The bottleneck has moved

For years, organizations treated execution as the primary constraint. Ideas were plentiful. Building them was difficult.

Today, AI is steadily removing friction from creation. Software development is becoming more automated. Prototypes can be generated in minutes. Content production scales at unprecedented speed. The barriers between concept and execution are shrinking.

The bottleneck hasn't disappeared—it has simply moved.

The challenge is no longer whether teams can create something. The challenge is determining whether they should.

That's where customer insights become indispensable.

Companies have historically gained advantages through operational efficiency. But when access to powerful AI tools becomes universal, efficiency begins to look less like a competitive moat and more like table stakes. What remains scarce is genuine customer understanding.

Baran Erkel, UserTesting's Chief Strategy Officer, framed the moment as a fork in the road. Organizations can pursue a future driven by automation and synthetic feedback, or they can double down on human-centered design and human understanding.

The choice seems obvious. Yet many organizations remain tempted by the illusion that more data and more automation will naturally produce better decisions.

History suggests otherwise.

The rising value of human judgment

One of the most striking themes from the event was that AI isn't replacing human judgment. It's increasing its value.

Travis Isaacs of Cisco described the transition many professionals are experiencing—from makers to supervisors. As AI handles more execution, human contributions increasingly come from empathy, creativity, context, taste, and storytelling.

Think of AI as a high-powered engine. The engine matters. But direction matters more.

Without judgment, speed simply gets you to the wrong destination faster.

This shift has profound implications for UX research, product strategy, and customer experience leadership. The professionals who thrive won't necessarily be the ones who generate the most outputs. They'll be the ones who ask better questions, identify meaningful customer problems, and connect insights to business decisions.

Former Twitter and Netflix executive Nikkia Reveillac captured this idea succinctly: "The teams with the most influence are never the ones with the most data. They're simply the ones who learn to ask better questions earlier."

That's a powerful distinction.

Data can inform decisions. Questions shape them.

Research becomes a strategic advantage

For years, user research has sometimes been viewed as a validation exercise—a final checkpoint before launch.

That mindset feels increasingly outdated.

When product development accelerates, the cost of building declines. Ironically, that makes research more important, not less. If organizations can launch ten ideas instead of one, understanding which idea deserves attention becomes exponentially more valuable.

Eliel Johnson pointed to a lesson popularized by product pioneer Tony Fadell: build painkillers, not vitamins.

In other words, solve real customer problems.

The metaphor feels particularly relevant today. AI can manufacture an endless supply of vitamins—interesting features, clever experiences, and impressive demos. But discovering genuine pain points still requires talking to people, observing behavior, and understanding motivations.

That's the work of customer insight.

Brad Carrera, who leads design research at McDonald's, argued that researchers play a larger role than many organizations realize. Research isn't merely about producing reports. It's about creating shared understanding.

"Getting a report is not the same as experiencing your customers," Brad explained.

The distinction matters because influence rarely comes from information alone. It comes from helping people see what customers see and feel what customers feel.

The future belongs to human-centered AI

Perhaps the most encouraging takeaway from Crafted Seattle was that the rise of AI does not diminish the importance of humanity. It highlights it.

As routine tasks become automated, uniquely human capabilities move closer to the center of value creation. Empathy. Curiosity. Creativity. Judgment. Connection.

These qualities have always mattered. They're simply becoming easier to see.

Stanford professor Jennifer Aaker expanded the conversation beyond products and technology entirely. She challenged attendees to consider how organizations can use AI while strengthening authenticity, purpose, and human connection.

The companies that succeed with human-centered AI won't be the ones that automate the most. They'll be the ones that understand people the best.

Technology may dramatically change how products are built. It may even redefine the composition of teams and organizations.

But the fundamental challenge remains remarkably consistent: understanding what matters to humans and acting on that understanding.

Or, as Jennifer concluded, "AI can't replace that sense of human connection, but we can take this moment of time and understand what is truly important for us."