Beyond AI adoption: Leveling up to a Design-Enhanced Organization
In my previous post, I explored how AI is changing the way we think about design maturity. AI has lowered the barrier to producing design. More people can create prototypes, explore concepts, generate user journeys, work with research material, and produce polished interfaces with very little effort.
But there is a catch.
More design output does not necessarily mean more design maturity.
In fact, AI can create something that looks like design maturity without actually building the organizational capabilities behind it. So, if AI alone isn't the answer, what does it actually take to become a Design-Enhanced Organization?
The blue box
Let's return to the model from the previous post.
On one axis, we have design maturity: how well an organization uses design knowledge, user-centered methods, and design competence in its decision-making.
On the other, we have AI capability: how effectively the organization uses AI to support design work and decision-making.
This gives us four different situations.
An organization with low design maturity and low AI capability is still largely working in an ad hoc way.
An organization with strong design maturity but low AI capability has good design thinking and practices, but may be slower to explore and produce.
An organization with high AI capability but low design maturity can produce an impressive amount of design — but risks falling into what I called design illusion: lots of output, without a corresponding increase in understanding.
And then there is the blue box: The Design-Enhanced Organization.
This is where strong design capability and strong AI capability reinforce each other.
The interesting question is:
“How do you get there?”
And the answer isn't simply "use more AI."
1. Build design knowledge before scaling AI
The first step is surprisingly unglamorous: build an understanding of design across the organization. That doesn't mean turning everyone into a designer. It means giving people enough design knowledge to understand questions like:
What do we actually know about our users?
What are we assuming?
What problem are we trying to solve?
When should we research rather than start designing?
What can a prototype help us learn?
What does it mean to validate an idea?
What makes evidence useful?
What are the limitations of our research?
This becomes especially important when AI is involved. If someone asks an AI tool to create a persona, a customer journey, or a prototype, they need enough design knowledge to understand what they are looking at. Otherwise, the AI-generated artifact can easily be mistaken for evidence.
AI can generate a plausible answer. Design knowledge helps you decide whether the answer is useful.
That distinction is fundamental.
2. Move design upstream
One of the clearest signs of design maturity is when design enters the process. In a less mature organization, the sequence often looks something like:
Business decision → requirements → development → design
Design is brought in to help shape a solution that has already largely been decided.
A more design-mature organization starts earlier:
Context → user needs → problem framing → exploration → validation → decision
The more mature way includes more steps, and yes, it can take a while before solutions are decided upon. Time is money, so this way is not always easy to sell. AI can help speed things along, but it can also make you go in the wrong direction faster. If a team can generate a prototype in 20 minutes, there is a temptation to skip the “uncomfortable parts”, like researching user needs and problem framing, and start building immediately. But speed isn't always progress. Sometimes the fastest way to build the wrong thing is to start building it very quickly.
A Design-Enhanced Organization therefore uses AI after — and alongside — problem framing, not as a replacement for it.
3. Make assumptions visible
This might be one of the most important design skills in an AI-enabled organization. AI is extremely good at filling in gaps. Ask it to create a user persona and it will happily create one. Ask it to describe a customer journey and it will produce one. Ask it what users probably need and it will give you an answer.
The problem isn't that the answer is necessarily wrong. The problem is that we may forget that it is an assumption.
A mature organization needs to distinguish between different kinds of information:
What we know
Based on research, data, observation or other evidence.
What we interpret
Our understanding of what that evidence might mean.
What we assume
Something we believe to be true but haven't validated.
What AI generates
A plausible possibility that still needs to be assessed.
That distinction becomes increasingly important as AI becomes part of everyday design work.
One simple practice is to explicitly ask:
“What do we know, what do we think, and what are we assuming?”
AI can actually help here — by challenging assumptions, generating alternative interpretations, and identifying questions that need investigation. But the organization has to make that practice part of the way it works.
4. Use AI to expand exploration, not replace judgment
One of the most obvious benefits of AI is that it can dramatically increase the number of alternatives we can explore. Instead of developing one concept, a team can explore ten. Instead of spending an afternoon creating a prototype, they can create several versions in minutes. That's powerful. But more alternatives don't automatically lead to better decisions.
Someone still needs to ask:
Which options are actually relevant?
What assumptions does each option make?
Which user needs does it address?
What are the consequences?
What would make us reject this idea?
What do we need to test?
This is where design judgment becomes more important, not less. AI can widen the scope of exploration. Humans still need to provide the context and judgment. The goal isn't to have AI make all the decisions. It's to make it easier for people to explore the decision space before making them.
5. Change what you measure
Organizations tend to optimize what they measure, and this creates an interesting challenge.
