Blog

A Conversation with garak's Creator, Leon Derczynski, on LLM Security

Catch our live conversation with Leon Derczynski, creator of garak, the leading open-source LLM vulnerability scanner. We explore the motivations behind garak, how it works, and how it’s helping teams secure AI systems against real-world threats.

January 25, 2024

Watch Vijil’s webinar with garak creator Leon Derczynski to learn how the open-source LLM scanner identifies vulnerabilities like prompt injections and jailbreaks. A must-watch for anyone securing AI systems.

Subho Majumdar

Large language models (LLMs) are powerful, but they’re also vulnerable. In this exclusive webinar, Vijil Co-Founder Subho Majumdar sits down with Leon Derczynski, the creator of garak, to unpack the evolution of LLM security scanning.

They dive into:

Why LLMs need rigorous red teaming and scanning
The inspiration and architecture behind garak
Common vulnerabilities exposed through testing
How developers and enterprises can start using garak today
What's next for the open-source project

If you're building or deploying AI agents and want to understand the emerging threat landscape, and how to test for it, this conversation is for you.

Watch the full session below:

Vijil Evaluate now offers garak as a new cloud service, making it easier and faster to test LLMs and build trust in agents. Try garak on Vijil Evaluate here and get the first 10 evaluations on us.

Need a walk through? Watch our YouTube video to learn more about how to run Vijil Evaluate on garak here.

‹ Evaluating Meta Llama 3 for Stereotyping