Transparent labeling of training data may boost trust in artificial intelligence

Showing users that visual data fed into artificial intelligence (AI) systems was labeled correctly might make people trust AI more, according to researchers. These findings may also pave the way to help scientists better measure the connection between labeling credibility, AI performance, and trust, the team added.

ai performance artificial artificial intelligence boost data intelligence labeling machine learning & ai people performance researchers scientists systems team training training data trust visual data