Concept·Design Exploration·Automation·Developing·CON-022
Design Variants
Value hypothesis
Parallel generation of alternative designs against controlled requirements. Allows multiple directions to be explored at the same time and bootstraps iteration.
Velocity · Innovation
Designer sets parameters - purpose, content, navigation, style - and the tool produces multiple variants simultaneously. Options are reviewed against requirements, and selected for further development or stakeholder review.
Risks in application
Homogenization
Variants tend to aggregate around statistically common design patterns, creating sets that appear diverse on the surface but share underlying compositional logic drawn from training data.
Pseudoproductivity
Reviewing a large variant set creates an appearance of thorough design exploration; if the variants share the same underlying logic, no real exploration of the design space has occurred.
Expertise that differentiates
Creative Direction
Evaluating which have functional merit, aesthetic value, brand coherence, and emotional resonance versus those that are fit the prompt but don't solve the problem.
Design System Logic
Assessing variants against token constraints, component standards, and scalability requirements before advancing.
AI Fluency that assures
Performance Description
Quality depends on the ability to configure the variable dimensions, the constraints to keep fixed, and the requirements to be satisfied.
Transparency Diligence
Disclosing to stakeholders that variant options were AI-generated and may cluster toward statistically common patterns, so they can participate in selection with accurate expectations about what the option space represents.
Related
Possible Indicators
Iteration speed
Number of testable variants in a fixed timeframe relative to a manual exploration baseline.
Sources
Noltenius (2025). Advancing User Experience Design through Generative AI. TU Wien.
Cattapan (n.d.). How my design workflow is changing with AI. Medium.
Wen (2025). Don't Trust the Design Process. Anthropic.
Author unknown (2023). AI as a UX Assistant: Adoption and Attitudes. Nielsen Norman Group.
Fanny (n.d.). Integrating AI Into Real Design Work. UX Planet.