Quantifying Digital Intelligence: An Info-Metrics Approach to Measuring Human–AI Interaction

In an era of accelerated digital transformation, interaction between humans and artificial intelligence has become a central
channel through which information is created, transmitted, and interpreted. Despite the growing role of conversational AI
systems, standardized frameworks for quantitatively assessing the informational efficiency of their outputs remain
underdeveloped. This paper proposes a conceptual info-metrics approach for measuring the efficiency of information
exchange between humans and AI as a novel way of interpreting digital intelligence. The approach is based on three
indicators: informational density, relevance, and redundancy, which are integrated into a unique efficiency index (E). The
planned study includes an analysis of interactions between two large language models and users across three types of
queries: factual, analytical, and creative. Info-metric indicators will be used to compare models and task types, as well as
to examine differences in informational efficiency. The paper serves as a preliminary communication that establishes the
theoretical foundation for the empirical validation of index E. The proposed approach is expected to contribute to the
development of reliable metrics for assessing digital intelligence and to provide a basis for applying info-metrics in the
evaluation of educational and business AI systems.