Modern AI models perform strongly on many established benchmarks yet still fail at tasks that humans find trivial, like manipulating a string or drawing a dog with five legs. These examples suggest that existing benchmarks may under-measure persistent blind spots in current systems.
We introduce \texttt{blind-spots-bench}, a benchmark designed to expose such blind spots through tasks that appear simple for humans but remain challenging for modern AI. We collect raw questions from students in an AI course, clean and annotate them with structured reference solutions, and propose a task taxonomy tailored to the resulting dataset of 235 samples.
We further develop an automated grading pipeline to evaluate a wide range of models, including open-weight and closed-source language, vision-language, and image-generation models. Our analysis on \texttt{blind-spots-bench} reveals that closed-source frontier models can substantially outperform open-weight models with even an approximate 10\% gap, even when they attain comparable performance on existing benchmarks.
A more fine-grained analysis shows that no single model dominates across all task types, and that some tasks remain challenging for all evaluated models. These results highlight the value of \texttt{blind-spots-bench} as a diagnostic stress test for identifying concrete weaknesses in current modern models.
Blogger's Review: This research introduces a blind spot benchmark that offers a fresh perspective on evaluating multimodal models, emphasizing the limitations of existing assessment methods. By revealing deficiencies in models on seemingly simple tasks, it pushes for continuous improvement of AI systems, which is noteworthy.