100,000 Humans Tested Against ChatGPT—The Results Changed Everything We Know About Creativity
The debate about whether artificial intelligence can truly be "creative" just got blown wide open—and the answer is way more complicated than anyone expected.A massive new study comparing more than 100,000 people with today's most advanced AI systems delivers a surprising result: generative AI can now beat the average human on certain creativity tests. But before you panic about robots taking over art studios and writing rooms, here's the twist: when the AI systems were stacked against the top 10% of creative people, every AI model failed to measure up. ScienceDailyStudyFinds
The internet is having a full-blown existential crisis about it.
What Actually Happened
Researchers from Université de Montréal, including AI legend Yoshua Bengio (one of the godfathers of deep learning that powers ChatGPT), just published the largest human-versus-AI creativity study ever conducted. Published in Scientific Reports in January 2026, the research tested several leading AI models—ChatGPT, Claude, Gemini—against over 100,000 real people on standardized creativity measurements.
Models like GPT-4 showed strong performance on tasks designed to measure original thinking and idea generation, sometimes outperforming typical human responses. ScienceDaily
Let that sink in. GPT-4—the AI you can access for $20/month—beat the median human at generating creative ideas.
The study focused on something called divergent thinking, which psychologists consider a core component of creativity. The test itself was deceptively simple: name 10 words as different from each other as possible. StudyFinds
For example, someone who writes "car, dog, tree" shows less creative range than someone who comes up with "microscope, volcano, whisper." The further apart the words are in meaning, the higher the creativity score.
The Results That Broke the Internet
Here's where it gets wild.
GPT-4 consistently scored higher than the average person. Not close. Higher.
But the story doesn't end there. The average performance of the most creative half of participants exceeds that of all AI models tested, and the top 10% of the most creative individuals open an even wider gap. EurekAlert!
Translation: if you're an average creative worker, you're now competing directly with AI. If you're in the top tier of creative talent? You're more valuable than ever.
The middle just disappeared.
What This Means for Real People
Let's be blunt about what's happening here.
For years, creative professionals have argued that creativity is the one thing AI can never truly replicate—the uniquely human superpower that would always keep us relevant. This study just nuked that comfortable assumption for half the population.
If you're an average creative professional, you're now competing with software that costs $20/month and never sleeps. Ucstrategies News
Marketing copywriters, junior graphic designers, content creators, brainstormers—anyone whose value proposition is "I can come up with ideas" just got put on notice. Not because AI is replacing them tomorrow, but because AI just proved it can match their output today.
But here's the counterbalance that everyone needs to understand: elite creators are fine. Better than fine, actually.
The Top 10% Are Untouchable
When researchers tested more complex creative tasks—writing haikus, movie synopses, short fiction—the pattern became even clearer.
Human writers still produced work with greater variety and originality, especially in poetry and storytelling. But only the best human writers. The researchers wrote in their paper that the persistent gap between the best-performing humans and even the most advanced LLMs indicates that the most demanding creative roles in industry are unlikely to be supplanted by current artificial intelligence systems. StudyFinds
The top 10% of creative professionals operate on a different level entirely. They bring something AI fundamentally can't replicate: lived experience, cultural context, emotional depth, intentionality, and the ability to break rules in ways that create genuine surprise.
How the Test Actually Worked
The Divergent Association Task (DAT) was the primary measurement tool. Created by researcher Jay Olson, it's surprisingly elegant in its simplicity.
The DAT asks participants, whether human or AI, to list ten words that are as unrelated in meaning as possible. An example of a highly creative response includes words like "galaxy, fork, freedom, algae, harmonica, quantum, nostalgia, velvet, hurricane, photosynthesis." ScienceDaily
The test is scored using computational methods that measure semantic distance—how far apart the meanings of words are from each other in conceptual space. The wider the spread, the higher the creativity score.
Here's what makes this test meaningful: performance on the DAT correlates strongly with performance on other established creativity benchmarks. Although the task is language-based, it goes well beyond vocabulary. It engages broader cognitive processes involved in creative thinking across many domains. ScienceDaily
It's not just about knowing weird words. It's about your brain's ability to make unexpected connections.
The AI Models Tested
The research team evaluated several major players:
GPT-4 and GPT-3.5 (OpenAI)
Claude (Anthropic)
Gemini (Google)
Vicuna (open-source model)
GPT-4 came out on top among AI models, consistently beating GPT-3.5 on creative tasks. But there was a fascinating twist: OpenAI released GPT-4-turbo after the original GPT-4, presumably as an improvement. On this creativity test, though, it performed worse. StudyFinds
The researchers suggest newer versions optimized for speed and cost might be trading creativity for efficiency. Bigger and newer doesn't automatically mean more creative.
