The State of Adversarial Machine Learning and Its Relevance to Cybersecurity


  • Kai Mackall
  • Earl Pike
  • Chase Smith
  • Justice Thompson


Machine learning has been utilized for a number of applications in both the public and private sectors and is slated to become even more widely used in years to come. For this reason, the importance of security in this field cannot be understated. There are various ways to exploit and trick machine learning models into behaving in unexpected or unfavorable ways. Adversarial machine learning, the field of study of such methods, can be understood as a form of cyber threat intelligence, which describes the tools, knowledge, and experiences used to better understand cyber-related threats. We will further explain the security implications of adversarial machine learning, evaluate current progress in the field, and discuss a recent example of a related attack and its impact.





Computer Science