Short answer

Designers and engineers should move beyond purely technical performance metrics and actively involve end-users and stakeholders in the iterative design and validation of AI systems, especially in critical fields like environmental science.

Field
User-Centred Design
Source
Risk Analysis (2023)
Method
Literature review and synthesis, conceptual analysis, and proposal of a research agenda.
Evidence
Strong effect

Developing trustworthy AI in environmental sciences necessitates active engagement with users and stakeholders to address contextual and social dependencies of trust. This user-centred design research insight is drawn from a 2023 study published in Risk Analysis. Using Literature review and synthesis, conceptual analysis, and proposal of a research agenda., researchers explored how this design variable affects real-world outcomes. The key design takeaway: Designers and engineers should move beyond purely technical performance metrics and actively involve end-users and stakeholders in the iterative design and validation of AI systems, especially in critical fields like environmental science.

Study
User-Centred DesignRecentStrong effect

Co-designing AI for Environmental Science: Prioritizing User Trust Through Stakeholder Engagement

Developing trustworthy AI in environmental sciences necessitates active engagement with users and stakeholders to address contextual and social dependencies of trust.

Risk Analysis · 2023

01

Key Findings

  • 01Existing research on AI trust and trustworthiness in environmental sciences has persistent ambiguities and measurement shortcomings.
  • 02Contextual and social dependencies of trust are often underappreciated in AI development.
  • 03Engaging AI users and other stakeholders is crucial for developing trustworthy AI.
  • 04Co-development strategies can help align performance-based standards with dynamic notions of trust.
02

Application

Design takeaway

Designers and engineers should move beyond purely technical performance metrics and actively involve end-users and stakeholders in the iterative design and validation of AI systems, especially in critical fields like environmental science.

How to apply

When designing AI tools for environmental monitoring, prediction, or management, establish a process for regular consultation and co-creation with scientists, policymakers, and affected communities.

Project actions

  • 01When designing an AI system, consider who the end-users are and what their concerns might be regarding trust.
  • 02Think about how the AI will be used in its real-world environment and how that context might affect trust.
03

Method & Evidence

AimHow can engaging users and stakeholders in the co-development of AI systems for environmental sciences enhance trust and trustworthiness?
MethodLiterature review and synthesis, conceptual analysis, and proposal of a research agenda.
ProcedureThe researchers reviewed and evaluated existing research on trust and trustworthiness of AI in environmental sciences, identifying gaps and proposing future research directions focused on stakeholder engagement and contextual factors.
ContextArtificial Intelligence in Environmental Sciences

Variables

IVStakeholder engagement strategies (e.g., co-design, consultation frequency).
DVUser trust in AI systems, perceived trustworthiness of AI.
CVDomain of AI application (environmental sciences), AI system complexity, regulatory environment.
04

Strengths & Limitations

Strengths

  • +Addresses a critical and timely issue in AI development.
  • +Synthesizes existing research to propose a forward-looking agenda.

Limitations

Directly measuring 'trust' can be challenging; focus on observable behaviors and stated preferences related to AI reliability and usability.

Reliability & validity

The reliability of measuring trust can be improved through triangulation of methods (e.g., surveys, interviews, behavioral observation). Validity is enhanced by ensuring the AI system and its context are representative of real-world scenarios.

Think critically

To what extent can AI truly be considered 'trustworthy' if its development is not deeply embedded with the needs and perspectives of its intended users and the communities it impacts?

05

Design Principles

"Trustworthy AI is built through collaborative design and a deep understanding of user context."

Effective AI integration in complex fields like environmental science hinges on user adoption and confidence. By involving end-users and stakeholders throughout the design process, developers can create AI systems that are not only technically sound but also perceived as reliable and appropriate for their intended use, thereby increasing their practical impact.

06

What This Means for Your Design

To make AI that people trust for environmental work, you need to ask the people who will use it and those it affects what they think and involve them in making it.

How to use in your project

  • 1.Reference this research when discussing the importance of user research and stakeholder engagement in your design process, particularly for complex or sensitive applications.
07

Add to My Project

08

Quick Cite

Paragraph starter

This research highlights the critical need for user-centered design principles in the development of AI systems, particularly within specialized domains like environmental sciences. By actively engaging end-users and stakeholders throughout the design process, developers can foster greater trust and ensure the practical applicability and reliability of AI solutions, moving beyond purely technical performance to address the contextual and social dynamics that underpin user confidence.

09

Source

Risk Analysis

Trust and trustworthy artificial intelligence: A research agenda for AI in the environmental sciences

journal · 2023

View source

Questions About This Research

What does the research say about co-designing ai for environmental science: prioritizing user trust through stakeholder engagement?
Designers and engineers should move beyond purely technical performance metrics and actively involve end-users and stakeholders in the iterative design and validation of AI systems, especially in critical fields like environmental science. Evidence: Risk Analysis (2023).
Why does "Co-designing AI for Environmental Science: Prioritizing User Trust Through Stakeholder Engagement" matter for design?
Effective AI integration in complex fields like environmental science hinges on user adoption and confidence. By involving end-users and stakeholders throughout the design process, developers can create AI systems that are not only technically sound but also perceived as reliable and appropriate for their intended use, thereby increasing their practical impact.
How can designers apply this research?
Designers and engineers should move beyond purely technical performance metrics and actively involve end-users and stakeholders in the iterative design and validation of AI systems, especially in critical fields like environmental science.
What were the main findings?
Existing research on AI trust and trustworthiness in environmental sciences has persistent ambiguities and measurement shortcomings.. Contextual and social dependencies of trust are often underappreciated in AI development.. Engaging AI users and other stakeholders is crucial for developing trustworthy AI.. Co-development strategies can help align performance-based standards with dynamic notions of trust.
What research method was used?
Literature review and synthesis, conceptual analysis, and proposal of a research agenda..
How strong is the evidence?
Evidence strength is rated Strong effect, based on a 2023 journal from Risk Analysis.
What should I do differently in my next project?
When designing AI tools for environmental monitoring, prediction, or management, establish a process for regular consultation and co-creation with scientists, policymakers, and affected communities.
What are the limitations?
The paper focuses on a research agenda and does not present empirical data from direct user studies.