AI red-teaming is a security exercise adapted from traditional cybersecurity practices, focused explicitly on challenging AI systems, including machine learning (ML) models and generative AI ...