Design AI mediators with negotiable intervention, participant-controlled synthesis, and transparent role transitions that preserve human ownership of team decisions¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article's design implications hold that future AI mediators should treat intervention, representation, and role authority as negotiated rather than fixed. Systems should offer lightweight options the team can "accept, defer, or dismiss", let participants "review, revise, or withhold" how private reflection becomes public representation, and make role transitions transparent so AI contributions "remain contestable and should not carry authority equivalent to that of accountable human members". This preserves human ownership of interdisciplinary decisions while allowing flexible role-switching.

Design Implications¶

Context¶

Requirements¶

• Teams must be able to distinguish when AI is summarizing, translating, advising, or challenging, and the system must disclose the evidence and contextual assumptions underlying its contributions

Constraints¶

• The article states these implications arise from a probe study with a scripted, 15-minute disagreement in one disciplinary configuration (technology, business, design), not from sustained negotiation over a project lifecycle

Target Learners¶

• university students in interdisciplinary project-based learning teams

Target Learning Goals¶

• negotiating disciplinary differences and making collective decisions in text-based team collaboration

Related Strategies¶

• Address equity, transparency, and teacher professional judgment when deploying AI-mediated collaboration
• Team Contracts
• AI-assisted learning systems should scaffold cognitive development rather than replace it, and require active student contribution to preserve ownership

Examples¶

-

Key Sources¶

• Ching-Jung Cheng, Yu-Chan Chung, Bing-Chen Chiu, Yu-Hsuan Lin, and Jia-Wei Liao. 2026. Exploring AI-Supported Disciplinary Mediation in Student Project Teams' Text-Based Communication: A Probe-Grounded Co-Design Study. OzCHI '26. https://arxiv.org/abs/2608.07503