2024-07-15 - Lecture at the ARTISAN summer school Towards Machine Learning Models that We Can Trust: Hacking and (properly) Testing AI