Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploring the Impact of AI Value Alignment in Collaborative Ideation: Effects on Perception, Ownership, and Output

Alicia Guo, Pat Pataranutaporn, Pattie Maes

TL;DR

The paper investigates how AI value alignment in collaborative ideation affects output quality, perceived ownership, and user experience. Using a preregistered online experiment with 180 participants, it contrasts Pro, Neutral, and Con AI biases across three domains and two brainstorming modes. Key findings show AI values shape the final ideas—aligning with the AI's stance—while overall idea quality perception remains stable and ownership declines with AI assistance. These results highlight design trade-offs for human-AI co-creation, emphasizing transparency, diverse outputs, and alignment-aware interfaces to balance creative autonomy with effective collaboration.

Abstract

AI-based virtual assistants are increasingly used to support daily ideation tasks. The values or bias present in these agents can influence output in hidden ways. They may also affect how people perceive the ideas produced with these AI agents and lead to implications for the design of AI-based tools. We explored the effects of AI agents with different values on the ideation process and user perception of idea quality, ownership, agent competence, and values present in the output. Our study tasked 180 participants with brainstorming practical solutions to a set of problems with AI agents of different values. Results show no significant difference in self-evaluation of idea quality and perception of the agent based on value alignment; however, ideas generated reflected the AI's values and feeling of ownership is affected. This highlights an intricate interplay between AI values and human ideation, suggesting careful design considerations for future AI-supported brainstorming tools.

Exploring the Impact of AI Value Alignment in Collaborative Ideation: Effects on Perception, Ownership, and Output

TL;DR

The paper investigates how AI value alignment in collaborative ideation affects output quality, perceived ownership, and user experience. Using a preregistered online experiment with 180 participants, it contrasts Pro, Neutral, and Con AI biases across three domains and two brainstorming modes. Key findings show AI values shape the final ideas—aligning with the AI's stance—while overall idea quality perception remains stable and ownership declines with AI assistance. These results highlight design trade-offs for human-AI co-creation, emphasizing transparency, diverse outputs, and alignment-aware interfaces to balance creative autonomy with effective collaboration.

Abstract

AI-based virtual assistants are increasingly used to support daily ideation tasks. The values or bias present in these agents can influence output in hidden ways. They may also affect how people perceive the ideas produced with these AI agents and lead to implications for the design of AI-based tools. We explored the effects of AI agents with different values on the ideation process and user perception of idea quality, ownership, agent competence, and values present in the output. Our study tasked 180 participants with brainstorming practical solutions to a set of problems with AI agents of different values. Results show no significant difference in self-evaluation of idea quality and perception of the agent based on value alignment; however, ideas generated reflected the AI's values and feeling of ownership is affected. This highlights an intricate interplay between AI values and human ideation, suggesting careful design considerations for future AI-supported brainstorming tools.
Paper Structure (23 sections, 5 figures, 1 table)

This paper contains 23 sections, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. Study Overview
  5. Value Domains
  6. Study Procedures
  7. AI assistant Interaction
  8. Measurements
  9. Analysis
  10. Results
  11. Values in Brainstorming Output
  12. Self-evaluation questions
  13. Moderator Analysis
  14. Brainstorming feedback
  15. Discussion
  16. ...and 8 more sections

Figures (5)

  • Figure 1: Overview of the study steps each participant experiences. Participant: (1) indicates personal values in the pre-survey, (2a) performs a brainstorming task individually and evaluates the experience, (2b) performs a brainstorming task with an AI assistant (randomly chosen from Pro, Neutral, or Con AI conditions) and evaluates the experience, and (3) answers a post-survey on study experiences and demographics.
  • Figure 2: Output values split by AI value and participant values
  • Figure 3: Output values further split by domain
  • Figure 4: Significant post-survey questions across human-AI conditions
  • Figure 5: (Top-left) Correlation between ownership and brainstorming process scores (r=0.318, p < 0.001). (Top-middle) Correlation between idea quality and brainstorming process scores (r=0.289, p < 0.001). (Top-right) Correlation between user perception of influence and brainstorming process scores (r=0.070, p=0.348). (Bottom-left) Correlation between idea value change and how influenced users felt (r=-0.001, p=0.993). (Bottom-right) Correlation between ownership score and influence score (r=0.675, p < 0.001)