Touchstone Research finds Gen Z judges AI products on trust and transparency first, novelty second, and quietly abandons tools that overpromise.
Building an AI product that wows Gen Z isn’t the same as building one they’ll actually keep using. According to research from Touchstone Research, a market research firm focused on youth and consumer insights, young users evaluate AI products on trust and transparency first, and novelty a distant second.
The core finding is straightforward but easy for product teams to miss: Gen Z doesn’t want AI to feel magical. They want it to feel honest.
Transparency Beats Impressiveness
According to the research, Gen Z responds better to AI tools that are upfront about their limitations than to tools that project confidence on everything. A system that flags uncertainty or lower-confidence answers builds more trust with young users than one that states every answer with the same flat, unwavering confidence. Overconfidence, in other words, reads as untrustworthy rather than impressive to this audience.
That preference extends beyond raw accuracy. What earns Gen Z’s trust, according to the research, is a product that’s upfront about what it can and cannot do, responds in a way that matches how they actually communicate, and gives them enough visible control to feel like they’re steering the interaction rather than guessing at it.
Young Users Test Products Before They Trust Them
The research also found that younger users engage in a distinct form of trust calibration early in their use of a new AI tool. Rather than easing into a product gradually, Gen Z users are more likely to deliberately probe a system: asking odd or unexpected questions, or trying to break it, before deciding whether to rely on it for real tasks.
That testing phase matters more than it might seem. Products that perform inconsistently while a young user is still forming their first impression tend to lose credibility quickly, and that early skepticism can be difficult to undo later.
They Don’t Complain. They Just Leave
Perhaps the most important finding for anyone building AI products is what Gen Z does when an experience disappoints them. According to the research, Gen Z users rarely file a complaint or give explicit negative feedback when an AI feature misses the mark. Instead, they quietly adjust their own behavior: they stop using the feature, work around it, or simply go back to doing the task manually.
That silent abandonment means problems with an AI product can remain invisible to a team relying only on typical feedback channels or complaint volume. Structured, dedicated research aimed specifically at young users is often needed to catch issues that broader studies, or a quiet drop in usage, would otherwise miss.
Asking the Wrong Questions Misses the Real Signal
The research argues that many companies evaluate their AI features by asking whether users find them impressive or engaging, and that this framing misses what actually predicts whether Gen Z will keep using a product. More useful questions, according to the research, focus on trust, control, and whether the tool behaved the way a young user expected it to.
For product and research teams building AI features aimed at younger audiences, the implication is that measuring delight or novelty in isolation isn’t enough. Tools that feel impressive in a first demo can still fail to earn lasting use if they don’t also feel honest, predictable, and responsive to the user’s own sense of control.