An Interaction Framework for Studying Co-Creative AI
Project Overview
The document presents a framework for co-creative AI systems designed to enhance collaboration between human users and AI agents in educational settings, particularly in creative tasks like storytelling and game design. It emphasizes the transformative potential of generative AI to enrich learning experiences and foster creativity among students. The framework seeks to improve the interaction quality between humans and AI, making it more intuitive and user-friendly, while recognizing and addressing the limitations of current AI systems. By promoting a deeper understanding of how to effectively integrate AI into creative processes, the document highlights key applications of generative AI in education, showcasing its ability to assist learners in developing innovative ideas and solutions. The findings suggest that when properly implemented, generative AI can significantly enhance creative engagement, leading to improved educational outcomes and a more collaborative learning environment. Overall, the document underscores the importance of designing AI systems that not only support but also inspire human creativity, thereby paving the way for more effective and engaging educational experiences.
Key Applications
Co-creative AI systems for creative tasks such as storytelling, music, and game design.
Context: Creative domains where users collaborate with AI to create artifacts.
Implementation: The framework organizes the interaction in a turn-based manner, allowing both the AI and human to take turns in shaping the creative output.
Outcomes: Enhanced user experience in creating artifacts, increased user engagement and satisfaction, and improved learning for the AI system through user interactions.
Challenges: Designing effective interaction paradigms and ensuring the AI can support creativity without requiring high technical expertise from users.
Implementation Barriers
Technical barrier
High technical knowledge requirement for users to effectively interact with existing creative AI systems.
Proposed Solutions: Develop co-creative systems that require less technical expertise, enabling broader user participation.
Design barrier
Lack of frameworks to adequately represent and compare different interaction paradigms between humans and AI.
Proposed Solutions: Introduce structured frameworks like the proposed turn-based interaction framework to clarify interaction designs.
Project Team
Matthew Guzdial
Researcher
Mark Riedl
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Matthew Guzdial, Mark Riedl
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai