As generative AI becomes increasingly embedded in creative practice, concerns are growing about its impact on creative diversity. While AI tools may support individual creativity, they could simultaneously contribute to a broader creative monoculture in which ideas, designs and texts become more alike.
In a new pre-print study, Alwin de Rooij, associate professor within the CARADT Situated Art, Design and Technology research group, and assistant professor at Tilburg University, together with Michael Mose Biskjaer of Aarhus University, investigated this question through a systematic review and meta-analysis of 19 empirical studies on human-AI co-creation.
The results show a small but significant reduction in creative diversity when people work with generative AI systems. In other words: creative outputs produced when using AI tend to become more similar to one another compared to outputs created without AI.
The effect appeared strongest in more structured creative tasks, such as idea generation and problem-solving, where participants worked on challenges like improving public transport systems. In more open-ended divergent thinking tasks, such as inventing alternative uses for everyday objects, the effect was less pronounced.
The study also suggests that these homogenizing effects may persist for some time even after people stop using AI tools. The researchers also found early indications that these homogenizing effects are not just laboratory phenomena, but may also emerge in real-world creative practices.
The findings resemble well-known psychological “fixation effects,” where exposure to examples unconsciously steers people toward similar solutions, now amplified by the scale and synchronicity at which AI systems are used. Since many users rely on the same pre-trained AI models, optimized toward widely preferred responses, creative outputs may gradually converge.
Rather than suggesting that people become less creative, the research points to a gradual shift in how creative diversity is distributed when many people work with the same AI systems. While the overall effect is relatively small, the researchers stress that it becomes significant due to the widespread adoption of generative AI across education, design, art, writing and research.
The study contributes to ongoing discussions about the societal role of AI within creative and knowledge practices, a topic that strongly connects to CARADT’s focus on critical imagination, situated technology and practice-based research.
The pre-print paper is titled Does Generative AI Make Us Think Alike? A Systematic Review and Meta-Analysis of Homogenization Effects in Human-AI Co-Creation and is available via OSF.
Read the pre-print here:
OSF Pre-print Publication
Recent media coverage via PsyPost:
PsyPost article on the research
‘‘Understanding how creativity and imagination emerge from interactions with our environment will lead to improved innovation processes, tools and technologies.’’
Alwin de Rooij is Associate Professor in Situated Art, Design, and Technology at the Avans Centre of Applied Research for Art, Design and Technology (CARADT), and Assistant Professor in Creativity Research in the department of Communication and Cognition, Tilburg University.
Thank you for your subscription! Please check your email inbox to confirm.