Meta-structural intelligence: AI as architect of organizational meta-structures

We introduce Meta-Structural Intelligence (MSI), a theoretical framework that elevates AI systems to the role of architects of organizational meta-structures. Departing from traditional theories of organizational design that confine artificial intelligence to decision-support or control, MSI frames AI as Metacoders that proactively generate, formalize and propagate design blueprints and executable governance ontologies across firms and industries. We theorize two key mechanisms (algorithmic isomorphism and generative institutionalization) to explain how AI-driven templates can accelerate field-level convergence, reshape hierarchies and blur firm boundaries. In contrast to existing paradigms, MSI uniquely establishes AI's macro-level agency that co-designs organizational meta-structures in transforming organizational design in industries. This theory reframes AI's role and argues that AI not only influences behavior and decisions but also co-designs the foundational meta-structures of organizations and industries. Hence, the contributions in this research are (1) MSI, a mid-range theory that repositions AI as a macro-level architect of organizational form rather than a micro-level task agent; (2) three original constructs (AI Metacoders, Structural Templates and Design Ontologies); (3) two generative mechanisms (algorithmic isomorphism and generative institutionalization); and (4) six testable propositions that connect template diffusion driven by AI to changes in firm boundaries, authority structures and industry-wide institutional logics.