01 / 09
In a recent conversation, Dylan Field, co-founder and CEO of Figma, said something that captures a shift I think is becoming increasingly difficult to ignore.
As coding models improve, the value moves toward the decisions surrounding the code. Field describes how that value eventually shows up in a company's brand, product design, marketing, point of view, judgment, and strategy. And at the center of much of it sits product design.
It's an interesting way to think about the future of our industry, because it reframes what design actually is—and why it matters.
For years, product design has often been treated as one discipline within a larger process. Business decides what to build. Engineering figures out how to build it. Design makes it usable, useful, and hopefully beautiful.
That model was always incomplete. But as AI changes the economics of building software, its limitations are becoming much more obvious.
The cost of making things is falling
We're entering a world where the distance between an idea and a working product continues to shrink. AI can generate interfaces, write code, assemble functional prototypes, and help people explore ideas that once required entire teams and weeks of work.
More people can build. More ideas can be tested. More software can exist.
That's a remarkable opportunity, but it creates a new problem: when the ability to execute becomes less of a constraint, execution alone becomes less of a differentiator.
Imagine a market filled with products that are technically competent, reasonably usable, and visually polished. Each solves a similar problem. Each can be built quickly. Each can imitate the conventions of the category.
What separates them?
Not necessarily who wrote the best code or who shipped the fastest. Those things still matter, but they become less decisive when everyone has access to increasingly capable tools.
The question shifts from Can we build this? to Should we build this, and what should it be?
That is a fundamentally different problem.
The value rises to the top
Field's observation that "the value rises to the top" is especially compelling because it points to where the scarce resource is moving.
The code is a means to an end. As the cost of producing it falls, more of the value concentrates in the decisions that shape the outcome.
What problem are we choosing to solve? Who are we solving it for? What should the experience feel like? What do we deliberately leave out? Where should we challenge convention, and where should we follow it? What makes this product distinct from everything else competing for someone's attention?
These aren't implementation questions. They're questions of judgment.
They require an understanding of people, context, behavior, culture, business objectives, and the subtle differences between something that works and something that resonates.
And they are deeply connected.
A company's point of view influences its product. The product reinforces—or contradicts—its brand. The brand shapes the expectations created by marketing. Those expectations influence how people experience the product. Every decision either strengthens the whole or introduces another inconsistency.
This is why brand, product design, marketing, and strategy cannot be treated as entirely separate exercises. They are different expressions of the same underlying decisions about what a company believes, what it values, and how it intends to matter in the world.
Product design is where those decisions become real
This is where I think product design becomes particularly important.
Product design isn't simply the process of translating requirements into screens. At its best, it's the discipline that connects business intent, human needs, technology, and the experience a company wants to create.
It makes abstract strategy tangible.
A company's values mean very little if they never influence the choices people encounter. A brand promise means very little if the product experience contradicts it. A compelling vision means very little if the resulting product is confusing, frustrating, or indistinguishable from its competitors.
Design is where those intentions meet reality.
It determines what receives attention, what feels effortless, what earns trust, what creates friction, and what makes an experience memorable. It shapes not only how a product looks, but also how it behaves, what it prioritizes, and what it communicates without saying a word.
These decisions compound. A thoughtful interaction reinforces a feeling of care. A coherent system creates familiarity. A deliberate omission can make a product feel simpler and more focused. Over time, thousands of decisions form something people recognize as a brand—even when they can't articulate exactly why.
That's a much bigger responsibility than making interfaces look good.
And it's why product design belongs closer to the center of strategy, not at the end of the process waiting to receive instructions.
AI lowers the floor. Judgment raises the ceiling.
One of the more useful distinctions in Field's conversation is the idea that AI can lower the floor without necessarily raising the ceiling.
AI makes it easier to produce something acceptable. It can generate a plausible interface, follow familiar patterns, and quickly explore conventional solutions. But producing an acceptable answer isn't the same as finding the right answer, particularly in design, where there may be many possible solutions and no single objective measure of success.
The first output is rarely the finished product. It is a starting point.
The real work is deciding what's missing, recognizing what feels wrong, identifying the possibility that wasn't obvious, and pushing beyond the expected result. It means understanding when a familiar pattern is appropriate and when it is holding the product back. It means knowing when to simplify, when to take a risk, and when a seemingly small detail has a disproportionate effect on the experience.
AI can help explore more possibilities in less time. But exploration only creates value when someone can recognize which possibilities are worth pursuing.
That requires taste, context, experience, and judgment.
Taste, in this sense, isn't simply a preference for a particular visual style. It's the ability to make distinctions: between good and great, generic and distinctive, functional and meaningful. It's the ability to recognize quality and articulate why one direction is more compelling than another.
As the average quality of generated work rises, these distinctions become more important, not less.
When anyone can produce ten options, choosing the right one becomes a more valuable skill than producing the ten options.
The designer's role has to expand
None of this means that designers should own every decision or that other disciplines suddenly become less important. Great products still require excellent engineering, deep customer understanding, strong business judgment, and collaboration across teams.
But it does mean that designers have an opportunity—and perhaps an obligation—to operate differently.
If the work stops at producing mockups, refining components, or handing off specifications, much of the opportunity will be missed. Those activities remain useful, but increasingly capable tools can accelerate large portions of them.
The greater opportunity is to become more involved in defining the problem, shaping the product's direction, establishing its principles, and determining what good actually looks like.
That means getting closer to business strategy and customer research. It means being willing to challenge assumptions rather than simply fulfill requests. It means thinking in systems rather than isolated screens, and considering how individual decisions reinforce the broader experience.
It also means learning to use AI as a way to think and explore more ambitiously—not as a substitute for thinking in the first place.
A designer who uses AI to produce more work is gaining leverage. A designer who uses AI to investigate more possibilities, test assumptions, sharpen a point of view, and push further into the details of a chosen direction is doing something more consequential.
The tools are changing the mechanics of the work. They don't eliminate the need to decide what the work should accomplish.
The companies that stand out will have something to say
As producing software becomes easier, the number of available products will grow. Similar capabilities will become increasingly common. Interfaces will converge around familiar patterns. The distance between a good idea and a working implementation will continue to narrow.
In that environment, differentiation will depend increasingly on the choices companies make—and on whether those choices add up to something coherent.
A recognizable point of view. A clear sense of whom the product serves. A distinctive experience. A brand that reflects what the company actually delivers. A willingness to make decisions that not everyone would make.
These things cannot be bolted on at the end. They have to inform the product from the beginning.
This is ultimately what I find most interesting about Field's observation. AI isn't simply changing how products get made. It's changing which parts of making a product create the most value.
When execution becomes easier, judgment becomes more visible. When output becomes abundant, taste becomes more consequential. And when everyone can build, the question of what deserves to exist becomes harder to avoid.
Product design sits at the intersection of those questions.
Not because designers should own the entire process, but because design is one of the disciplines that brings strategy, technology, brand, and human experience together—and turns them into something people can actually use, understand, and care about.
The opportunity isn't just to design faster.
It's to shape better products, make better decisions, and create things that could not have emerged from execution alone.
The code may be the material. But the judgment behind it is what makes the product matter.