We’re formally sending out a petition to change the phrase “Jack of all trades” to “Hyrum of all trades” in honor of this episode’s guest, Hyrum Anderson. In this episode, hosts Natalia Godyla and Nic Fillingham sit down with Hyrum Anderson who, when he’s not fulfilling his duties as the Principal Architect of the Azure Trustworthy ML group, spends his time playing accordions, making cheese, and founding impressive technology conferences. He does it all!   Rather than chatting with Hyrum about a specific capability that he’s helped to develop, or a blog post that he co-authored – because, believe us, the episode would last for hours – we decided to have a chat with him about his life, how he first got into the world of technology, and his thoughts on the current state of cyber security.   In This Episode You Will Learn:   The differences between a risk and a threat  Why it’s easier to attack than defend  What a Principal Architect of the Azure Trustworthy ML group does in his spare time  Some Questions We Ask:   How does Hyrum think about adversarial machine learning and protecting A.I. systems?  What is it like for Hyrum to oversee both the red teaming and defensive side of operations?  Why are we better at finding holes in security than we are at making sure they don’t exist in the first place?     Resources:   Hyrum Anderson’s LinkedIn Hyrum Anderson’s Twitter Conference on Applied Machine Learning in Information Security (CAMLIS) Machine Learning Security Evasion Competition Microsoft Security Blog Nic’s LinkedIn Natalia’s LinkedIn    Related: Listen to: Afternoon Cyber Tea with Ann Johnson Listen to: Security Unlocked: CISO Series with Bret Arsenault  Security Unlocked is produced by Microsoft and distributed as part of The CyberWire Network.