A Look at AI Security with Mark Russinovich скачать в хорошем качестве

A Look at AI Security with Mark Russinovich

Title: A Look at AI Security with Mark Russinovich Date: December 3, 2025 Duration: 1HR ABSTRACT Join Mark Russinovich as he explores the evolving landscape of generative AI risks and safeguards, with a focus on large language models. We’ll take a close look at three fundamental vulnerabilities inherent in these systems—hallucination, indirect prompt injection, and jailbreaks (or direct prompt injection)—examining where they come from, how they can impact systems and users, and what strategies exist to mitigate them. Together, we’ll explore how to harness the immense potential of LLMs while responsibly managing the risks that come with them. SPEAKER Mark Russinovich CTO, Microsoft Azure MODERATOR Scott Hanselman Vice President of Developer Community, Microsoft; ACM Practitioner Board