Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Design Principles for Generative AI Applications

Justin D. Weisz, Jessica He, Michael Muller, Gabriela Hoefer, Rachel Miles, Werner Geyer

TL;DR

The paper tackles the design challenges of generative AI UX by proposing six design principles, each with concrete strategies, to harmonize user goals of optimization and exploration with the generative nature of artifacts. It employs a rigorous, four-iteration process—literature review, external/internal feedback, a modified heuristic evaluation, and real-world application with two product teams—to validate the principles and ensure practical applicability. Key contributions include reframing traditional AI design concerns (e.g., responsible design, mental models, trustworthy reliance) for generative systems and introducing generative-specific considerations (generative variability, co-creation, and imperfection) with actionable UX strategies. The work demonstrates concrete adoption within industry teams, outlines a robust evaluation framework, and discusses limitations and avenues for integrating design with broader AI development lifecycle decisions, offering practitioners a practical, adaptable toolbox for safer and more effective generative AI experiences.

Abstract

Generative AI applications present unique design challenges. As generative AI technologies are increasingly being incorporated into mainstream applications, there is an urgent need for guidance on how to design user experiences that foster effective and safe use. We present six principles for the design of generative AI applications that address unique characteristics of generative AI UX and offer new interpretations and extensions of known issues in the design of AI applications. Each principle is coupled with a set of design strategies for implementing that principle via UX capabilities or through the design process. The principles and strategies were developed through an iterative process involving literature review, feedback from design practitioners, validation against real-world generative AI applications, and incorporation into the design process of two generative AI applications. We anticipate the principles to usefully inform the design of generative AI applications by driving actionable design recommendations.

Design Principles for Generative AI Applications

TL;DR

The paper tackles the design challenges of generative AI UX by proposing six design principles, each with concrete strategies, to harmonize user goals of optimization and exploration with the generative nature of artifacts. It employs a rigorous, four-iteration process—literature review, external/internal feedback, a modified heuristic evaluation, and real-world application with two product teams—to validate the principles and ensure practical applicability. Key contributions include reframing traditional AI design concerns (e.g., responsible design, mental models, trustworthy reliance) for generative systems and introducing generative-specific considerations (generative variability, co-creation, and imperfection) with actionable UX strategies. The work demonstrates concrete adoption within industry teams, outlines a robust evaluation framework, and discusses limitations and avenues for integrating design with broader AI development lifecycle decisions, offering practitioners a practical, adaptable toolbox for safer and more effective generative AI experiences.

Abstract

Generative AI applications present unique design challenges. As generative AI technologies are increasingly being incorporated into mainstream applications, there is an urgent need for guidance on how to design user experiences that foster effective and safe use. We present six principles for the design of generative AI applications that address unique characteristics of generative AI UX and offer new interpretations and extensions of known issues in the design of AI applications. Each principle is coupled with a set of design strategies for implementing that principle via UX capabilities or through the design process. The principles and strategies were developed through an iterative process involving literature review, feedback from design practitioners, validation against real-world generative AI applications, and incorporation into the design process of two generative AI applications. We anticipate the principles to usefully inform the design of generative AI applications by driving actionable design recommendations.
Paper Structure (58 sections, 3 figures, 4 tables)

This paper contains 58 sections, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Guidelines for human-computer interaction and user interface design
  4. Guidelines for human-AI interaction
  5. Why generative AI needs design principles
  6. Design principles for generative AI applications
  7. Methodology
  8. Iteration 1: Crafting initial design principles
  9. Iteration 2: External and internal feedback
  10. Iteration 3: Modified heuristic evaluation
  11. Method
  12. Results
  13. Relevance
  14. Clarity
  15. Coverage
  16. ...and 43 more sections

Figures (3)

  • Figure 1: Evolution of the design principles across four iterations. During Iteration 3, we recognized that some principles offered new interpretations of existing AI system characteristics whereas others identified new characteristics of generative AI.
  • Figure 2: Portion of an evaluation of AIVA for the principle of Design for Optimization.
  • Figure 3: Evaluators' ratings of the (a) relevance and (b) clarity of each principle and its strategies to their application in the modified heuristic evaluation.