Why Teams That Use AI Are Getting Less Creative — And How to Fix It

Generative AI has become the go-to brainstorming partner for teams across every industry. From marketing campaigns to product design, workers are leaning on tools like ChatGPT, Claude, and Gemini to generate ideas faster than ever. But a growing body of research reveals a troubling paradox: while AI makes individual workers more creative, it is quietly making teams less innovative overall. At Metaintro, we track how emerging technology reshapes the labor market — and this creativity gap is one of the most important workforce trends of 2026. Here is what the science says and what your team can do about it.

What Does the Research Actually Show About AI and Creativity?

The most influential study on this topic comes from researchers Anil R. Doshi at UCL School of Management and Oliver Hauser at the University of Exeter. Their paper, published in Science Advances in July 2024, assigned 300 participants to three groups: one with no AI help, one that received a single AI-generated starting idea, and one that could choose from up to five AI-generated ideas. The results were striking.

Writers who used AI produced stories that independent evaluators rated as more creative, better written, and more enjoyable to read. The effect was strongest among participants who scored lowest on baseline creativity assessments — AI essentially leveled the playing field, bringing weaker writers up to the standard of the most creative participants in the study.

But here is the catch. When researchers analyzed the full pool of AI-assisted stories, they found something alarming: the stories were significantly more similar to each other than those written without AI. Individual creativity went up, but collective diversity went down. The researchers described this as a "social dilemma" — each writer is individually better off using AI, but when everyone does it, the overall range of novel ideas shrinks.

This finding has been replicated across multiple contexts. A 2025 study published in the Journal of Creative Behavior by researchers at multiple universities found that ChatGPT-assisted creative work exhibited more uniform stylistic patterns and reduced conceptual diversity compared to purely human output. Another study from Georgetown University examining large language models confirmed what they called the "homogenizing effect" — LLM outputs converge around similar themes, vocabulary, and narrative structures because they draw from the same statistical patterns in their training data.

Why Does AI Reduce Creative Diversity in Teams?

Understanding why AI homogenizes creative output requires looking at how these tools actually work — and how humans interact with them.

Algorithmic monoculture is the root cause. Large language models like GPT-4, Claude, and Gemini are trained on massive but ultimately finite datasets. They learn to predict the most statistically likely next word or idea based on patterns in their training data. This means their suggestions naturally gravitate toward the median — the most common, most "acceptable" creative choices. When every team member draws from the same AI well, they all start with variations of the same central idea.

The anchoring effect compounds the problem. Behavioral research has long shown that people anchor to the first piece of information they receive. When a team starts a brainstorming session by asking AI for ideas, those AI-generated suggestions become the anchor point. Even when team members try to diverge, they tend to orbit around the AI's initial framing. A 2025 study from Cornell University found that AI suggestions made writing measurably more generic, with participants gravitating toward and building upon AI-generated starting points rather than developing truly independent ideas.

Motivation decay accelerates the cycle. Research from Zhejiang University published in Harvard Business Review, involving over 3,500 participants across four studies, found that using generative AI reduced intrinsic motivation by 11% and increased boredom by 20%. The mechanism is straightforward: AI removes the cognitively demanding elements of creative work — the struggle, the uncertainty, the moments of friction that produce breakthrough ideas. When work feels less challenging, people engage less deeply, and when they engage less deeply, they default to whatever the AI suggested rather than pushing further.

Skill atrophy creates long-term damage. Perhaps the most concerning finding comes from a 2025 ScienceDirect study that tracked students over time. Researchers found that participants who initially relied on ChatGPT continued to produce less diverse, AI-like content even when working without AI afterward. Their individual creativity reverted to baseline levels, but the homogeneous patterns persisted — suggesting that heavy AI use may actually train people to think more like AI over time.

How Does AI Homogenization Affect the Workplace and Innovation?

The implications of AI-driven creative homogenization extend far beyond writing exercises in research labs. For employers, hiring managers, and workers navigating the 2026 job market, this trend carries real consequences.

