal company knowledge across formats (documents, guidelines, product materials, etc.) to generate dynamic learning content in real time. The objective is to:

• Increase learning frequency and relevance

• Reduce production and coordination costs of traditional training

• Enable faster qualification cycles

• Build transparent, data-driven skill insights

Reasons Behind

The insurance industry faces increasing regulatory complexity, rapid product innovation, demographic change, and technological disruption. Traditional eLearning formats are too slow, too generic, and rarely embedded in real workflows. ERGO’s ambition to become a digital leader by 2025 required a scalable and intelligent learning solution aligned with this strategy.

State of Competition

Most corporate learning platforms remain content libraries or LMS-based systems requiring manual content production and periodic updates. Even AI-supported platforms often recommend pre-produced courses rather than generating hyper-personalized content dynamically from proprietary knowledge bases. The chunkx–ERGO solution differentiates itself through real-time content generation, continuous adaptation, and full integration into daily work channels.

Sources of Inspiration

The project draws inspiration from adaptive learning systems, advanced AI models including RAG architectures, consumer-grade personalization logic found in streaming platforms, and workflow-integrated productivity tools, merging these concepts to deliver a unique learning experience.

Departments Involved

• HR / Learning & Development

• Innovation & Digital Transformation

• IT / Architecture

• Data & AI Teams

Main Results So Far

The pilot engaged 199 registered users with over 70% active participation, generating 1,625 learning interactions and accumulating 174 learning hours within four months. On average, users completed six learning interactions per month, reducing content production and coordination efforts while eliciting strong qualitative feedback. Users also voluntarily created learning content, demonstrating the system’s adoption and engagement potential. The pilot phase concluded successfully, with plans for organizational scaling in 2026 already underway.