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

No ratings

Presented at t2 Infosec 2024 by

Jossef Kadouri

In a world where generative AI and Large Language Models (LLMs) have become integral to business operations, companies are confronted with a unique set of challenges. In this talk, we will demonstrate how AI models are built, with a focus on their vulnerabilities. We will then give a live demo of how an attacker might exploit these weaknesses to build malicious models and publish them to be consumed by victims. In addition, we'll be addressing takeaways related to the use of generative AI and LLMs within companies: What challenges do companies face today when it comes to employee use of generative AI and LLMs? How can companies understand their company’s specific risks regarding generative AI and LLMs, and best mitigate against them? Can companies benefit from the productivity gains of generative AI and LLMs while still meeting compliance, privacy, and security standards?