AI news ·
Is AI Really Creative? How Seeing the Process Shapes Our Judgment of Creativity
People judge AI creativity higher when they see more of its process and creation. This reveals how perception shapes views on AI and human creativity alike.
Is AI Truly Creative? Turns Out Creativity Is In The Eye Of The Beholder
May 11, 2025
Eurasia Review
What makes people believe an AI system is creative? Recent research reveals that it depends largely on how much of the creative act people observe. This insight has important implications for designing and researching creative AI systems and raises questions about how we perceive creativity in others.
AI increasingly participates in creative activities. But should we call it creative? That remains a separate question. Researchers from Aalto University and the University of Helsinki explored whether people judge a robot as more creative when they see more of its creative process.
Participants first rated creativity based only on still life drawings produced by robots. They were told the drawings were AI-generated, but in reality, the robots were programmed to replicate commissioned artist drawings. This approach controlled for variability in creativity while measuring perception.
Next, participants viewed a video showing the drawing process—lines appearing on the page—but without seeing the robot itself. Finally, they saw the complete picture: the final drawing, the drawing process, and the robot creating it. Results showed creativity ratings increased as more aspects of the creative act were revealed.
“The more people saw, the more creative they judged it to be,” said one of the lead researchers. This study is among the first to analyze how product, process, and producer individually and collectively affect creativity perception—not just for AI, but generally.
The Power of Perception
Knowing how people assess creativity in robots and artificial systems is crucial for their design. The study suggests revealing more about the creative process and producer can boost the perceived creativity of AI systems. But is that always positive?
Enhancing perception might help users stay engaged with co-creative systems. However, it could also create misleading impressions about an AI’s true creative capabilities. Recognizing this tension highlights the role of human biases in evaluating creativity and stresses the importance of transparent design.
These findings also impact research methodologies. If creativity judgments depend on presentation, future studies must control for how AI systems are shown. Past comparisons between systems might need revisiting to avoid misleading conclusions caused by differences in presentation.
Interestingly, this research prompts reflection on human creativity perception too. Does how we see the act, product, and producer influence how we judge creativity in other people? This question opens new avenues for exploring creativity perception beyond AI.
Does Shape Matter?
The study also tested whether the robot’s physical design influences creativity ratings. Two robot types were used: a sleek arm-like robot and a more mechanical plotter. Keeping drawings consistent across robots and participants was challenging but necessary for a fair comparison.
Surprisingly, no significant difference emerged in creativity scores between the two robot shapes. This counterintuitive result encourages further investigation into what factors truly shape our judgments of creativity in embodied AI systems.
Researchers plan to explore additional biases affecting evaluations of creative AI and how these effects occur. They also aim to test these findings across various artistic genres and creative forms. The study followed open science practices to facilitate replication and further research.
As AI systems become more common collaborators, understanding what influences our perception of their creativity matters for practical design and offers insight into how we recognize creativity in humans.