Designing with AI: Knowing When to Go Generative or Predictive

Much like design, AI has its levels. At the feature level, Generative AI can help teams move faster — creating wireframes, writing content, generating code snippets, or suggesting design variations. It’s where creativity meets automation, allowing us to bring ideas to life quickly and test them just as fast.

But when you zoom out to the journey level, Predictive AI becomes far more powerful. When done right, predictive models can help identify patterns, anticipate needs, and shape broader strategies that guide experience design, business priorities, and customer outcomes. It’s the equivalent of designing the entire ecosystem, not just the interface.

Feed AI the Right Data

AI is only as good as what we feed it. Whether you’re prompting a generative model or training a predictive one, make sure it has access to complete, relevant, and contextual data. This includes behavioral insights, business metrics, customer feedback, and operational patterns. The more holistic the data, the more reliable the output.

Take AI Outputs with a Grain of Salt

AI can be remarkably insightful — and equally misleading. Every output should be validated for its accuracy, feasibility, and value. Does the recommendation align with your business goals? Can it be implemented with existing capabilities? And most importantly, does it make sense for your users?

Treat AI as a partner in thinking, not a decision-maker.

People Still Make the Decisions

Even the best AI insights won’t move the organization forward unless people buy in. After reviewing outputs, we still need to engage leaders and stakeholders, aligning them on what the data means and why it matters. AI can surface possibilities — but it takes human judgment and organizational will to make them real.

Efficiency Doesn’t Replace Collaboration

AI brings efficiency, not replacement. We might move faster, but the work doesn’t end there.

At the journey level, influencing change requires alignment across multiple functions — strategy, operations, technology, and design. At the feature level, collaboration continues within squads and stakeholder groups, ensuring the work connects to real outcomes.

The promise of AI isn’t in doing our jobs for us — it’s in giving us more space to do the parts that truly matter: thinking strategically, collaborating effectively, and designing experiences that move people and businesses forward.