This paper proposes knowledge engines as a framework for understanding how intelligent systems—both human and artificial—systematically discover, integrate, and generate knowledge. We argue that history’s greatest scientific minds functioned as knowledge engines, processing information through iterative cycles of ingestion, analysis, synthesis, and communication, guided by curiosity and willingness to challenge established beliefs. We propose a taxonomy of nine integrated capabilities—ingestion, digestion, analysis, calculation, comparison, connection, association, analogy, and multimodal communication—that any serious knowledge engine must combine systematically. The argument is deliberately integrative: achieving ambitious research goals requires orchestrating all nine capabilities within a durable infrastructure, not merely scaling up foundation models alone. We present CopernicusAI as a working proof-of-concept of the knowledge engine framework, demonstrating feasibility through a fully deployed system with 59,499 indexed research papers, 692 process diagrams across five scientific disciplines, an operational knowledge graph, vector search, and retrieval-augmented generation (RAG) capabilities. To address reviewer requests for empirical content, we report a preliminary retrieval pilot (30 queries, lexical TF-IDF baseline, frozen corpus of 59,499 documents) yielding a mean nDCG@10 of 0.545—a useful early benchmark, with dense-vector evaluation deferred pending infrastructure resumption. While extensive validation remains necessary, the system demonstrates that the knowledge engine framework can be instantiated in practice.
Gary Welz (Mon,) studied this question.