Human-Centered AI Communication in Co-Creativity: An Initial Framework and Insights
How AI communicates with humans is crucial for effective human-AI co-creation. However, many existing co-creative AI systems cannot communicate effectively, limiting their collaborative potential. This paper presents the initial design of the Framework for AI Communication (FAICO) for co-creative AI, developed through a systematic review. FAICO highlights key aspects of AI communication and their impact on user experience, offering preliminary guidelines for designing human-centered AI communication. To improve the framework, we conducted a formative study with two focus groups involving skilled individuals in AI, HCI, and design. These sessions sought to understand participants' preferences for AI communication, gather their perceptions of the framework, collect feedback for refinement, and explore its use in co-creative domains like collaborative writing and design. Our findings reveal a preference for a human-AI feedback loop over linear communication and emphasize the importance of context in fostering mutual understanding. Based on these insights, we propose actionable strategies for applying the framework in practice and future directions, marking the first step toward developing comprehensive guidelines for designing effective human-centered AI communication in co-creation.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 83%
Productive vs. Reflective: How Different Ways of Integrating AI into Design Workflows Affect Cognition and Motivation
CHI '25· Human-LLM Collaboration +1
- 83%
Studying Collaborative Interactive Machine Teaching in Image Classification
IUI '24· Human-LLM Collaboration +1
- 71%
CoPrompt: Supporting Prompt Sharing and Referring in Collaborative Natural Language Programming
CHI '24· Immersion & Presence Research +2
- 71%
AI-Augmented Brainwriting: Investigating the use of LLMs in group ideation
CHI '24· Generative AI (Text, Image, Music, Video) +1
- 71%
Collaborative Document Editing with Multiple Users and AI Agents
CHI '26· Human-LLM Collaboration +2
- 71%
Preference-Guided Prompt Optimization for Text-to-Image Generation
CHI '26· Generative AI (Text, Image, Music, Video) +2
- 71%
GraftMind: Facilitating Group Ideation with AI-Mediated Idea Sharing
CHI '26· Human-LLM Collaboration +2
- 71%
Towards AI as Colleagues: Multi-Agent System Improves Structured Ideation Processes
CHI '26· Human-LLM Collaboration +2
- 71%
Partnering with Generative AI: Experimental Evaluation of Model-Led and Human-Led Interaction in Human-AI Co-Creation
CHI '26· Generative AI (Text, Image, Music, Video) +2
- 71%
ClearFairy: Capturing Creative Workflows through Decision Structuring, In-Situ Questioning, and Rationale Inference
CHI '26· Human-LLM Collaboration +2
Based on Jaccard similarity of research subtopics & professions (≥60%)