Hybrid AI for Explainable and Accurate Conversational Agents in eGovernment

2026-08-02

AI Summary

This study addresses the complexity of legal and policy consultation in e-government by proposing CHAI, a hybrid artificial intelligence architecture that uniquely integrates a large language model with a symbolic controller based on Dynamic Condition Response (DCR) graphs. The large language model serves as a natural language interface, while the DCR-based component formally models legal obligations, conditions, and administrative workflow logic, thereby ensuring conversational fluency, regulatory compliance, and interpretability. Validated through two real-world prototypes—COVID-19 policy consultation and case management for disability student grants—CHAI demonstrates high accuracy and strong explanatory capabilities in authentic e-government settings, offering a novel paradigm for trustworthy, explainable government dialogue systems.

Abstract

We present a so-called Conversational Hybrid AI (CHAI) architecture for building explainable and accurate conversational agents for eGovernment. We exemplify the architecture with a running prototype of a Covid-19 Chatbot based on a governmental guideline directed to citizens. We also describe an ongoing case on case management for supplementary grants for students with disabilities. We use large language models (LLMs) as a bounded conversational interface to a rule-based (symbolic AI) controller that executes a logical model expressing the provisions and obligations of the law and/or guidelines. As logical modelling language we use Dynamic Condition Response (DCR) graphs, a symbolic declarative process-modeling language developed with the aim to be able to express both deontic, defeasible and temporal logic properties, making it suitable for expressing both the rules of the law and the steps of the legal case management processes.

Problem

• Explainable AI
• Conversational Agents
• eGovernment
• Legal Reasoning
• Hybrid AI

Innovation

• Hybrid AI
• Explainable AI
• Large Language Models
• DCR Graphs
• eGovernment