Output isn't design (4 minute read)
Design is fundamentally about understanding the fit between form and context, not just generating outputs, and AI tools risk creating polished but brittle products by skipping the iterative thinking process.
What: An opinion piece by Karri Saarinen arguing that design is the process of deeply understanding a problem and resolving the forces that shape it, rather than simply producing visual outputs or interfaces.
Why it matters: As AI tools become more capable of generating designs quickly, there's a growing risk that teams will mistake output generation for actual design work, leading to products that appear polished but fail when users encounter edge cases and real-world complexity.
Takeaway: Use AI for prototyping and exploration, but preserve the slow, visual, iterative design process that forces you to think through underlying complexities and understand what should exist and how.
Deep dive
- The core misunderstanding in the industry is treating design as the act of producing artifacts rather than understanding problems deeply enough to know what should exist and how
- Christopher Alexander's concept defines design as finding good fit between form and context, where context is the full set of forces including human needs, technical constraints, conflicting requirements, and edge cases
- AI tools generate plausible outputs quickly but don't help you understand underlying problems, and often do the opposite by encouraging you to skip problem-shaping
- Products built this way look impressive initially but unravel during actual use because they're brittle, poorly integrated, and full of unresolved decisions
- Visual design work is valuable because it's slow enough to allow thinking, and the act of moving things around and testing relationships is part of how clarity emerges
- The process parallels writing: asking AI to write produces text but doesn't rearrange your thinking, whereas writing yourself forces you to organize ideas
- The gradual understanding that comes through doing the work is where design value lives, not just in the final output
- AI can still be useful for prototyping, exploration, and generating surprises, but that's different from design itself
- Real design still requires judgment, conversation, tension, and time to work through the complexities
- The risk is mistaking generated form for solved problems when the underlying fit hasn't been achieved
Decoder
- Form and context fit: Christopher Alexander's framework where design is the search for alignment between what you're creating (form) and all the forces that shape the problem (context: needs, constraints, edge cases, relationships)
- Misfits: Points where the form doesn't properly address the contextual forces, resulting in bad design
Original article
Design isn't about producing outputs but about deeply understanding a problem and achieving a good fit between form and context. Tools (including AI) can generate results quickly, but they don't replace the thinking required to resolve underlying complexities. Overreliance on AI risks creating polished yet fragile products, because real design value comes from the slow, iterative process that clarifies understanding—not just the final output.