Didact: A Cross-Domain Capability Discovery System for Defence

Didact is a prototype system that integrates Australian defence reports, policy documents, and research publications into a unified knowledge graph to help policymakers discover defence capabilities faster. The system uses retrieval-augmented generation (RAG) and natural language conversations to surface fragmented information across heterogeneous sources, with an interactive Evidence Rail for visualizing source relationships.

Didact addresses a genuine operational challenge in defence policy: the fragmentation of critical information across disparate repositories and formats. Policymakers require rapid access to evolving research relevant to strategic priorities, yet current workflows rely on manual monitoring of siloed data streams. This prototype demonstrates how artificial intelligence and knowledge graph technology can consolidate unstructured information into actionable intelligence.

The system's architecture combines public defence reports, policy documents, and academic research into a unified interface with natural language query capabilities. The Evidence Rail feature—visualizing retrieved evidence and source relationships—acknowledges a critical need in policy work: auditability and transparency in how conclusions derive from source material. This design choice reflects lessons from institutional requirements that decisions be traceable and defensible.

Beyond defence applications, Didact's framework has broader implications for sectors where knowledge fragmentation creates operational friction. Healthcare, financial regulation, and infrastructure planning similarly face challenges synthesizing scattered information sources. The academia-industry collaboration model suggests a pathway for domain-specific AI applications where specialized knowledge graphs outperform general-purpose language models.

The system's success metrics—output quality and runtime performance—matter for real-world deployment. Policymakers will adopt tools only if they deliver faster, more accurate capability discovery than existing workflows. As defence institutions increasingly recognize that strategic advantage depends on information synthesis speed, systems like Didact become infrastructure investments rather than experimental prototypes.

• →Didact consolidates fragmented defence information sources into a unified knowledge graph with natural language interfaces for policy workflows.
• →The interactive Evidence Rail ensures transparency by visualizing source relationships, addressing auditability requirements in policy decisions.
• →The system demonstrates how domain-specific RAG pipelines can outperform general AI models when tackling fragmented knowledge across specialized sectors.
• →The academia-industry collaboration model suggests scalable pathways for deploying AI systems in other knowledge-intensive institutional contexts.
• →Real-world adoption depends on demonstrating faster capability discovery and output quality compared to manual monitoring and existing workflows.