Co-Creative AI Research Program

Enactive AI, Human–AI Co-Creation, and Adaptive Intelligence Under Structural Drift

My research investigates how intelligent co-creative systems maintain coherent behavior under structural drift—changes in the underlying structure of dynamic environments. Drawing on enactive cognitive theory and AI system design, I develop adaptive architectures that regulate perception, sense-making, action, and co-creativity over time.

The Co-Creative AI Research Program investigates how intelligent systems maintain coherent behavior in dynamic environments through interaction, regulation, and adaptive sense-making. The work spans more than a decade of interdisciplinary research by Nicholas Davis and collaborators at the intersection of human–AI interaction, co-creative AI, cognitive science, computational creativity, adaptive systems, and enactive cognition.

Traditional artificial intelligence has largely focused on static optimization, prediction, classification, and autonomous generation. In contrast, this research program explores intelligence as an ongoing process of interaction between agents and environments. Rather than treating cognition as isolated computation, the work examines how meaning, creativity, coordination, and adaptive behavior emerge through continuous participation within evolving systems.

This perspective draws heavily from enactive cognitive science, participatory sense-making, dynamical systems theory, and human-centered AI. Across the research collected here, these ideas are translated into computational architectures, co-creative systems, interaction frameworks, and adaptive AI platforms capable of operating under conditions of uncertainty and structural change.

Core Research Themes

Enactive AI

This research develops Enactive AI as an interaction-centered paradigm for artificial intelligence. Rather than treating intelligence as static representation or offline optimization, enactive systems maintain coherence through continuous feedback loops connecting perception, interpretation, regulation, and action.

The work explores how AI systems:


Human–AI Co-Creation

A major focus of the research program investigates how humans and AI systems collaborate in shared creative and cognitive processes. This work helped establish many of the foundational concepts underlying modern co-creative AI, including:

Through systems such as the Drawing Apprentice, Aether, and other co-creative platforms, this research examines how creativity emerges dynamically through interaction rather than isolated generation alone.


Creative Sense-Making

The program introduces Creative Sense-Making as a framework for understanding and quantifying how meaning emerges during human–AI interaction. Rather than evaluating AI solely through outputs, this work studies:

This perspective reframes collaboration as an evolving process of shared sense-making unfolding across time.


Structural Drift and Adaptive Intelligence

Real-world environments rarely remain stable. Financial systems, organizations, social systems, and human–AI collaborations continuously evolve as underlying structures shift over time.

This research investigates how intelligent systems can:

The broader goal is the development of adaptive intelligent systems capable of maintaining coherence within dynamic environments rather than optimizing only under fixed assumptions.

Experimental Systems and Research Platforms

The research program includes the design and development of experimental computational systems that function as research instruments for studying adaptive intelligence and human–AI interaction.

Key platforms include:

These systems operationalize theoretical concepts from enactive cognition and allow them to be empirically studied within real-time interactive environments.

Bridging Cognitive Science and AI

A central goal of the research program is to bridge cognitive theory and computational system design. Rather than treating cognitive science as purely descriptive, this work explores how theories of:

can directly inform the design of next-generation AI architectures.

This approach positions co-creative systems and adaptive AI platforms not merely as applications, but as epistemic testbeds for investigating cognition itself.

Toward Interaction-Centered AI

The broader vision emerging from this research is an interaction-centered paradigm for artificial intelligence.

In this paradigm:

As AI systems become increasingly embedded within dynamic human environments, the future of AI may depend less on autonomous generation alone and more on the ability to sustain meaningful coordination between humans, machines, and evolving environments.

This section documents the theoretical foundations, experimental systems, architectural frameworks, and ongoing research directions contributing to that emerging paradigm.