Innovation pipelines are narrowing. When product teams, marketing departments, and strategy groups all feed their brainstorming through the same AI tools, the range of ideas entering the pipeline shrinks. A survey by Deloitte found that 91% of respondents who reported strong AI outcomes also hired for varied skill sets — suggesting that organizations already recognize the link between diversity and AI effectiveness. But many companies are moving in the opposite direction, using AI to standardize rather than diversify their creative processes.

Job roles are shifting toward curation, not creation. As AI handles more of the initial ideation work, the most valuable human skill is becoming the ability to recognize when AI output is too generic and push beyond it. This represents a fundamental shift in what employers need. The World Economic Forum's 2025 Future of Jobs Report identified creative thinking and analytical thinking as the two most in-demand skills globally — but the kind of creativity that matters now is not idea generation from scratch. It is the ability to evaluate, challenge, and diversify AI-generated starting points.

Team composition matters more than ever. Research published in Frontiers in Psychology in 2025 found that teams where only some members had access to AI actually outperformed teams where everyone used AI. The selective use of generative AI by specific team members helped generate diverse cognitive inputs and reduced homogeneous ideas. This has direct implications for how organizations structure their teams and distribute AI tools.

62% of creative directors report less diversity. According to industry surveys, nearly two-thirds of creative leadership has noticed decreased conceptual diversity in team outputs following AI adoption. This is not a theoretical concern — it is showing up in the work that organizations are producing right now, from advertising campaigns that feel interchangeable to product features that look like they came from the same template.

When Does AI Help Creativity — and When Does It Hurt?

Not all AI use leads to homogenization. The research points to specific contexts where AI genuinely enhances creative work and others where it actively undermines it.

AI helps most when used by individuals working on well-defined problems. The Science Advances study found that AI had the greatest positive impact on writers who started with lower baseline creativity. For individual workers tackling specific, bounded tasks — writing a product description, drafting an email, summarizing research — AI consistently improves output quality without the homogenization risk that emerges in group settings.

AI hurts most when used as the starting point for group ideation. The convergence problem is worst when entire teams begin their creative process by asking AI for ideas. Instead of each person bringing their unique perspective, experience, and cognitive style to the table, everyone begins with the same AI-generated foundation. The result is a narrower range of ideas that all sound vaguely similar.

AI helps when it is used to expand, not replace, human thinking. Research from the ACM Conference on Creativity and Cognition found that AI was most beneficial when used after humans had already generated their own ideas independently. In this configuration, AI served as a tool for expansion — offering adjacent possibilities that humans might not have considered — rather than as a replacement for original thought.

AI hurts when teams stop measuring diversity of output. Most organizations track productivity metrics — how many ideas generated, how fast deliverables are completed, how polished the final product looks. Very few track the diversity of their creative output. Without that measurement, homogenization creeps in unnoticed because every individual piece of work looks good on its own. It is only when you compare across a portfolio that the narrowing becomes visible.

How Can Teams Stay Creative While Using AI in 2026?

The solution is not to abandon AI — it is to use it strategically. Here are seven evidence-based approaches that organizations and individuals can implement immediately.

1. Adopt a "human-first, AI-second" ideation process. Before opening any AI tool, require team members to spend 15 to 20 minutes generating ideas independently. Write them down, sketch them out, or discuss them in pairs. Only after this initial divergent thinking phase should AI be introduced — and only as a tool to expand on human-generated ideas, not to replace them. This preserves the cognitive diversity that comes from different people thinking about the same problem in different ways.

2. Distribute AI access unevenly across teams. The Frontiers in Psychology research suggests that teams perform best when AI access is selective rather than universal. Designate specific team members as "AI scouts" who use generative tools to surface possibilities, while others focus on purely human ideation. This creates natural cognitive diversity within the team and prevents the convergence that happens when everyone starts from the same AI-generated foundation.