Even more interesting: Vicuna, a smaller open-source model, beat several larger commercial alternatives. The lesson? Model size isn't everything.
Why This Matters Right Now
This study isn't happening in a vacuum. It's dropping at a moment when creative professionals across industries are already dealing with massive AI disruption.
Graphic designers are watching Midjourney and DALL-E evolve at breakneck speed. Writers are seeing ChatGPT draft blog posts, ad copy, and social media content. Musicians are grappling with AI-generated compositions. Even programmers—often considered safe from automation—are using AI coding assistants that handle increasingly complex tasks.
The 2026 creativity study gives us hard data on something people have been feeling intuitively: AI has crossed a threshold.
It's no longer just a tool. For many tasks, it's a competitor.
The Human Prompt Problem
Here's something crucial that often gets buried in the hype: AI creativity isn't autonomous.
AI creativity is not autonomous. It is shaped by human instruction. A Square Solutions
Temperature settings, prompt engineering, iterative refinement—without that human guidance, AI defaults to generic output. At low temperature, it's cautious and boring. At high temperature, it's varied but often incoherent.
The sweet spot requires human judgment. And that human judgment? That's a skill. One that separates people who get amazing results from AI from people who get mediocre slop.
This is where the practical opportunity lies. The top creators aren't competing against AI—they're amplifying themselves with it. They're using AI to handle the grunt work while they focus on the strategic, emotional, and culturally resonant decisions that machines can't make.
What Makes Top Creators Different
So what exactly are the top 10% doing that AI can't replicate?
Context and intentionality. Elite creators don't just generate ideas—they generate ideas for a specific purpose, audience, and cultural moment. They understand subtext. They know what's been done before and why breaking that pattern will land.
Emotional intelligence. The best creative work makes people feel something. Not just "that's clever"—real emotional resonance. AI can mimic emotional language, but it doesn't understand grief, joy, nostalgia, or rage.
Cultural fluency. Top creators read the room. They know what's happening in the zeitgeist and can riff on it in ways that feel native to the moment. AI trained on historical data is always playing catch-up.
Rule-breaking judgment. Knowing when to break the rules is a form of mastery AI doesn't possess. The best creators know which conventions to smash and which to preserve.
Iteration with taste. Anyone can generate 100 ideas. Elite creators know which three are worth pursuing and how to refine them into something that actually works.
The Uncomfortable Truth About Average Creativity
Here's the part nobody wants to talk about: maybe we've been overvaluing median creative output all along.
Creative professionals join the growing list of high-skill jobs AI is targeting, but the threat isn't replacement—it's commodification of average performance. Ucstrategies News
For decades, companies paid decent salaries for competent creative work. Brainstorming sessions, first drafts, mood boards, concept sketches—these have economic value because they take time and effort.
But what if that value was based on scarcity, not quality? What if average creative work has always been somewhat interchangeable, and we're only realizing it now because AI can do it faster and cheaper?
That's the existential question facing entire industries.
How to Use AI as a Creativity Amplifier
If you're in the creative field, here's the playbook emerging from early adopters who are thriving:
Use AI for the boring stuff. Let it handle first drafts, variations, format conversions, and administrative creative tasks. Save your brain for the decisions that actually matter.
Treat it like a junior collaborator. You wouldn't just accept whatever a junior team member hands you. You'd refine it, redirect it, and apply editorial judgment. Do the same with AI.
Focus on taste, curation, and strategy. AI generates options. You pick the winners and explain why they work. That discernment is increasingly valuable.
Get better at prompting. This is a real skill. Learning how to coax excellent output from AI is becoming as important as traditional creative skills.
Double down on your unique perspective. The more your work reflects specific cultural knowledge, lived experience, or a distinctive point of view, the harder it is for AI to replicate.
The Reaction From Creative Communities
Social media lit up when this study dropped.
On Reddit and Twitter, reactions split predictably along experience lines. Junior creators expressed panic. Seasoned professionals felt validated.
"I've been telling my team for months that AI isn't the threat—it's becoming obvious who actually has talent versus who's been coasting," one creative director posted.
Others pushed back: "This study just proves what I've always suspected—most 'creativity' is pattern matching, which is exactly what AI is good at."
The conversation is getting philosophical fast. What even is creativity if a language model can beat half of humanity at it?
What the Researchers Actually Said
Professor Karim Jerbi, who led the study, was careful not to sensationalize the findings.
