AI Security: New Risks, New Attack Vectors, New Defences

Join us for the July AI Security Collective meet-up, where we will explore new risks, new attack vectors and new defences.

This month we're joined by Dr. Peter Garraghan who will share his academic research on the subject of AI security. The research that led him to create the company Mindgard to support security teams as they navigate new attack vectors and emerging risks.

We'll then dive into an interactive discussion about Where to start as a security team wrestling with AI Sprawl, joined by expert panelist Sonya Moisset from Snyk.

Schedule:

• ​​1800: Doors Open
• ​​18:20: Kick Off Introduction
• ​​18:30: AI Red Teaming: Stress-Testing AI Systems for Enhanced Security by Dr. Peter Garraghan
• ​19:15: Panel Discussion: Where to start as a security team wrestling with AI Sprawl (Hannah Foxwell, Peter Garraghan and Sonya Moisset)
• 20:00: Networking
• ​​21:00: Home Time

Speaker:
Dr. Peter Garraghan is CEO & CTO of Mindgard, Professor in Computer Science at Lancaster University

Talk: AI Red Teaming: Stress-Testing AI Systems for Enhanced Security
This talk pulls back the curtain on "AI red teaming". We'll dive into how to approach AI red teaming to uncover hidden risks and attack routes. Sharing strategies for mitigating the risks you inevitably uncover. You'll learn practical ways to bake security testing right into AI development from the start, so you can build innovative solutions with AI that are also incredibly tough and ready for production. Plus, we'll share some real-world stories where this approach caught important problems, before it was too late!

• Examine the critical role of red teaming in identifying vulnerabilities and enhancing security of AI systems
• Explore various AI red teaming methodologies and their effectiveness in uncovering potential risks and attack vectors
• Discuss strategies for integrating red teaming practices into the AI development lifecycle without hindering innovation
• How do we balance thorough security testing with the rapid pace of AI advancement
• Share real-world case studies where AI red teaming has successfully improved system robustness and mitigated potential threats

Bio:
Dr. Peter Garraghan is CEO & CTO of Mindgard, Professor in Computer Science at Lancaster University, and fellow of the UK Engineering Physical Sciences and Research Council (EPSRC). He is an internationally recognised expert in AI security, Peter has dedicated years of scientific and engineering expertise to create bleeding-edge AI red teaming technology to understand and overcome growing threats against AI. He has raised over €11.6 million in research funding and published over 60 scientific papers, and has featured in media such as the BBC, Forbes, and CISO Online.

Expert Panel: Where to start as a security team wrestling with AI Sprawl

• Hannah Foxwell (Founder, AI for the rest of us)
• Peter Garraghan (Founder and CEO, Mindgard)
• Sonya Moisset (DevRel, Snyk)

Huge thanks for [ControlPlane.io](https://control-plane.io/) and [Mindgard.ai](https://mindgard.ai/) for supporting our community, providing the venue, food and drinks.