Study
User-Centred DesignNew This WeekStrong effect

User-Centred Design Framework for Explainable AI Systems

A structured user-centered design framework can guide the development of explainable AI (XAI) systems, ensuring transparency and user trust.

Artificial Intelligence Review · 2025

01

Key Findings

  • 01A five-stage framework for user-centered XAI system design (UCXAISD) was identified: contextual inquiry, explanation needs identification, XAI method selection, user interface design, and evaluation and refinement.
  • 02The framework integrates traditional UCD processes with specialized XAI elements like user-centricity, transparency, and actionability.
02

Application

Design takeaway

Integrate a dedicated user-centered design process into the development lifecycle of explainable AI systems, ensuring that user needs for transparency and understanding are met at every stage.

How to apply

Use the five identified stages (contextual inquiry, explanation needs identification, XAI method selection, user interface design, evaluation and refinement) as a checklist and guide for designing any AI system where explainability is a key requirement.

Project actions

  • 01Clearly define the user group for your AI system and their specific needs regarding explanations.
  • 02Consider how the AI's explanations will be presented visually and interactively to the user.
03

Method & Evidence

AimTo develop a comprehensive framework for user-centered design of explainable AI systems.
MethodSystematic Literature Review
ProcedureThe researchers conducted a systematic review of 27 studies published between 2020 and 2024 focused on user-centered design for explainable AI. They analyzed these studies to identify key stages and develop evidence-based guidelines for each stage.
Sample27 studies
ContextArtificial Intelligence System Design

Variables

IVUser-centered design process stages and XAI-specific considerations
DVTransparency, user trust, and usability of XAI systems
CVAI system complexity, application domain, user demographics
04

Strengths & Limitations

Strengths

  • +Systematic approach ensures comprehensive coverage of relevant literature.
  • +Provides a structured, actionable framework for designers.

Limitations

The specific XAI methods and UI elements will vary greatly depending on the AI's complexity and application domain.

Reliability & validity

The reliability of the systematic review depends on the rigor of the search strategy and inclusion/exclusion criteria. Validity is enhanced by synthesizing findings from multiple studies.

Think critically

How might the 'actionability' aspect of XAI explanations differ across various user expertise levels and task complexities?

05

Design Principles

"Prioritize user needs for transparency and understanding throughout the design and development of complex AI systems."

As AI systems become more prevalent, their complexity can lead to a lack of user understanding and trust. Implementing a user-centered approach ensures that the design of XAI prioritizes user needs, making AI more accessible and reliable.

06

What This Means for Your Design

When you design AI systems that need to explain themselves, make sure to involve users from the start to understand what they need to know and how they want to see the explanations.

How to use in your project

  • 1.Reference this framework when justifying your design process for AI-related projects, especially when discussing user research and interface design for explainability.
07

Add to My Project

08

Quick Cite

(2025). Developing user-centered system design guidelines for explainable AI: a systematic literature review. Artificial Intelligence Review. https://doi.org/10.1007/s10462-025-11363-y Retrieved from https://designdex.org/study/13b3d1f9-e476-40ef-9436-893db38d06e1/user-centred-design-framework-for-explainable-ai-systems

Paragraph starter

This research provides a valuable user-centered design framework for developing explainable AI (XAI) systems. The framework, comprising five key stages—contextual inquiry, explanation needs identification, XAI method selection, user interface design, and evaluation and refinement—offers a structured approach to ensure AI transparency and build user trust. By integrating specialized XAI elements with traditional user-centered design principles, this framework serves as a blueprint for creating AI systems that are both technically advanced and highly user-centric.

09

Source

Artificial Intelligence Review

Developing user-centered system design guidelines for explainable AI: a systematic literature review

journal · 2025

View source

Questions about this research

What does the research say about user-centred design framework for explainable ai systems?
Integrate a dedicated user-centered design process into the development lifecycle of explainable AI systems, ensuring that user needs for transparency and understanding are met at every stage. Evidence: Artificial Intelligence Review (2025).
Why does "User-Centred Design Framework for Explainable AI Systems" matter for design?
As AI systems become more prevalent, their complexity can lead to a lack of user understanding and trust. Implementing a user-centered approach ensures that the design of XAI prioritizes user needs, making AI more accessible and reliable.
How can designers apply this research?
Integrate a dedicated user-centered design process into the development lifecycle of explainable AI systems, ensuring that user needs for transparency and understanding are met at every stage.
What were the main findings?
A five-stage framework for user-centered XAI system design (UCXAISD) was identified: contextual inquiry, explanation needs identification, XAI method selection, user interface design, and evaluation and refinement.. The framework integrates traditional UCD processes with specialized XAI elements like user-centricity, transparency, and actionability.
What research method was used?
Systematic Literature Review with 27 studies.
How strong is the evidence?
Evidence strength is rated Strong effect, based on a 2025 journal from Artificial Intelligence Review.
What should I do differently in my next project?
Use the five identified stages (contextual inquiry, explanation needs identification, XAI method selection, user interface design, evaluation and refinement) as a checklist and guide for designing any AI system where explainability is a key requirement.
What are the limitations?
The framework is based on a review of existing literature and may not encompass all emergent XAI design challenges.
Is there evidence that design affects design outcomes?
A five-stage framework has been developed to guide the user-centered design of AI systems that are explainable, focusing on understanding user context, their need for explanations, selecting appropriate AI methods, designing user interfaces, and iterative refinement. As AI systems become more prevalent, their complexit Source: Artificial Intelligence Review (2025).
Where does this user research apply?
Artificial Intelligence System Design It sits within user-centred design research on designdex.org.

Related research topics

design design research · evidence on design · does design improve design outcomes · user studies for designers · design and user findings · user-centred design research evidence