A UI/UX Design Workflow for Using AI While Preserving Creativity
1. Define the problem and broaden exploration with AI
Understanding the user and product context gives you criteria for assessing AI suggestions.
- Clarify user goals, frustrations, business goals, and the next action.
- Use AI to explore potential problems and feature ideas quickly.
- Narrow the direction by selecting, rejecting, and combining ideas instead of accepting the first result.
2. Use AI as a starting point for research and learning
AI explanations and hypotheses become useful when checked against real evidence.
- Treat personas and user journeys as unvalidated hypotheses, and prepare interview questions and issues to investigate.
- Verify actual problems through interviews, surveys, observation, support records, and analytics.
- Ask for examples of unfamiliar concepts and question why a particular design approach is needed.
- Deepen your understanding by comparing answers with real products and your own experience.
3. Shape the structure yourself and use AI for repetitive work
Sketch the flow and reduce the time spent drafting.
- Sketch user flows and information hierarchy yourself, then design layouts, spacing, and styles in Figma.
- Use AI to generate UX copy alternatives, research-note summaries, and presentation outlines.
- Check whether buttons make actions clear, error messages explain how to recover, and wording fits the product’s tone.
4. Seek critique and decide with real feedback
Use AI reviews to surface blind spots and bring product context into the final decision.
- Check for buried primary actions, overloaded screens, and empty states that lack guidance.
- Consider user needs, accessibility, brand, technical constraints, and development cost together.
- Refine the design with feedback from real users. An AI review alone does not complete usability validation.
