Artificial intelligence is everywhere—from writing emails to generating business ideas in seconds. With that kind of speed, it’s tempting to let AI take over creative work like brainstorming, ideation, and innovation. But research and classroom experiments suggest something surprising: when it comes to true creativity, humans still have the edge .
The Creativity Test That Challenged Assumptions
In a classroom experiment at Kellogg School of Management, students were asked to take a simple creativity test. Their task was to list ten words that were as different from one another as possible. Afterward, they asked an AI chatbot to complete the same task.
Most students expected the AI to outperform them. After all, machines have access to massive vocabularies and process information far faster than humans. But that didn’t happen. On average, human scores matched the AI’s—and in several cases, students produced far more original results than the chatbot.
The reason? While humans draw from personal experience, culture, and perspective, AI tends to recycle ideas from a relatively narrow pool. It often produces “safe” and familiar answers—the greatest hits—rather than unexpected combinations.
The Hidden Cost of Speed
Despite seeing that AI wasn’t more creative, many students still preferred its answers. Why? Speed.
AI delivers responses in seconds, while humans may need minutes of effort. That efficiency feels productive—but it can quietly dull creative thinking. Once people see an AI-generated answer, they often stop exploring their own ideas. Psychologists call this anchoring: we latch onto the first option presented, even if it’s mediocre.
In the experiment, students who edited AI-generated answers spent far less time thinking and made only small changes. As a result, their final ideas were often less original than what they could have produced on their own.
Creativity Is About Process, Not Just Output
True creativity isn’t about finishing first. It’s about differentiation—coming up with ideas others wouldn’t easily copy.
One key insight from the research is that AI works best when it helps shape how we think, not what we think. When students asked the AI for a process instead of direct answers, their creativity improved dramatically.
For example, instead of asking AI to generate ten different words, they asked it how to approach the task. The AI suggested a two-step method: first choose broad categories (such as science, art, nature, or business), then select one word from each category. This approach reduced mental traps and sparked more original thinking.
Automation, Answers, and Alliances
A useful way to think about AI is as a pyramid with three levels:
Automation: Using AI to handle repetitive tasks and scale efficiency.
Answers: Using AI to find information that already exists.
Alliances: Collaborating with AI on complex, creative, and strategic problems.
The highest value lies at the top—alliances. This is where human judgment, imagination, and experience combine with AI’s analytical power. Instead of replacing human creativity, AI becomes a thinking partner.
Why Teams Still Beat Bots
The research also showed that teams outperform individuals—and AI—when it comes to creativity. Groups of people, working together without AI, produced more original ideas than individuals or chatbots. However, when teams relied on AI to tell them what to think, creativity dropped again.
The most successful teams were those that used AI as a guide, not a crutch—asking it for frameworks, questions, and feedback while keeping human insight at the center.
The Bottom Line
Creativity remains a deeply human skill. It thrives on collaboration, curiosity, and lived experience. AI can support that process, but it cannot replace it.
If you want fast answers, ask AI what to think.
If you want a real competitive advantage, ask AI how to think—and do the creative work yourself.
That’s where the magic still happens.
Comments
Post a Comment