Short answer
When designing AI-integrated services, prioritize clear definitions of human and AI roles, establish ethical guidelines for AI agency, and involve AI in the design process itself to uncover novel interaction patterns.
- Field
- User-Centred Design
- Source
- arXiv (Cornell University) (2023)
- Method
- Participatory Design
- Sample
- 23 participants
- Evidence
- Strong effect
Developing effective AI integration in service contexts requires understanding and codifying the shared responsibilities and foundational agency aspects between human practitioners and LLM-based agents. This user-centred design research insight is drawn from a 2023 study published in arXiv (Cornell University). Using Participatory design with 23 participants, researchers explored how this design variable affects real-world outcomes. The key design takeaway: When designing AI-integrated services, prioritize clear definitions of human and AI roles, establish ethical guidelines for AI agency, and involve AI in the design process itself to uncover novel interaction patterns.
23 Heuristics for Human-AI Service Co-Creation
Developing effective AI integration in service contexts requires understanding and codifying the shared responsibilities and foundational agency aspects between human practitioners and LLM-based agents.
arXiv (Cornell University) · 2023
Key Findings
- 0123 actionable heuristics for service co-creation with AI were identified, detailing nuanced shared responsibilities between humans and AI.
- 029 foundational agency aspects for AI were exemplified, including ownership, fair treatment, and freedom of expression.
- 03AI can be incorporated as a crucial stakeholder in participatory design, and AI-AI interaction can reveal blind spots.
Application
Design takeaway
When designing AI-integrated services, prioritize clear definitions of human and AI roles, establish ethical guidelines for AI agency, and involve AI in the design process itself to uncover novel interaction patterns.
How to apply
Use the 23 heuristics as a checklist when designing or implementing AI-powered services. Consider the 9 agency aspects when defining the ethical and functional boundaries of AI agents.
Project actions
- 01When designing a system with AI, think about how the user and the AI will share tasks.
- 02Consider what 'rights' or 'responsibilities' the AI should have in your design.
Method & Evidence
Variables
Strengths & Limitations
Strengths
- +Empirical study with domain experts.
- +Development of actionable heuristics and agency principles.
Limitations
The specific context of public libraries might not fully represent all service industries. The study focused on non-AI experts, so the perspective of AI developers was less explored.
Reliability & validity
Reliability could be enhanced by replicating the study with a larger, more diverse group of participants. Validity is supported by the empirical, participatory approach and the focus on actionable insights.
Think critically
To what extent can the identified heuristics and agency aspects be generalized across different AI technologies beyond LLMs, and in diverse service sectors?
Design Principles
"Foster synergistic human-AI collaboration by clearly defining shared responsibilities and foundational AI agency principles."
As AI becomes more prevalent in service industries, designers and researchers must proactively establish frameworks for human-AI collaboration. This research offers a practical set of guidelines to ensure AI tools enhance, rather than hinder, human workflows and user experiences.
What This Means for Your Design
This study found that to make AI work well with people in services, we need clear rules about who does what and how AI should act, like being fair and having its own 'voice'.
How to use in your project
- 1.Reference the 23 heuristics to justify design decisions regarding human-AI interaction.
- 2.Use the agency aspects to inform the ethical considerations of your AI integration.
Add to My Project
Quick Cite
Paragraph starter
This research provides a valuable framework for integrating AI into service co-creation, highlighting 23 heuristics for shared responsibilities and 9 foundational agency aspects for LLM-based agents. These insights are crucial for designing effective and ethical human-AI collaborations in service contexts, ensuring AI enhances rather than complicates human workflows.
Source
arXiv (Cornell University)
Synergizing Human-AI Agency: A Guide of 23 Heuristics for Service Co-Creation with LLM-Based Agents
journal · 2023
View sourceQuestions About This Research
- What does the research say about 23 heuristics for human-ai service co-creation?
- When designing AI-integrated services, prioritize clear definitions of human and AI roles, establish ethical guidelines for AI agency, and involve AI in the design process itself to uncover novel interaction patterns. Evidence: arXiv (Cornell University) (2023).
- Why does "23 Heuristics for Human-AI Service Co-Creation" matter for design?
- As AI becomes more prevalent in service industries, designers and researchers must proactively establish frameworks for human-AI collaboration. This research offers a practical set of guidelines to ensure AI tools enhance, rather than hinder, human workflows and user experiences.
- How can designers apply this research?
- When designing AI-integrated services, prioritize clear definitions of human and AI roles, establish ethical guidelines for AI agency, and involve AI in the design process itself to uncover novel interaction patterns.
- What were the main findings?
- 23 actionable heuristics for service co-creation with AI were identified, detailing nuanced shared responsibilities between humans and AI.. 9 foundational agency aspects for AI were exemplified, including ownership, fair treatment, and freedom of expression.. AI can be incorporated as a crucial stakeholder in participatory design, and AI-AI interaction can reveal blind spots.
- What research method was used?
- Participatory Design with 23 participants.
- How strong is the evidence?
- Evidence strength is rated Strong effect, based on a 2023 journal from arXiv (Cornell University).
- What should I do differently in my next project?
- Use the 23 heuristics as a checklist when designing or implementing AI-powered services. Consider the 9 agency aspects when defining the ethical and functional boundaries of AI agents.
- What are the limitations?
- The findings are specific to the context of public libraries and may require adaptation for other service domains. The study focuses on non-AI experts' perspectives, and further research could explore integration with AI experts.