I will discuss Membership Inference Attacks, which are designed to determine whether a specific data point was part of a model's training set. The effectiveness of these attacks shows how much information a model inadvertently leaks about its training data. Such attacks form the foundation of privacy auditing in machine learning.