If you measure:
number of prototypes
number of concepts generated
speed of delivery
number of AI tools adopted
amount of content produced
AI will probably make those numbers look impressive. But none of them tell you much about design maturity.
A Design-Enhanced Organization needs to pay attention to a different set of questions:
How much user evidence influenced this decision?
How quickly do we discover that an assumption was wrong?
How often do we test important ideas before committing significant resources?
How well do teams understand the problem before developing the solution?
How often does research actually change a decision?
This is a subtle but important shift. Measure learning, not just output.
AI makes output cheap. That makes learning more valuable.
6. Change the role of designers
If AI can produce design artifacts faster, designers may need to spend less of their time being the people who produce every artifact themselves. That doesn't make design craft irrelevant. It changes where designers can create the most value.
Designers can increasingly become:
problem framers
researchers
facilitators
critical reviewers
systems thinkers
design coaches
stewards of user-centered practice.
Instead of being the people who own design production, they can help the organization develop its ability to think through design. That might mean challenging a product brief before any screens are designed. It might mean helping a team distinguish evidence from assumptions. It might mean facilitating a workshop where different perspectives are brought together. It might mean reviewing AI-generated concepts and asking the questions nobody else has thought to ask.
In other words, the designer's role can move from:
"I will design this for you."
Towards:
"I will help you understand what needs to be designed, why, and how we can learn whether it works."
7. Democratize design — but also democratize design judgment
AI makes it possible to democratize design production. More people can create. More people can prototype. More people can explore. That's a good thing. But if we distribute the tools without distributing the understanding, we risk creating an organization where everyone can generate design, but nobody knows whether it is good design.
So the goal shouldn't be that everyone becomes a designer. Rather, it should be that everyone becomes better equipped to contribute to design decisions. This requires shared language, shared methods, and shared principles. It also means maintaining access to deeper design expertise when the problems become complex.
Democratization shouldn't mean replacing specialists. It should mean making design capability more distributed while keeping design expertise available where it matters most.
8. Create a learning loop
Ultimately, a Design-Enhanced Organization isn't defined by the tools it uses. It's defined by how it learns.
A mature design process creates a loop:
Understand → Frame → Explore → Test → Learn → Decide → Repeat
AI can accelerate almost every part of this loop. It can help synthesize research and identify patterns. It can generate alternatives and can create prototypes. It can suggest test scenarios and help analyze feedback. But the loop only works if the organization is genuinely willing to learn. That means being prepared to discover that you were wrong about something and then change direction.
This is perhaps where organizational culture matters most. No amount of AI capability can compensate for an organization where decisions are already made, and research is only used to justify them.
So, how do you move to the blue box?
The journey will look different depending on where you start.
If you're already strong in design (green box)
Then your challenge isn’t to learn design. Rather, it's to learn how AI can augment and scale the design capabilities you already have.
Look for opportunities to:
reduce repetitive production work
explore more alternatives
accelerate prototyping
make research easier to synthesize
support designers with AI throughout the design process
free up time for deeper research, strategy and facilitation.
The question becomes:
“How can AI make our existing design practice more powerful?”
If you're already using lots of AI (yellow box)
Your challenge may be almost the opposite. Don't start by adding another AI tool.
Start by asking:
Are we actually involving users?
Are we framing problems before generating solutions?
Which of our "insights" are actually assumptions?
Do we test what AI generates?
Do we know why we choose one concept over another?
Does design influence decisions, or are we simply producing more artifacts?
The question becomes:
“How can we turn AI-enabled design output into genuine design capability?”
That is a much harder transformation. But it is also where the real opportunity lies.
The goal isn't AI-powered design. It's a better organization.
It's tempting to think of the future as a simple combination of human creativity and AI resulting in better design. But I think that's too narrow. The more interesting transformation happens when an organization combines strong design capability with AI to make better-informed decisions.
The goal isn't to create an organization where AI generates more designs. It's to create an organization that can:
understand problems better,
explore possibilities faster,
make more informed decisions,
and learn continuously from the people it serves.
That's what makes an organization Design-Enhanced. And perhaps the most important part is this:
“AI doesn’t create design maturity. It amplifies the maturity that is already there.”
If your organization has strong design practices, AI can help you go further.
If your organization has weak design practices, AI can help you move faster — but it can also help you move faster in the wrong direction.
The journey to the blue box therefore isn't primarily about adopting AI. It's about building an organization that knows how to think, learn, and decide through design — and then using AI to make that capability stronger.
“How was AI used in the writing of this article?” you may ask. Well, I wrote a first draft of the article, then used ChatGPT to improve the language. Then I asked it to structure the article so that it was easier to read. I also asked it to come up with some of the example questions.
