This paper develops the MIRACLE framework, a stage-based conceptual model that examines how human and artificial cognitive agents interact across the entrepreneurial process, how artificial intelligence (AI) transparency affects entrepreneurial agency and how heterogeneous knowledge bases and distributed agency influence these dynamics across venture creation stages.
The paper adopts a conceptual, theory-building approach that integrates entrepreneurship research, philosophy of science, human–computer interaction and AI transparency studies to examine transparency of AI in depth and detail, derive stage-specific modes of human–AI interaction and discuss associated transparency requirements.
The interdisciplinary analysis highlights that, beyond a binary opacity-transparency distinction, AI transparency is contextually situated, relational and distributed, and it operates at both individual and collective levels. This finding indicates a spectrum of human-technology relationships that varies across cognitive profiles, knowledge bases and resources, as well as across different stages of entrepreneurial journey. The resulting MIRACLE framework shows how differences in entrepreneurial expertise and AI familiarity shape interaction patterns with AI tools across stages. It highlights that transparency of AI tools varies by user expertise across entrepreneurial stages thereby influencing how entrepreneurs use AI outputs.
Building on an interdisciplinary synthesis, this paper introduces the MIRACLE framework, a stage-based model that maps human–AI interaction patterns across entrepreneurial stages, accounting for heterogeneous knowledge bases and varying levels of AI familiarity. It advances existing work by specifying stage-specific transparency requirements while integrating ethical, privacy and bias considerations. Rather than treating AI use or transparency in aggregate, the MIRACLE framework reveals why AI-supported outcomes vary across entrepreneurial stages by linking them to the relational and distributed nature of AI transparency and offers a novel, interaction-aware perspective on AI-augmented entrepreneurship. By breaking down human–AI interactions into stage-specific components, the framework also generates actionable insights for entrepreneurs to navigate AI tools more effectively.
