On the Definition of Intelligence
arXiv:2507.22423 · doi:10.1007/978-3-032-00800-8_1
Abstract
To engineer AGI, we should first capture the essence of intelligence in a species-agnostic form that can be evaluated, while being sufficiently general to encompass diverse paradigms of intelligent behavior, including reinforcement learning, generative models, classification, analogical reasoning, and goal-directed decision-making. We propose a general criterion based on \textit{entity fidelity}: Intelligence is the ability, given entities exemplifying a concept, to generate entities exemplifying the same concept. We formalise this intuition as \(\varepsilon\)-concept intelligence: it is \(\varepsilon\)-intelligent with respect to a concept if no chosen admissible distinguisher can separate generated entities from original entities beyond tolerance \(\varepsilon\). We present the formal framework, outline empirical protocols, and discuss implications for evaluation, safety, and generalization.
Accepted at AGI-25. v2: Enhanced mathematical rigor and conceptual clarity; terminology was refined from "category"/"sample" to "concept"/"entity", and the concept-fibre-entity relationship was clarified. v3: Added a reference to subsequent mathematical development in Similarity Field Theory (arXiv:2509.18218); main results and conclusions unchanged