The Dark Side of AI: Hidden Supply Chain Risks in Open-Source AI Models

Dive into a 25-minute conference talk that exposes critical security vulnerabilities in open-source AI models and their supply chains. Examine the architecture of AI tools while focusing on potential security weaknesses through multiple live demonstrations of model manipulation techniques. Learn from a detailed case study of the "Malicious Copilot" IDE plugin that reveals how code-completion models can be weaponized to target specific users and embed harmful code. Gain practical insights and defensive strategies for organizations implementing generative AI and Large Language Models (LLMs) in their development workflows.