Show notes
Hello all.I admit, I didn't expect much. In fact, I was fully prepared to dismiss it as just another AI gimmick—a shiny distraction from real user research. But after experimenting with using AI-generated personas (or "synthetic personas," as they're being called), I found myself in a strange new territory: genuinely impressed.If you're skeptical, trust me—I get it. Personas traditionally demand significant legwork: user interviews, surveys, digging into analytics. Cutting corners here can feel uncomfortable, even reckless. Yet, despite my reservations, the AI method delivered insights deeper and richer than I'd anticipated.But let's slow down for a second. What exactly am I talking about?How AI Generates Synthetic PersonasThe method is straightforward:First, you prompt AI to do deep research into your target audience online. You ask it to dig into questions your users might have, tasks they're trying to accomplish, their emotional state, influences shaping their perception, and so on.Next, you instruct it to distill this mass of data into clear, structured personas. No manual research needed—no lengthy user interviews, no distributing surveys, and definitely no trawling through endless analytics reports.On paper, this is a dream. But it also feels a bit... risky, doesn't it? After all, aren't we skipping the very essence of UX: truly understanding real people?That's exactly why I approached it cautiously.Unexpected Discoveries from AI PersonasHere's what surprised me though—while the AI didn't independently generate personas I hadn't considered, it asked incredibly insightful questions about my target audience that prompted me to explore additional user segments. As we discussed each potential persona, it suggested ways to add more depth and nuance to my initial ideas.During a recent client project, this questioning approach led me to expand beyond my basic buyer personas. The AI's prompts helped me flesh out variations I might have glossed over—like the specific concerns of nervous first-time buyers. Because creating additional personas was so effortless, I ended up developing a much richer, more nuanced set of user profiles than I'd originally planned.In short, I got a more detailed picture of my users—and it took just a fraction of the usual research time.So, Should We Trust Synthetic Personas?Here's the honest answer: not entirely.AI is scouring web data, but it's doing so without human empathy, intuition, or nuance. It's excellent at synthesizing vast amounts of data but struggles with subtleties like hidden motivations or evolving emotional contexts.Still, as a starting point, it's impressive. Think of synthetic personas as a sophisticated first draft, but still in need of human validation to avoid slipping into clever illusions.Synthetic Personas Must Be ValidatedPeople don't always express their true needs and desires online. What appears in digital spaces is just a fraction of human experience. This limitation becomes even more pronounced when dealing with niche products or services - there simply might not be enough relevant online data for AI to generate meaningful personas. Without sufficient training data, these synthetic personas risk becoming oversimplified or misaligned with actual user needs.That's why validation becomes crucial. One of the most effective ways to ensure your synthetic personas reflect reality is to share them with actual end users for feedback. Do these AI-generated profiles genuinely represent their experiences? If not, how do they differ? This real-world validation can highlight gaps or misconceptions in the synthetic personas while providing valuable additional insights.Even if direct access to end users proves challenging, there's still room for validation. Internal stakeholders who regularly interact with your target audience—such as customer service representatives, sales teams, or account managers—can provide valuable feedback on how well these synthetic personas align with their day-to-day observations of real users. Their practical experience can help refine and ground these AI-generated profiles in reality.Don't Lose Touch with Real UsersThere's another critical downside: synthetic personas create a barrier between you and your users. One of the most rewarding aspects of UX is direct interaction with real people. Face-to-face conversations, interviews, and observational research not only give you authentic insights but also help you build genuine empathy.When you rely solely on AI-generated personas, you miss out on that human connection. Your understanding becomes theoretical rather than experiential. This lack of direct interaction can subtly but significantly impact your ability to truly empathize with users' genuine struggles, needs, and aspirations.In short, synthetic personas shouldn't replace direct conversations—they should complement them.My Verdict: Useful, With CautionSo, where do I land on this?I believe synthetic personas have genuine value. They're a practical, insightful shortcut that can enrich your UX toolkit—provided they're handled responsibly. Used thoughtfully, they offer a fast track to initial insights without sacrificing too much depth.But—and this is a big "but"—they're not an end point. Think of AI-generated personas like work from an intern: while it can be valuable and insightful, you still need to review and validate their output before making any significant decisions based on it.So yes, give synthetic personas a try. Embrace them as a tool to kickstart your UX strategy, spark new ideas, and streamline your research workflow.Just don't forget that UX is ultimately about real human experiences, not clever AI illusions.Because in UX, shortcuts can save time—but real connections always win out.