How Crayola, Samsung and Generous Brands utilize AI in consumer insights
Editor’s note: Jessica Corbett is the director of product marketing at Suzy, an AI-powered marketing research platform headquartered in New York City. This is an edited version of an article that was originally published under the title “Human edge: What Crayola, Samsung, and Generous Brands actually think about AI.”
Every insights leader right now is navigating the same impossible question: How much do you trust the machine?
At the Quirk's Event – Dallas 2026, Suzy brought together three practitioners who are living that question daily – Matt Frampton, senior director of consumer, shopper and customer insights and analytics at Crayola; Aneesha Nilakantan, director of behavioral science and innovation at Samsung; and Lissa Crisp, director of consumer insights at Generous Brands. What followed was one of the most candid conversations we've had about the real state of AI in consumer insights.
Here's what we learned.
AI is a behind-the-scenes player – not the star of the show.
Across the board, each panelist described the same pattern: AI doing its best work out of the spotlight.
At Crayola, a brand that has been synonymous with imagination for 125 years, Frampton was clear that the brand's core hasn't changed – and won't.
"The hands-on, tactile feel is always going to remain relevant," Frampton said.
Where AI has genuinely moved the needle is in the research and insight work that happens before anything reaches a consumer. More qualitative, conversational research. Faster synthesis. Smarter concept development.
Crisp, a team of one supporting four brands at Generous Brands, uses AI as a brainstorming partner, not a replacement for judgment.
"I use Suzy templates to start, export it, add a few things and I might turn to ChatGPT to say, hey, I want to ask this survey question – give me five ways I could ask this," Crisp explained.
Nilakantan framed this idea as an acceleration of an existing vision.
"Technology will be successful when it's intuitive, embedded in your day-to-day life and acts as your personalized companion. AI just accelerates that," Nilakantan said.
The goal isn't to make the AI visible – it's to make the experience feel seamless.
The risk isn't bad AI. Overconfident humans are
When asked where automation creates risk, none of the panelists pointed at the technology itself. They pointed at us. This was the thread that tied the whole conversation together.
Nilakantan articulated it most sharply.
"The biggest risk is overconfidence leading to quicker decision-making,” Nilakantan said. “It's so easy to be like, 'I have this question, now I have an answer.' And I think that is always going to be a challenge when we're building experiences for people."
Frampton echoed the sentiment from an organizational standpoint.
"Folks can prompt an engine, it spits out insight and now all of a sudden they're a subject expert on what that insight is. It sounds right – and then they run with it," Frampton said.
The answer, he argued, is guardrails – clear internal policies about how AI outputs are presented, with appropriate confidence levels attached.
Crisp went a step further, naming the cultural dynamic that makes this harder to solve.
"I think there are people using AI with good intentions but not the right way – and not saying anything because they're worried about being perceived as taking shortcuts," Crisp said.
The unlock, she argued, isn't more AI policy. It's more transparency.
"I've made sure to tell people: here's when I use it, here's when I don't, here's what data I give it, here's what I don't,” Crisp said.
The competitive advantage is how you use it, not whether you use it
When the panel was asked what happens when AI becomes the baseline – when every competitor has access to the same tools – the answers converged on the same insight.
The edge isn't the technology. It's the discipline around it.
"The competitive advantage is having AI built into your process at very specific points," Crisp said. "So that you're moving fast without taking the risk."
For Samsung, the north star is a single word: companion.
"That helps anchor us in terms of how we think about how we want the user to interact with it, but also how we want ourselves to interact with it," Nilakantan explained.
Terminology shapes culture, and culture shapes decisions.
For Crayola, the edge lies in using AI as an inspiration engine for kids – not a replacement for creativity. Frampton described a phenomenon that will resonate with every parent in the room.
As kids get older, they self-identify as "not creative" and disengage from creative activities. "AI can inspire in ways where that's not truly the case. Joy over judgment." That's not a tech story. That's a brand story powered by technology.
What human-centered AI looks like in practice
Theory is easy. What does this look like when you're in a sprint, under-resourced and staring down a product launch deadline?
Crisp offered the clearest example.
During a recent product innovation project, her team used foundational qualitative research to identify territories and unmet needs – all human-led. Then, in the brainstorming phase that typically happens between qual and concept testing, they brought in an AI tool to generate ideas.
"It did give us some really good ideas that tested well. But would we go to it just because we're in a hurry and on a budget? Absolutely not. That's big stakes for a company," Crisp said.
Nilakantan described how Samsung uses AI not just to model past behavior, but to surface intent, the questions consumers are asking. Basically, using AI as a signal detector and human judgement as the signal interpreter.
"If they're looking for an off-label use, that's actually a product opportunity," Nilakantan said.
Frampton pointed to Crayola's upcoming Story Creator as an example of technology that enhances the human experience rather than replacing it. Story Creator is a creative tablet where kids draw however they want – crayons, markers, paint – and then add their own voice to bring their story to life.
"That blob on the sheet? It's a superhero flying through the sky. Now you understand," Frampton said.
What does a true partnership between AI and humans look like?
"Swim lanes," said Crisp. Know where AI plays, know where humans play and know where they intersect, because the intersection can be the most generative space of all, as long as it's intentional.
"Complementary," said Frampton. AI that gets kids to spend more time being creative, not less.
"Swim lanes plus transparency," said Nilakantan. Define the roles and then be radically honest about how you're using the tools.
AI and brand-side research: The bottom line
The brands doing this well aren't the ones with the most sophisticated AI stack. They're the ones who've answered a harder question, What is AI for, in my specific context, for my specific consumer?
Speed and efficiency are table stakes. The real advantage is knowing when to trust the output and when to trust the human. That judgment can't be automated and is compounded over time.
The strongest business decisions come from truly understanding people. AI can get you there faster, but empathy, context and judgment are still the variables that matter most.