Can We Rely on AI?
This is an incremental review paper addressing the problem of AI reliability for researchers in applied and computational mathematics.
The paper provides an overview of adversarial attacks in deep learning, highlighting their role in exposing instabilities that affect safety and reliability in AI, particularly in high-risk settings, without presenting new experimental results or specific numerical findings.
Over the last decade, adversarial attack algorithms have revealed instabilities in deep learning tools. These algorithms raise issues regarding safety, reliability and interpretability in artificial intelligence; especially in high risk settings. From a practical perspective, there has been a war of escalation between those developing attack and defence strategies. At a more theoretical level, researchers have also studied bigger picture questions concerning the existence and computability of attacks. Here we give a brief overview of the topic, focusing on aspects that are likely to be of interest to researchers in applied and computational mathematics.