Team Learning as a Lens for Designing Human-AI Co-Creative Systems
Project Overview
The document explores the transformative potential of generative AI in education, highlighting its ability to turn human-computer interactions into collaborative creative processes. It emphasizes the need for learners to develop skills that enable effective collaboration with AI, advocating for team learning strategies that enhance these interactions. The authors introduce a framework aimed at designing generative AI systems that improve the quality of collaboration by promoting shared mental models and facilitating active communication among users. They underscore the significance of these systems in fostering a conducive learning environment where students can creatively engage with AI. Furthermore, the text calls for continued research into optimizing human-AI collaboration in educational and creative contexts, suggesting that such advancements could lead to more effective learning outcomes and innovative educational practices.
Key Applications
Systems that support creative thinking and making, including Image-Sense and SimuLearn.
Context: Creative professionals such as designers using generative AI tools in their workflows.
Implementation: Designing systems to support ideation and production of creative outputs, while also providing learning support for collaboration.
Outcomes: Improved collaboration effectiveness and quality of creative processes, enhanced user awareness of the system's capabilities.
Challenges: Existing systems often support narrow ranges of tasks and lack comprehensive collaboration learning support.
Implementation Barriers
Technical
Most existing generative systems are designed to support a narrow range of creative tasks, limiting their applicability. To address this, it is crucial to develop modular systems that unite various generative capabilities for broader creative processes.
User Experience
Users often struggle with understanding the capabilities of generative AI systems and how to collaborate with them effectively. This can be improved by implementing team learning strategies that help users develop shared mental models and enhance communication with the system.
Project Team
Frederic Gmeiner
Researcher
Kenneth Holstein
Researcher
Nikolas Martelaro
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Frederic Gmeiner, Kenneth Holstein, Nikolas Martelaro
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