Leading Through the AI Wave—With Clarity, Not Confusion
The leadership challenge behind the AI hype
If you’re a product leader quietly wondering whether you’re already behind on AI, you’re not alone.
Many leaders I coach feel the same: overwhelmed by the expectations, unsure where to start, and guilty for not leading their teams more proactively. They’ve read some articles and books, but don’t feel they have the practical experience yet to confidently set direction. And meanwhile, stakeholders and peers are pushing: Where is our AI strategy? Where is it on the roadmap?
Recently, Teresa Torres reminded us: you’ll learn far more by doing than by scrolling. I couldn’t agree more. But as a product leader, your role isn’t just to encourage exploration—it’s to create a frame for purposeful exploration. And to help your organization avoid patterns that could easily slow you down or add unnecessary complexity.
So where can AI meaningfully help you—today—as a product leader? And how will your products and services change as you learn to apply these new capabilities?
This diagram shows how AI influences both how product teams work and how products evolve—connecting internal practices like discovery and delivery with external shifts in experience, model, and scope.
How AI can change how we work as product people
AI can already support and accelerate key parts of our craft.
In discovery, it can help teams analyze large volumes of qualitative and quantitative data more quickly, summarizing user research and surfacing patterns that might otherwise go unnoticed. It can assist in market and competitive intelligence, scanning competitors’ moves, monitoring industry shifts, and tracking macro trends. These insights help us identify new opportunities and uncover bigger, better user problems to solve—work that too often gets deprioritized when teams are heads down in delivery.
AI can also play a valuable role in solution space exploration. It can help teams brainstorm how to leverage emerging technology to innovate on behalf of the user. This doesn’t replace product thinking, but it can expand the idea space and help us imagine new possibilities. The key is to stay anchored in user needs: using AI to help us innovate for the user, not just for the sake of novelty.
In delivery, AI tools are already creating real impact. Coding assistants like GitHub Copilot are accelerating development and improving code quality. AI-driven test automation is helping teams suggest edge cases, write and maintain test cases, and automate regression testing, which frees up valuable engineering time. Code review assistance is becoming smarter, catching bugs and suggesting improvements. Automated documentation generation is reducing the manual burden on teams while improving documentation coverage. And new AI tools like Lovable and n8n that allow you to build fully functioning apps through a chat interface are disrupting the world of software development as we know it.
Learning is another critical area where AI can help us as product people. We can use AI tools to capture knowledge across retrospectives, analyze learning loops, and surface patterns that inform future work. Encouraging teams to explore AI playfully—to understand both its limitations and possibilities—builds confidence and competence. And it helps teams avoid blindly trusting outputs they don’t fully understand.
How AI can transform the products and services we offer
For those of us working on internal products, this is perhaps one of the biggest areas of immediate opportunity—and one that is often underestimated. AI agents are already radically changing what is possible in business process optimization and internal tooling. They can automate repetitive tasks far beyond what traditional workflow automation offered. They can serve as intelligent assistants (or “interns” as Henrik Kniberg framed it in his Product at Heart talk), helping internal users navigate complex processes, summarize large amounts of information, and even proactively suggest actions.
If you are leading a product team working on products used internally (by colleagues or a few trusted partners), I encourage you to pause and ask: how could agentic AI fundamentally transform what we offer to our colleagues? Not just incremental improvements, but entirely new ways of working. The shift is happening fast. Organizations that move early will not only improve internal efficiency—they will free up human capacity for more strategic work, and fundamentally change the value their internal products deliver.
For external products (by customers and users), AI can open up new possibilities, but we must proceed thoughtfully. Capabilities such as personalization, summarization, intelligent routing, and conversational agents can improve the user experience. More importantly, AI can enable users to accomplish things they previously couldn’t. But we must guard against shipping AI-powered features simply because the market expects it.
How our products evolve when AI is thoughtfully embedded
As AI becomes more deeply integrated into the tools and services we build, it’s also reshaping the very nature of our products. The most forward-thinking teams are not simply adding AI features—they’re reconsidering what their products fundamentally do and enable.
Smarter products: AI is making it possible to embed decision support, predictive assistance, and autonomous workflows. Products are becoming more proactive—able to guide, recommend, or even act on behalf of users.
Hyper-personalization: Interfaces and experiences can now adapt dynamically to user behavior, preferences, and context. This changes not just UX but also how we think about segmentation and customer journeys.
New business models: The value created by AI—whether through insights, automation, or platform capabilities—can unlock new monetization paths, from premium AI add-ons to data-powered APIs.
AI-first design: In some cases, AI becomes the foundation rather than a layer. Products are being re-architected around what AI can uniquely enable, from automating workflows to redefining user roles.
This shift requires strategic thinking, not just technical enablement. It raises questions about differentiation, ethical design, and long-term positioning. And it challenges us to define not only what’s possible—but what’s worth building.
Beware the "AI on the roadmap" trap
This is a trap I see far too often. Many product leaders—especially those running mature, stable products—are being pushed to “get AI on the roadmap” quickly. If no one on your team can clearly articulate the user value being created, you should pause. Adding AI for its own sake risks confusing your product positioning, wasting resources, and eroding trust.
And it puts unnecessary stress on your teams, which many leaders are already feeling.
Be intentional with AI Product Manager roles
Another pattern I’m seeing is the rise of the AI Product Manager role. This happens with every new technology wave. We saw it with APIs, with mobile, and now with AI.
My general rule is simple: avoid specialized roles unless absolutely necessary. We want one product, one product-market fit, one accountable product manager. There is no AI product unless your entire product is AI.
That said, there are cases where a temporary specialized role can make sense. Hiring skilled AI talent is often easier if you offer a focused role. And having a dedicated role can help buffer the organization while the broader team absorbs the complexity of working with AI—from ethical considerations to new usability patterns and technical architectures. I saw this happen during the rise of mobile, when temporary mobile PM roles helped teams manage the learning curve.
But these roles must remain temporary scaffolding. The long-term goal must be to reintegrate AI understanding across the product organization. Specialized roles left in place for too long create more handovers, more tension, and slower progress.
Lead with clarity, not confusion
You do not need to have all the answers today. You do not need to pretend to be an AI expert. But you do need to help your teams explore AI with purpose. Equip them with a simple frame: How can AI help us discover, deliver, and learn—and where might it meaningfully change what we offer to our users?
And, just as importantly, help your organization avoid hype-driven decisions and unhelpful structures.
The opportunity is real. So is the responsibility to lead well. Let’s do it with purpose.
P.S. We had five speakers at Product at Heart—including keynotes by Henrik Kniberg and Elena Verna and themed sessions by Jonathan Evens, Dominik Faber, and Zamina Ahmad—whose talks offered in-depth AI case studies and predictions. We will be publishing the recordings in our video archive and detailed recaps on our blog in the upcoming weeks.