Technology Category
Humans and AI: User Experience and UsabilityCognitive Modeling & Cognitive Systems: Computational CreativityPhilosophy and Ethics of AI: Safety, Robustness & Trustworthiness
On the usability of generative AI: Human generative AI
🤖 AI Summary
Generative AI systems suffer from critical usability bottlenecks—including unpredictable outputs, challenging fine-tuning, low transparency, weak user control, and high cognitive load. To address these, this study introduces the first cross-domain usability evaluation framework for generative AI, systematically diagnosing issues along four dimensions: user experience, transparency, user control, and cognitive load. We propose a novel usability enhancement paradigm centered on explainability augmentation, intuitive interaction design, and closed-loop user feedback. Empirical validation integrates human-computer interaction evaluation, cognitive ergonomics analysis, and multidimensional metrics (efficiency, learnability, satisfaction). Results yield domain-specific best practices for content creation, education, and programming—demonstrating significant improvements in task completion rates and user satisfaction. This work delivers the first evidence-based usability guideline for designing trustworthy generative AI systems.
Technology Category
Application Category
📝 Abstract
Generative AI systems are transforming content creation, but their usability remains a key challenge. This paper examines usability factors such as user experience, transparency, control, and cognitive load. Common challenges include unpredictability and difficulties in fine-tuning outputs. We review evaluation metrics like efficiency, learnability, and satisfaction, highlighting best practices from various domains. Improving interpretability, intuitive interfaces, and user feedback can enhance usability, making generative AI more accessible and effective.
Problem
Research questions and friction points this paper is trying to address.
Enhance usability of generative AI
Address unpredictability in AI outputs
Improve user experience and control
Innovation
Methods, ideas, or system contributions that make the work stand out.
Enhancing interpretability of generative AI
Developing intuitive user interfaces
Incorporating user feedback mechanisms
🔎 Similar Papers
No similar papers found.
