By Michael D. Brown

We are offering AI/ML safety and security training in summer and fall of this year!

Recent advances in AI/ML technologies opened up a new world of possibilities for businesses to run more efficiently and offer better services and products. However, incorporating AI/ML into computing systems brings new and unique complexities, risks, and attack surfaces. In our experience helping clients safely and securely deploy these systems, we’ve discovered that their security teams have knowledge gaps at this intersection of AI/ML and systems security. We’ve developed our training to help organizations close this gap and equip their teams with the tools to secure their AI/ML operations pipelines and technology stacks.

What you will learn in our training

Our course is tailored for security engineers, ML engineers, and IT staff who need to understand the unique challenges of securing AI/ML systems deployed on conventional computing infrastructure. Over two days, we provide a comprehensive understanding of Al safety and security that goes beyond basic knowledge to practical and actionable insights into these technologies’ specific dangers and risks. Here’s what you will learn through a blend of instructional training and hands-on case studies:

1. Fundamentals of AI/ML and cybersecurity: In this module, you will learn how AI/ML models/techniques work, what they can and cannot do, and their limitations. We also cover some essential information and software security topics that may be new for ML engineers.
2. AI/ML tech stacks and operations pipelines: In our second module, you will learn how AI/ML models are selected, configured, trained, packaged, deployed, and decommissioned. We’ll also explore the everyday technologies in the AI/ML stack that professionals use for these tasks.
3. Vulnerabilities and remediation: In this module, you will learn about the unique attack surfaces and vulnerabilities present in deployed AI/ML systems. You’ll also learn methods for preventing and/or remediating AI/ML vulnerabilities.
4. Risk assessment and threat modeling: The fourth module covers practical techniques for conducting comprehensive risk assessments and threat models for AI/ML systems. Our holistic approaches will help you evaluate the safety and security risks AI/ML systems may pose to end users in deployed contexts.
5. Mitigations, controls, and risk reduction: Finally, you will learn how to implement realistic risk mitigation strategies and practical security controls for AI/ML systems. Our comprehensive strategies address the entire AI/ML ops pipeline and lifecycle.

Equip your team to work at the intersection of security and AI/ML

Trail of Bits combines cutting-edge research with practical, real-world experience to advance the state of the art in AI/ML assurance. Our experts are here to help you confidently take your business to the next level with AI/ML technologies. Please contact us today to schedule an on-site (or virtual) training for your team. Individuals interested in this training can also use this form to be notified in the future when we offer public registration for this course!