Interaction-Required Suggestions for Control, Ownership, and Awareness in Human-AI Co-Writing

2025-04-11

AI Summary

This paper addresses the problem of passive user engagement and lack of process control in generative AI writing. To remedy this, we propose the “Interaction-Required” design paradigm, embedding real-time human-AI collaboration throughout the entire text generation pipeline. Methodologically, we develop a predictive input interface that automatically detects and highlights edit points using semantic and structural awareness, and introduce a human-AI collaborative state synchronization mechanism to enable fine-grained intervention and visualization of the writing possibility space. Empirical evaluation demonstrates significant improvements in users’ sense of control over AI outputs, clarity of revision intent, and trust in collaboration. This work is the first to systematically establish that mandatory, continuous human interaction enhances cognitive engagement, perceived autonomy, and situational awareness. It provides both theoretical grounding and an empirically validated design framework for explainable, intervenable generative AI writing tools.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Natural Language Processing: Generation
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

This paper explores interaction designs for generative AI interfaces that necessitate human involvement throughout the generation process. We argue that such interfaces can promote cognitive engagement, agency, and thoughtful decision-making. Through a case study in text revision, we present and analyze two interaction techniques: (1) using a predictive-text interaction to type the assistant's response to a revision request, and (2) highlighting potential edit opportunities in a document. Our implementations demonstrate how these approaches reveal the landscape of writing possibilities and enable fine-grained control. We discuss implications for human-AI writing partnerships and future interaction design directions.

Problem

• Designing human-involved AI interfaces for co-writing
• Promoting engagement and control in AI-assisted writing
• Exploring interaction techniques for text revision

Innovation

• Predictive-text interaction for response generation
• Highlighting potential edit opportunities in documents
• Promoting cognitive engagement through human involvement