AI is not a replacement for product discipline. It’s a force multiplier for it.

This one’s a little different.

We’ve been sharing our AI POV over the last several weeks, exploring how AI is reshaping product management, why customer proximity matters more than ever, what excellence requires in this new reality, and why ignoring product fundamentals has become expensive. If you’ve been following along, thank you. The conversations that have come from these posts have been incredible.

Last week, we took those conversations live. Vidya and I hosted a webinar called AI as a Force Multiplier for Product Discipline, and we brought a powerhouse along for the ride: Elena Luneva, a product, go-to-market, and AI practitioner with over twenty years of product leadership experience at companies like GoFundMe, Braintrust, Nextdoor, OpenTable, and BlackRock.

Elena currently advises companies and teaches PMs to be more impactful business leaders. She’s brilliant, practical, and not afraid to be direct. In other words, she fits right in with us.
Together with Elena, we dug into the questions product leaders keep asking: How do I use AI without losing rigor? How do I make high-stakes decisions with AI as a partner, not a crutch? And how do I build real customer intelligence using the data that’s already flowing through my organization?

Here’s what came out of that conversation.

When AI can build anything, the only advantage is knowing what’s worth building

This was the core premise of the webinar. AI has collapsed the cost of execution. Teams can prototype faster, generate PRDs faster, and ship faster than ever before. That part is undeniable.

But here’s the problem we keep seeing: speed without discipline doesn’t create value. It creates polished garbage at scale.
We opened the session with a point that I think landed: AI is not making product teams more disciplined right now. It’s exposing those teams who never were disciplined. A little rebellious? Sure. But true.

Here’s what we’re seeing across the teams we work with: AI-powered tools like Rovo (the AI tool inside Jira) are enabling teams to pump out PRDs at a masterful rate. But when you look at what’s inside those PRDs, a majority of them lack a clear understanding of the problem being solved or the customer it’s solving for. They’ve got a good definition of the solution, but no understanding of what success looks like. The result? Those beautiful, polished documents end up in churn once development starts. Decisions get overturned, trade-offs get annihilated, and developers are left asking, “Why are we doing this?”

Without alignment around the foundations—problem, persona, outcomes—you’re just getting to terrible decisions faster.

The shift we want teams to make: use AI to build faster, and use it just as aggressively to challenge what you’re building.

The AI maturity curve: most teams are still at level one

Elena brought data to the conversation. She recently surveyed about eighty product leaders and asked whether they had a process for taking AI-generated prototypes and making decisions about what to actually ship and what will move the business. Eighty-three percent said they didn’t.

Let that sit for a moment. There’s a level of dabbling and trying, but the “so what” and “what do I do with it?” is missing for the vast majority. Elena described an AI maturity curve that she’s seeing across the teams she works with:

1. Prompting. Getting better at individual prompts. This is where most teams still live.
2. Projects. Setting up structured environments (like Claude projects or ChatGPT projects) where context is persistent and shared. Think onboarding docs, customer insights, business strategy—all in one place so your AI is couched in real data.
3. Interconnected workflows. Building agentic systems that pull from multiple sources and automate repeatable tasks. Elena shared an example: product leaders spend about four hours a week on product updates—chasing PMs, pulling from Jira, standardizing the message. That’s a relatively straightforward agentic workflow to automate so you can spend more time on strategy and with customers.
4. The executive stack. Reclaiming the sixty-plus percent of time that product leaders spend in meetings and regurgitating information, and reinvesting it in higher-level decision framing and strategy.

The takeaway here was clear: most teams are stuck at level one or two, and the real unlock isn’t in better prompts. It’s in building the right context and infrastructure so that whatever AI tool you use is grounded in real customer insight and business strategy.

Four ways to use AI for high-stakes product decisions

This was the heart of the webinar. So much of our work over the last ten years has been helping teams move from opinion-based to evidence-based decision-making. AI doesn’t change that mission. It gives us new tools to do it better—if we use them with intention.

We outlined three roles for AI in high-stakes decisions, and Elena added a powerful fourth:

1. AI as a Skeptic

Instead of using AI to generate more ideas and fill your backlog, use it to challenge the ones you already have. Feed it your ideal customer profile, your pain points, your competitive data, and your context. Then ask it: why shouldn’t I build this? We’ve found that this skeptic approach can help teams remove eighty to ninety percent of what they thought was important, because it points out the breaks in your logic.

We recently used this with a team creating a dashboard concept, and the rich, context-heavy prompting revealed that the one or two pain points the team said were important to the customer weren’t actually being addressed by the proposed solution. That’s the kind of insight that saves months of wasted effort.

2. AI as a Truth Seeker

We see so many teams stopping their discovery process too early. At Product Rebels, we’re all about scrappy, continuous learning. But with the shift into building, we’re seeing teams do a little bit of discovery, use AI to synthesize it, and then rush to solutions. The truth seeker mindset is different.

It means constantly pulling in your discovery notes, your customer data, and what you’re learning—and testing those shallow assumptions at every stage. It’s that “five whys” constantly in your development process, pushing the team to go deeper into root causes rather than accepting the first reasonable-sounding answer.

3. AI as a Strategic Thought Partner

Product leaders are often too close to their own problems to see internal contradictions. Strategic failures are rarely caused by bad execution. They’re caused by underlying assumptions that were never tested well. Use AI to pressure test the “why” at every stage. Red team your PR/FAQ. Red team your strategy docs. Red team the PRD. Hunt for those logical leaps that you didn’t realize you were making.

