Asking Better Strategic Questions About AI in Product Design

People are asking whether AI will replace jobs, reduce headcount, or make their role obsolete. I get it. But those aren't the most useful questions. In most cases, they're the wrong questions entirely.

If the only thing leadership is asking is "How many fewer people do we need?", you're already thinking too small. That's a cost-cutting lens on a capability-expansion moment. It assumes the opportunity is limited to doing the exact same work with fewer resources.

That's where the real value is. When the cost of execution drops, the game changes. Not because you spend less, but because you can suddenly build things, test things, and ship things that used to be off the table. New products. New services. New ways of making better decisions faster. The biggest opportunity isn't lower costs. It's new capability.

The companies that benefit most from AI won't be the ones figuring out how to preserve the status quo a little longer. They'll be the ones asking how to create more value.

What can we build now that we couldn't build before?

In a digital product org, this might mean faster prototyping and validation cycles. It might mean synthesizing user research at a speed and depth that wasn't feasible with a two-person research team. It could mean tighter design-to-code handoff, component generation, or automated accessibility testing baked into the workflow instead of bolted on at the end.

What I've seen firsthand: AI tools can eliminate hours of repetitive DesignOps work, the "boring" parts, while freeing people up to focus on the things that actually require a human brain. Judgment. Taste. Strategy. That's where the value lives.

And when the speed of exploration increases 10x, your team makes better decisions, beyond faster ones, because they can test more ideas before committing.

Where are smart people being blocked?

This is one I think about constantly. For years, great ideas inside companies never got built. Not because the ideas were bad, but because the people who had them were too far away from the people building the tools. There was always a translation layer.

Someone in user research or customer success knew exactly what was broken, but they had to explain it to a PM, write a PRD, wait in line for engineering resources, and hope it got built correctly. That gap killed momentum and killed ideas.

That gap is shrinking. The people closest to the problem can now play a much more direct role in building solutions. A UX researcher who understands user friction. A product manager who knows what data matters most in a feature review. A design system lead who sees exactly where component adoption breaks down.

Those people have real domain knowledge, and AI makes it easier for that knowledge to turn into working prototypes and real solutions, without waiting six months for an engineering allocation that may never come.

How do we empower the people who already understand the problems best?

This isn't a technology question. It's a leadership question.

If your organization punishes experimentation, people won't experiment. If every idea has to move through six layers of approval, nothing meaningful will ship. If AI gets treated like an isolated IT initiative instead of a core product capability, you'll get incremental results at best.

The real question is: Have we created an environment where good people can test useful ideas quickly?

Because speed matters. When iteration cycles get compressed from months to days, the entire operating model shifts. You no longer need to place a handful of big bets each year and hope they work out. You can run small bets, learn faster, and improve continuously. That's a fundamentally different way to build product.

The bottleneck has moved

That compression in speed moves the constraint.

That's a human question. It requires judgment. It requires deep user empathy. It requires clarity about what actually matters. So another question worth asking: Do we have enough judgment inside the organization to know what's worth building?

AI increases the value of good design leadership, it doesn't reduce it. In fact, it raises the bar. Because when the tools become more accessible, the differentiator becomes decision quality. Better thinking. Better priorities. Better understanding of the user. AI generates options. Humans choose.

Quality isn't a premium anymore

Here's another mental model shift worth making: Are we still operating as if quality is expensive?

For a long time, comprehensive documentation, thorough testing, clean design-to-code workflows, and polished execution were treated as luxuries. Things you'd invest in if you had the budget and the time. AI is collapsing the cost of all of those things.

Which means the baseline rises for everyone. Mediocre execution becomes harder to defend. If everyone can move faster, the advantage shifts to the teams that make better choices and create better experiences. The table stakes just went up.

Raise the ambition

This is the one that matters most to me. A lot of product teams are using AI to protect the old model. I think that's too defensive.

The better move is to ask what becomes possible now that the economics of building and operating have fundamentally changed.

The teams that win with AI won't be the ones clinging to the way things were. They'll be the ones creating more value, solving better problems, and operating at a higher level.

That's a conversation worth having.