Study finds AI beats average humans on creativity tests but still trails the most creative people

A 2026 study found AI beats the average human on creativity tests - but the top 10% of human creators still outpaced it by a wide margin. AI generates fluently; it has no reason to create.

Categorized in: AI News Creatives
Published on: Jun 08, 2026
Study finds AI beats average humans on creativity tests but still trails the most creative people

What AI Actually Creates (And What It Doesn't)

A major 2026 study found that generative AI now beats the average human on standardized creativity tests. The finding made headlines. It missed the point entirely.

The top 10% of human creators - those producing poetry, storytelling, and meaning-laden work - still left AI well behind. The real question is not whether AI can be creative. It is what kind of creativity we are actually talking about.

Two Different Engines

Human creativity runs on lived experience. Grief, joy, obsession, the slow accumulation of a life actually being lived. Vincent van Gogh painted out of emotional turmoil. Frida Kahlo's self-portraits emerged from surviving a near-fatal bus crash at eighteen.

Generative AI runs on pattern recognition. Feed it enough data and it develops a statistically-grounded sense of what tends to follow what. It becomes fluent in the grammar of creativity without ever having a reason to create.

Fluency Without Judgment

A 2025 study in Frontiers in Psychology tested ChatGPT-4o on the "egg task," a standard measure of creative thinking. The AI generated more ideas than humans. But it exhibited the same fixation bias - clustering around obvious categories - and could not distinguish between its original ideas and conventional ones. Humans made that distinction easily.

Ask an AI to brainstorm unusual uses for a brick and it will produce ten responses confidently and fluently. It cannot reliably tell you which ideas are genuinely surprising versus which ones everyone has already thought of.

Human creativity is inseparable from evaluation. A novelist feels which sentences are alive. A jazz musician hears the note not played and knows it was right. The curator inside the creator - that critical, intuition-driven filter - is largely absent in AI systems.

The Remix Problem

AI-generated output can feel original. An image can surprise you. A paragraph can move you. A melody can send a chill down your spine.

But there is a difference between outputs that feel original and outputs that are original in a deeper sense - generated from a genuinely new vantage point on the world. AI creativity is, at its core, a sophisticated remix of existing human expression. The combinations can be surprising, useful, and beautiful. They emerge from statistical relationships in existing work, not from a new perspective on experience.

When an AI generates a story about heartbreak, it trades in secondhand emotions. It has never felt the weight of a relationship ending, never written something desperate and true because the alternative was silence.

Where AI Actually Works

This does not mean AI creativity is trivial. A PNAS analysis of over 4 million artworks found that text-to-image AI increased creative productivity by 25% and increased the value of work by 50%. The artists who benefited most used AI to explore novel ideas, then applied their own judgment to filter and refine.

AI is a spectacular brainstorming partner. It never gets tired, never gets blocked, never runs dry of suggestions. It can rapidly generate a hundred visual variations, a hundred chapter openings, a hundred chord progressions.

Research shows AI provides the most benefit to less experienced creators, helping them close the gap with skilled practitioners. For highly skilled individuals, the benefit is smaller. Their own originality runs deeper than what AI can add.

The creative partnership model - human ideation plus AI fluency - produces better outcomes than either alone. AI for Creatives works best when the human creator has the taste, vision, and lived experience to know which suggestions are worth pursuing.

The Thing That Cannot Be Automated

Human beings create out of necessity. We create to process what we cannot otherwise understand. We create to communicate with people we will never meet. We create to leave some mark of having been here.

That urgency - that stakes-laden, mortality-aware, meaning-hungry impulse - gives creative work its power to connect. When we encounter art that genuinely moves us, we respond to the presence of another consciousness that looked at the world and felt something worth sharing.

AI has no life experience to draw from. No emotion. No reason to create beyond statistical probability.

The power of human creativity is not in what we make. It is in why we feel compelled to make it at all.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)