3. Measure collective novelty, not just individual output quality. Most organizations evaluate creative work one piece at a time. Start tracking the diversity of your team's output across projects. Use simple metrics like semantic distance between ideas, the range of approaches proposed, or the number of genuinely distinct concepts in a brainstorming session. If everything your team produces starts to sound the same — even if each individual piece is polished — you have a homogenization problem.

4. Rotate AI tools and prompt strategies regularly. Different AI models have different biases and tendencies. Teams that rely exclusively on one tool will converge around that tool's particular patterns. Rotate between different AI platforms, vary your prompting approaches, and periodically introduce constraints that force the AI to generate ideas outside its comfort zone. Ask it to think like a specific industry outsider, apply frameworks from unrelated fields, or generate deliberately contrarian positions.

5. Build "creative friction" into your workflow. The Zhejiang University research showed that AI reduces the cognitive effort that drives deep engagement. Counteract this by deliberately introducing friction into your creative process. Set constraints that AI cannot easily solve. Assign "devil's advocate" roles that challenge AI-generated ideas. Create stages where team members must justify why their idea is different from what AI would suggest. This friction is not inefficiency — it is the mechanism that produces genuinely novel thinking.

6. Invest in human skills that AI cannot replicate. The most AI-resistant creative skills are those rooted in lived experience, cultural context, and emotional intelligence — areas where AI has no genuine understanding. Encourage team members to develop deep domain expertise, cultivate diverse life experiences, and practice forms of creativity that draw on personal perspective rather than pattern matching. These human-only inputs are what prevent teams from producing AI-flavored output.

7. Audit your creative output regularly. Schedule quarterly reviews where your team examines its body of work for signs of homogenization. Compare recent output to work produced before AI adoption. Look for narrowing vocabulary, converging visual styles, or recurring structural patterns. If you spot these trends early, you can adjust your AI usage before the creative damage compounds. Remember the ScienceDirect finding: once homogeneous patterns take hold, they can persist even after AI is removed.

People Also Asked

Q: Does AI actually make you less creative?

A: Not individually — research shows AI can boost the creativity of individual work, especially for people who start with lower baseline creative skills. The problem is collective. When entire teams use AI, their combined output becomes more homogeneous because everyone anchors to the same AI-generated patterns. The key is how you use AI, not whether you use it.

Q: What is the AI homogenization problem?

A: AI homogenization refers to the tendency of AI-assisted creative work to converge around similar themes, vocabulary, and ideas. Because large language models generate output based on statistical patterns from their training data, their suggestions naturally gravitate toward the most common and conventional responses. When multiple people use the same AI tools, their work begins to look and sound alike — reducing the diversity of ideas that drives innovation.

Q: How can I use AI without losing my creative edge at work?

A: Always generate your own ideas first before consulting AI. Use AI to expand and challenge your thinking rather than to replace it. Rotate between different AI tools to avoid anchoring to one model's patterns. Set deliberate constraints that force you to think beyond what AI suggests. And regularly compare your recent work to your pre-AI output to check for signs of narrowing creativity or recurring patterns.

Related Articles

• Why AI Won't Replace Most Jobs in 2026 — But Will Reshape How You Work
• AI Is Reshaping Every Job
• HBR Research: How AI Is Actually Changing the Labor Market
• AI Hype vs. Reality: How Artificial Intelligence Is Changing White-Collar Work
• AI Is Coming for Entry-Level Jobs First — What New Grads Should Know
• AI Isn't Replacing That Many Jobs — Yet
• In a Jobs Apocalypse, Look to AI-Proof Skilled Trades
• 10 Jobs Most at Risk of AI Replacement in 2026
• The 10 Highest-Paying Jobs Resilient to AI Disruption
• Your Role Was Eliminated — Your Capability Wasn't: Career Resilience in 2026

Future-proof your career by staying ahead of how AI is changing the workplace. Metaintro delivers the latest insights on AI, hiring trends, and career strategy straight to your inbox. Sign up for free today and make sure you are using AI to stand out — not blend in.