A new paper on arXiv examines the common perception that AI-generated outputs are highly creative. The authors note that while these outputs are frequently judged as creative, they also tend to be more similar to each other—more homogenized—than human ideas.

The central question the paper addresses is whether the human heterogeneity that extends beyond AI's semantic blind spots reflects valuable thinking that AI lacks. Based on the paper's title, the answer appears to be that these human ideas are not inferior to AI-generated ones.

The research adds to the ongoing discussion about the nature of AI creativity and the value of human divergence from AI-generated patterns. It suggests that the differences between human and AI outputs may not be a weakness of human thinking, but rather a distinct form of value.