Mitigating Risks in Generative AI: Attack Vectors and Defense Strategies

Join us for an insightful webinar presented by Dr. Neil Daswani, Co-Director of the Stanford Advanced Cybersecurity Program and author of "Big Breaches: Cybersecurity Lessons for Everyone." In this session, Dr. Daswani will provide an in-depth analysis of recent vulnerabilities, breaches, and abuses affecting leading Generative AI (GenAI) systems, including OpenAI and Mistral.

You will learn about major GenAI vulnerabilities and abuses such as jailbreaks, prompt injections, hallucinations, adversarial examples, and deep fakes. Additionally, we will explore how these vulnerabilities can be protected and defended against. You will gain a clear understanding of some of the current capabilities and limitations of GenAI. Practical steps for enhancing GenAI security, applicable to both consumers and enterprises, will also be discussed.

Moderator:

• Tas Jalai, Chair, AI Subcommittee, Head of Cybersecurity, AC Transit, Oakland CA

Speaker:

• Dr. Neil Daswani, Director Stanford Advanced Cybersecurity, CISO QuantumScape