Tell me Mr. AI, what do you see in this image?
Tommaso Giacometti, Paola Surcinelli, Mariachiara Stellato, Nico Curti

TL;DR
This study applies the Rorschach inkblot test to 61 AI models to compare their interpretative responses with human profiles, revealing systematic differences in semantic richness and affective load.
Contribution
It introduces a novel framework for evaluating AI perception using psycho-semantic variables inspired by the Rorschach tradition, highlighting biases in AI models.
Findings
AI responses are highly non-random with systematic semantic convergence.
Human responses show higher affective load and variability.
AI models favor formal coherence over affective or symbolic elaboration.
Abstract
Background: The Rorschach inkblots are ambiguous stimuli developed to evoke subjective interpretations in humans, while modern artificial intelligence (AI) models are trained to recognize well established patterns and classes. The comparison of these two opposite systems arises a simple and provocative question: what happens when we ask an AI model to interpret an inkblot that "is not supposed to represent anything predefined"? Methods: We submitted the complete set of ten Rorschach inkblots to 61 AI models pretrained on the ImageNet dataset, spanning multiple architectural families. Model predictions were analyzed at the level of top-ranked classes and were quantified using a selected set of psycho-semantic variables inspired by the Rorschach tradition. Statistical analyses examined the effects of model family, computational complexity, and image conditions, comparing model-generated…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
