Co-Creative AI

Toward an Interaction-Centered Paradigm of Human-AI Co-Creation

Co-Creative AI is an interaction-centered paradigm of artificial intelligence focused on how humans and AI systems collaborate in shared processes of creativity, meaning construction, adaptive interaction, and participatory sense-making. This site documents more than a decade of interdisciplinary research by Nicholas Davis, PhD and collaborators spanning: co-creative AI, Creative Sense-Making, enactive AI, participatory sense-making, quantified co-creation, hybrid intelligence, and human-AI interaction.

Beginning with early work in computational creativity and human-computer co-creativity, this research evolved into one of the earliest sustained research programs investigating AI systems as collaborative participants rather than isolated generators or passive tools. Early publications such as:

helped establish foundational concepts for modern co-creative AI systems. This work introduced frameworks and concepts including: artistic computer colleagues, participatory interaction, creative trajectories, quantified co-creation, sense-making curves, conceptual shifts, and interaction-centered models of intelligence.

The research collected here helped establish many of the foundational concepts underlying modern co-creative AI systems, including: artistic computer colleagues, creative trajectories, sense-making curves, participatory interaction, and interaction-centered models of intelligence.

Rather than treating AI as merely a tool or autonomous generator, this research explores AI systems as dynamically coupled participants engaged in shared creative and cognitive processes with humans. Within this perspective, intelligence is not viewed as isolated computation occurring solely within humans or machines. Instead, intelligence emerges dynamically through interaction between coupled participants engaged in shared processes of sense-making.

“A vertically oriented infographic titled ‘The Co-Creative AI Research Ecosystem’ visualizing the evolution of interaction-centered AI research from 2013 to 2025+. The design uses clean academic styling with soft blues, greens, oranges, and purples. At the top is the subtitle: ‘Interaction-Centered AI for Collaborative Sense-Making, Quantified Co-Creation, and Enactive Intelligence.’ The infographic begins with a horizontal timeline showing six major phases of the research program: 2013 Co-Creativity Foundations, 2014–2016 Drawing Apprentice, 2017 Creative Sense-Making, 2018–2021 Quantified Co-Creation, 2022–2024 Explainable Co-Creative AI, and 2025+ Enactive AI. Each stage contains a brief description and icon representing the conceptual shift. Below the timeline is a large central framework diagram organized around ‘Interaction-Centered Co-Creative AI’ at the center. Surrounding nodes include Creative Sense-Making (CCSM), Quantified Co-Creation, Explainable Co-Creative AI, Enactive Regulation Theory (ERT), Participatory Sense-Making, and Enaction & Embodied Cognition. Arrows connect the nodes to illustrate conceptual relationships and mutual influence. On the left side is a ‘Systems & Implementations’ panel listing projects including Drawing Apprentice, AI Drawing Partner, Interaction Visualization Tools, Co-Creative Analytics, Provenance & Trace Systems, and Emergence Machine. On the right side is a ‘Key Contributions’ panel highlighting interaction-centered AI, quantified co-creation, explainable co-creative AI, participatory frameworks, enactive adaptive systems, and bridging theory with practice. Near the bottom are sections for Foundational Essays, Davis et al. Publications, and the broader Research Program. A final footer row labeled ‘Impact & Application Areas’ includes creative arts and design, human-AI collaboration, education and learning, wellbeing and health, scientific discovery and sense-making, and social good and public interest AI. The infographic presents the research ecosystem as an interconnected interdisciplinary framework combining human-AI interaction, enactive cognition, computational creativity, explainable AI, and adaptive co-creative systems.”

EnactiveAI.pdf

Toward Interaction-Centered Enactive AI

Traditional artificial intelligence has largely focused on: prediction, optimization, classification, automation, and autonomous generation. Co-Creative AI proposes a fundamentally different paradigm. Rather than treating cognition as isolated information processing, co-creative systems investigate how creativity, meaning, and adaptive intelligence emerge through interaction itself. This interaction-centered perspective draws from: enactive cognition, participatory sense-making, human-computer interaction, computational creativity, ecological psychology, adaptive systems research, and hybrid intelligence frameworks.

Over time, these ideas evolved into broader theoretical frameworks including: Creative Sense-Making, Enactive AI, quantified co-creation, and human-AI co-creation as a new interaction paradigm.

This research includes the development of some of the earliest co-creative AI systems designed to collaborate with humans in real time, including: The Drawing Apprentice and The AI Drawing Partner. These systems were not designed simply to generate outputs autonomously. Instead, they were developed as: interactive collaborators, improvisational partners, and quantified co-creative systems capable of: modeling interaction dynamics, logging activity traces, visualizing collaboration patterns, generating conceptual shifts, and quantifying the co-creative process itself.

A major contribution of this research has been the development of frameworks for quantifying co-creation in situ through: interaction dynamics, creative trajectories, activity traces, and sense-making curves.

These approaches helped establish a new subfield of co-creative AI research focused on understanding how meaning and creativity emerge dynamically during interaction between humans and AI systems. More recent work extends these frameworks into the era of generative AI and hybrid intelligence through: explainable co-creative AI systems, quantified artificial media, interaction-centered design frameworks, and human-AI co-creation models for contemporary generative systems.

Together, the theories, systems, and publications collected here document the emergence of an interaction-centered paradigm of artificial intelligence — one in which creativity and intelligence arise not from humans or machines alone, but through dynamic participation between them.

Why This Matters Today

The rise of generative AI systems — including ChatGPT, copilots, diffusion systems, and multimodal AI agents — has transformed public interest in human-AI collaboration. Yet most modern AI systems still primarily focus on autonomous generation, prompt-response interaction, and output optimization.

Co-Creative AI explores a fundamentally different paradigm.

Rather than treating intelligence as isolated generation, this research investigates how creativity, meaning, and intelligence emerge dynamically through sustained interaction between humans and AI systems. The focus shifts from outputs alone toward: interaction dynamics, participatory sense-making, adaptive collaboration, and quantified co-creation.

Long before the current generative AI wave, projects such as the Drawing Apprentice and AI Drawing Partner explored: real-time human-AI improvisation, shared creative workspaces, interaction-centered intelligence, explainable co-creative systems, and computational modeling of collaborative dynamics.

As AI becomes increasingly integrated into creative, scientific, educational, and professional environments, interaction-centered frameworks such as Creative Sense-Making, participatory sense-making, and Enactive AI may become increasingly important for understanding how humans and AI systems collaborate meaningfully together---understanding how humans and AI systems collaborate may become just as important as improving generation quality itself.

Page updated

Google Sites

Report abuse