Image classification algorithms are notoriously error-prone, but a novel method for spotting errors within incomprehensible AI code could help solve the problem.

When something does zero-shot image classification, that means it’s able to make judgments about the contents of an image without the user needing to train the system beforehand on what to look ...