Securing Generative AI - Back to the Future Edition

Securing Generative AI - Back to the Future Edition

Explore the security challenges of generative AI in this 34-minute conference talk from LASCON. Drawing parallels to the alternate 1985 in "Back to the Future 2," delve into concerns surrounding ChatGPT, including data handling, PII and IP leakage, LLM manipulation attacks, and legal issues. Discover a trust framework specifically designed for generative AI, incorporating access control, authenticity, auditing, and data protection. Learn how to navigate the complexities of establishing trust with GPT systems, using familiar solutions to address these emerging challenges in AI security.