machine learning security

This article explains how The Matrix’s idea of programs hacking programs mirrors real adversarial AI behaviour. It highlights risks, benefits and future implications of AI fighting, manipulating or improving other AI systems.