I’ll add a practical layer here: a big part of a product leader’s job is convincing. You have to talk to finance, to the CEO, to your boss—and they all speak differently than a product leader. Use AI extensively in that translation. If you’re talking about engagement and usability, have AI help you translate that into the numbers and framing that a financial stakeholder actually cares about. That’s how you get investment instead of glazed-over eyes.

4. AI as a Mirror

This one came from Elena and it was a standout addition. Companies frequently have a strategy, a roadmap, and then the stuff they’re actually doing—and these three things often don’t align. Ask AI: where are the gaps between your strategy and your roadmap? Where are the gaps between what your CEO cares about and what you’re actually focusing on? This kind of honest reflection can open up opportunities to pivot, change direction, and focus on the functionality that will actually move the business—and probably your career—forward.

But here’s the critical caveat I kept reinforcing throughout the session: all four of these are only as effective as the context you set. Have you actually uploaded or input the business strategy, the outcomes you’re accountable for, the latest customer insights, your current roadmap? The garbage in, garbage out dynamic is stronger with AI than it’s ever been.

There’s a lot more confidence in just throwing something together with AI and calling it done. We call it work slop. And it’s everywhere.

AI-powered customer intelligence: bridging product and go-to-market

Elena brought a perspective here that resonated with everyone on the call. We all agree that spending time with customers is non-negotiable. But there’s a massive source of customer insight that most product teams are underutilizing: the conversations that sales and customer success teams are already having every single day.

With tools like Gong, CS platforms, and call recorders, we now have access to those transcripts. And AI is exceptionally good at synthesizing all of that content. Elena described how teams can use Zapier or API integrations to pull those feeds into their LLM of choice (whether that’s OpenAI, Claude, or whatever your team uses) and start building a systematic, evolving customer understanding.

The power here isn’t just faster synthesis. It’s about closing the gap between product and go-to-market.

Elena called out a pattern most of us have lived: product builds what they think is the right thing, then turns to sales and says, “Why can’t you sell it?” Instead, by building shared customer understanding—using the customer’s own language, pulled from real conversations—product and sales start working off the same foundation. You develop a shared language. You build things that map to how customers actually talk about their problems. And you position them in a way that resonates because the language came from the customer, not from what we think it should be.

The practical advice: start small. Start with one source of information. Know what you want out of it. Then build from there.

From the Q&A: AI in regulated industries and the design question

The live questions were fantastic, and two stood out.

First, on AI in regulated industries: we’ve seen this firsthand in tax, insurance, healthcare, and beyond. AI is incredibly useful for understanding new regulatory changes fast—new tax codes, new state regulations, evolving HIPAA requirements.

I shared how teams we work with use it to establish regulatory context as decision-making criteria. Elena added that in SaaS companies operating across different regulatory environments, AI dramatically speeds up the creation of compliance documentation, release notes, and state-specific positioning. The PM still needs to be the judge. AI isn’t yet an expert in law or healthcare. But as a tool for learning what’s changed and thinking through implications? It’s been a game changer.

Second, on AI prototyping and design: someone asked how to move fast with tools like Figma Make without skipping necessary design expertise. Elena was honest: if you’re starting from scratch, you can vibe code your way into things. But if you’re at an established company with specific design patterns and principles, skipping over strong design leadership leads to more negative than positive results. The tools can extend, but they can’t replace design thinking.

The real answer? Get the foundations right (problem, persona, pain points) and then use AI to accelerate prototyping against those foundations, constantly testing that your solution is actually solving the problem you established in the beginning.

The Takeaway

AI is a force multiplier. That means it multiplies whatever is already there. Strong discipline? AI makes you unstoppable. Weak foundations? AI helps you fail faster and more confidently than ever before.

The shift we’re asking teams to make isn’t dramatic. It’s foundational:

• Use AI to build faster and judge more rigorously
• Set the right context before expecting useful output
• Treat AI as a skeptic, truth seeker, thought partner, and mirror—not an oracle
• Build customer intelligence that bridges product and go-to-market using the data that’s already flowing through your organization

When you can build anything, the only remaining competitive advantage is knowing exactly what is worth building. AI won’t tell you that. Your discipline will.

Want help putting this into practice?

At Product Rebels, we work with product leaders and teams navigating exactly this shift: using AI to accelerate delivery without losing product discipline or customer trust.

Is your team experiencing challenges in implementing AI into your product operating model or struggling in establishing the practices that enable the best outcomes from AI product building?
Schedule 30 minutes with us to learn a little bit about you and explore how we can help.

We’re also bringing back our AI Roundtables—small, confidential group conversations with peer product leaders where we dig into exactly these types of questions. And for those in the Pacific Northwest, keep an eye out for our in-person gatherings. Same conversations, same caliber of leaders, but with some wine. Look out for details in the coming weeks.

About Product Rebels

Product Rebels helps product leaders bring their teams from good to great. We work with established product organizations that already know the basics of product management but want to operate at a higher level. Our focus is not Product Management 101. It’s helping teams build strong customer foundations and outcome-oriented ways of working that consistently translate into better results; for customers and for the business.

We partner with leaders and teams to change how product work actually happens day to day: how customer insight is gathered and shared, how problems are framed, how tradeoffs are made, and how learning compounds over time. AI is infused throughout these practices as an accelerator, helping teams synthesize learning faster, explore more options, and move with greater confidence without sacrificing judgment or customer connection.

The result is product teams that don’t just ship more, they build the right things, make better decisions under pressure, and deliver meaningful, sustained impact.