Structure GenAI studio sessions as a two-phase create-then-edit design task linking open-ended ideation to context-sensitive refinement¶

Strategy · All strategies
Evidence · no claims cited

Description¶

A session recipe in which students first create imagery from scratch with text prompts (Phase A), then edit provided CAD-based visualisations or photographs using reference-image and hybrid in-painting modes (Phase B), so that later edits must cohere with or deliberately contrast against earlier designs. "This explicit linkage between Phase A (create) and Phase B (edit) is central to examining how the AI supports both open-ended ideation and context-sensitive refinement." The session combines demonstration, individual prompting in a shared space, and a reflective discussion with a usability questionnaire.

Design Implications¶

Context¶

Requirements¶

• Access to the bespoke workflow at computer workstations
• A facilitator demonstration of the three modes with example prompts
• A reflective discussion and usability questionnaire after the hands-on activity

Constraints¶

• Evaluated only as a single 90-minute exposure with a small purposive cohort; the article notes it "may not straightforwardly generalise to other design typologies, levels of complexity or less supported studio environments"

Target Learners¶

• Undergraduate architecture students with varied prior AI and drawing experience

Target Learning Goals¶

• Open-ended design ideation
• Context-sensitive design refinement
• AI-handling skills

Affordances¶

• Constructionist Microworld Genai Studio

Related Strategies¶

Examples¶

-

Key Sources¶

• Kapsalis, T. (2026). Gen-AI-tecture: using generative AI to support architectural students in design tasks. https://orcid.org/0000-0003-1598-6426