A bicycle-share dock kiosk provides a user story plus a GUI wireframe. Which GenAI analysis ask is most appropriate for that multimodal basis?
Select an answer to reveal the explanation.
Short Explanation
Story text plus a wireframe is a two-lane briefing pack. Steer GenAI with a structured multimodal ask aimed at well-formed acceptance criteria, not a lab grade. Pictures and words together should sharpen what “done” means.
Full Explanation
When a bicycle-share dock kiosk supplies both a user story and a GUI wireframe, the most appropriate GenAI analysis ask is a structured multimodal analysis whose intended output is well-formed acceptance criteria. Civic micromobility kiosks combine textual intent with visual layout, so multimodal prompting should synthesize testable “done” statements rather than treat either modality as decoration. Asking for an executed hands-on lab score with no acceptance criteria fails because grading a lab is not producing AC from the basis. Asking only for pixel-perfect CSS regeneration of the wireframe fails by goal: visual recreation is implementation styling, not testability analysis. Treating the wireframe image as adversarial input to jailbreak the LLM under test fails because the task is application analysis, not attacking the assistant model. Exam caveat: multimodal GenAI asks should declare the intended test artifact—here acceptance criteria—when text and images jointly form the civic basis. Operational check: feed story plus wireframe under a structured prompt, review generated AC for measurability against dock behaviors, correct gaps with product owners, and baseline the criteria before automation or exploratory charters for the public kiosk.