Rather than signaling the end of creative careers, the findings suggest a future where AI serves as a creative assistant. By expanding ideas and opening new paths for exploration, AI may help amplify human imagination rather than replace it. ScienceDaily
Yoshua Bengio, one of the study's co-authors and a pioneer in AI development, echoed that balanced view.
The message from the research team is clear: this isn't about humans versus machines. It's about understanding where each excels and building workflows that leverage both.
The Future of Creative Work
Here's what seems increasingly clear: creative work is stratifying.
At the bottom: routine creative tasks that AI handles easily. Commodity work. Generic content. First-pass ideas.
In the middle: competent creative execution. This is where the anxiety lives. This is the tier that's actively being disrupted.
At the top: exceptional creative vision. Strategic thinking. Cultural fluency. Emotional intelligence. Work that reflects genuine lived experience and taste.
The bottom is getting automated. The top is becoming more valuable because it's harder to find. The middle is being squeezed.
If you're in that middle tier, you have two options: level up to the top, or get comfortable using AI to multiply your output and compete on speed and cost.
Neither path is easy. But pretending AI isn't changing the game isn't an option anymore.
Tools You Can Use Right Now
Speaking of practical applications, if you're working with AI for creative projects, you'll want the right supporting tools.
Building presentations? You'll need easy ways to compress images without quality loss—try an Image Compressor to keep file sizes manageable while maintaining visual impact.
Working with lots of PDFs for research or client work? A good PDF Merger can save hours when you're consolidating documents.
And if you're a creator building your own brand, having a professional resume is non-negotiable. An AI Resume Builder can help you craft one that actually stands out to hiring managers.
The point isn't just to use AI for the creative work—it's to build a full toolkit that handles all the unglamorous stuff efficiently, so you can focus on the high-value creative decisions.
What This Means for the Next Generation
If you're a student or early-career creative, this study is both a warning and an opportunity.
The warning: baseline creative competence isn't enough anymore. You can't just be "pretty good at coming up with ideas." AI already does that.
The opportunity: if you develop exceptional taste, deep cultural knowledge, and the ability to guide AI toward genuinely original output, you'll be incredibly valuable.
Learn prompt engineering. Study creative strategy, not just execution. Build a point of view. Develop taste. Those are the skills that age well in an AI-augmented creative economy.
The Environmental Asterisk
One thing that doesn't get talked about enough: there's an environmental cost nobody's pricing in. Ucstrategies News
Every AI creativity test, every ChatGPT brainstorming session, every image generation—it all runs on massive server farms consuming enormous amounts of electricity. As AI creative tools scale, so does their carbon footprint.
It's worth considering whether the convenience is worth the climate cost, especially when human creativity literally runs on lunch.
The Bottom Line
Half of humanity just got outscored by AI on creativity tests. That's the headline that's going to dominate tech news for weeks.
But here's the real story: creativity is fracturing into tiers, and your position in that hierarchy matters more than ever.
The world's largest creativity study just revealed an uncomfortable truth: half of humanity is now less creative than a language model. But the top 10% of human minds still operate in territory AI cannot reach. AI Unfiltered
Where do you fall? And more importantly, what are you going to do about it?
Because one thing's certain: pretending AI isn't disrupting creative work won't make it go away. The only question is whether you're going to use it, compete with it, or get replaced by it.
The choice is still yours. For now.
FAQ SECTION
Q: Can AI really be creative, or is it just mimicking patterns?
A: AI exhibits what researchers call "divergent creativity"—the ability to generate varied and original ideas. However, it lacks lived experience, intentionality, and cultural context that inform the deepest forms of human creativity.
Q: Should creative professionals be worried about AI taking their jobs?
A: It depends on where you rank. If you're in the top tier of creative talent, AI makes you more valuable by handling routine tasks. If you're performing average creative work, AI is now a direct competitor.
Q: What was the Divergent Association Task (DAT) used in the study?
A: The DAT asks participants to generate ten words that are as semantically unrelated as possible. Performance is measured by computational methods assessing the conceptual distance between chosen words.
Q: Which AI models were tested in the study?
A: The study evaluated GPT-4, GPT-3.5, Claude, Gemini, and Vicuna. GPT-4 performed best among AI models but still fell short of top human performers.
Q: How can I improve my creativity compared to AI?
A: Focus on developing taste, cultural fluency, emotional intelligence, and strategic thinking. Learn to use AI as a tool for ideation while you provide the judgment and refinement.
Q: Does prompt engineering really matter?
A: Yes. The study found that AI's creative output is heavily influenced by how you prompt it. Temperature settings and instruction quality dramatically affect results.
