EP150 Taming the AI Beast: Threat Modeling for Modern AI Systems with Gary McGraw

Guest:

• Dr Gary McGraw, founder of the Berryville Institute of Machine Learning

Topics:

• Gary, you've been doing software security for many decades, so tell us: are we really behind on securing ML and AI systems?

• If not SBOM for data or "DBOM", then what? Can data supply chain tools or just better data governance practices help?

• How would you threat model a system with ML in it or a new ML system you are building?

• What are the key differences and similarities between securing AI and securing a traditional, complex enterprise system?

• What are the key differences between securing the AI you built and AI you buy or subscribe to?

• Which security tools and frameworks will solve all of these problems for us?

Resources:

• EP135 AI and Security: The Good, the Bad, and the Magical

• Gary McGraw books

• "An Architectural Risk Analysis Of Machine Learning Systems: Toward More Secure Machine Learning" paper

• "What to think about when you're thinking about securing AI"

• Annotated ML Security bibliography

• Tay bot story (2016)

• "Can you melt eggs?"

• "Microsoft AI researchers accidentally leak 38TB of company data"

• "Random number generator attack"

• "Google's AI Red Team: the ethical hackers making AI safer"

• Introducing Google's Secure AI Framework