Private by Design and Secure by Default AI Products

In this masterclass you'll design an AI product from conception, through architecture, risk and threat modeling and into your deployment and testing plan, ensuring that privacy, transparency and security are built in. Along the way, you'll learn about common privacy and security anti-patterns in large-scale deep learning/AI systems and design better approaches that both communicate and enforce better trust. By putting on your product, design, risk, architect, engineer and hacker hats, you'll leave the room with a more holistic and multidisciplinary perspective.

Expect hands-on exercises (and some code!) to: • Discover privacy and security antipatterns in AI Product design • Identify and evaluate (regulatory) privacy risk in AI systems • Map data and user flows to identify potential privacy issues • Evaluate AI-specific privacy and security threats/attacks • Design and review architectures, informed by risk and threat analysis • Evaluate and integrate use case specific guardrails and other potential technological solutions (i.e. leading privacy technologies) • Build evaluation datasets and pipelines • Define and measure success You leave the class informed by the latest best practices and information around building privacy-first, secure AI systems -- and hopefully inspired to take some ideas directly back to your AI, software or platform engineering work.