Design GenAI-in-the-loop XR tools with editable components, reproducible generation, continuity with existing artifacts, and shared-attention cues¶

Strategy · All strategies
Evidence · no claims cited

Description¶

Based on student interviews, the authors recommend future tool designs that preserve learner agency and collaboration. Students "described workflows in which contextual information could first be established and then smaller components could be generated and edited iteratively," preferring editable components, reproducible generation, and continuity with existing design artifacts over single black-box generation. For collaborative XR, the authors suggest "clearer indications of teammates' viewpoints, pointing or focus cues, and mechanisms that help groups deliberately return to the same design element for discussion," and note architectural students wanted to specify realistic size and scale and build from existing geometry.

Design Implications¶

Context¶

Requirements¶

• Tools must let students understand, modify, and take responsibility for generated representations to preserve authorship

Constraints¶

• Speed of generation alone does not address the need for control, scale fidelity, and continuity with existing design artifacts

Target Learners¶

• design students in collaborative studio or project-based courses

Target Learning Goals¶

• collaborative ideation and design evaluation
• learner agency and authorship in AI-assisted design

Related Strategies¶

• Pursue two future research directions in XR learning: studying learning designers' situated practice and critically investigating power, politics, and ideology in XR platforms

Examples¶

-

Key Sources¶

• Xiao, Y., Chen, M., Li, Y., Powers, N., Wiesenfeld, M., Smith, G., Farzin, S., Liu, S. (2026). Generative AI and Extended Reality in Collaborative Architectural Design Education: An Exploratory Studio Study. arXiv preprint arXiv:2609.13494. https://arxiv.org/abs/2609.13494. doi:10.48550/arxiv.2609.13494