Design practice
A practical approach to AI in design
Practical examples for reviewing research summaries, prototypes and draft copy.
Article summary
AI is my always-on assistant. I use it to help plan workshop activities and summarise the material from them. The examples below look at reviewing AI output in research, copy and prototypes.
For example, ask for a summary of authorised workshop notes with links to the passages behind each finding. Check the original notes for missing qualifications and disagreements before sharing the summary. Agree how client material may be used before putting it into an AI tool.
A prototype should help answer a design question. For example, build a screen where a reviewer can accept, edit or reject a proposed draft. Test whether someone can complete that task and recover from an error. A working screen does not prove the workflow is ready for production.
Give the model the service scope, audience and destination for each action. Remove benefits that the available evidence does not support. A polished sentence is still a draft: check its facts, read it aloud and see whether the link label predicts what opens.
Before testing an AI workflow, decide what a useful result would look like. Compare quality and effort with the current process, including time spent checking, correcting and handling exceptions. Use those findings to decide whether to continue, revise the experiment or stop.
Related links
- Related service — AI Discovery Workshop
- Get in touch — Talk with Tino about practical AI in product and design work.