David Hawthorne

O3 Solutions

BIO

David Hawthorne is the Director of Cloud Engineering at O3 Solutions, a Birmingham, AL-based company that develops advanced SaaS software for the industrial construction sector. With over 20 years of experience in the tech industry, David has held a variety of roles, including administration, development, and data & software architecture, cultivating a rich and diverse skill set. He is dedicated to mentoring the next generation of tech professionals and passionate about driving innovations in cloud technology, devops practices, and security.

TITLE

Navigating the New AI Security Landscape: OWASP Top 10 for LLMs

ABSTRACT

Recently, the OWASP Top 10 for LLMs has emerged as a pivotal resource for uncovering and mitigating the evolving security risks in the realm of generative AI. Join us for a dynamic 30-minute session that not only introduces the key concepts behind LLM-specific challenges but also draws compelling parallels with traditional web-era vulnerabilities. This talk is your gateway to understanding the critical intersections of cybersecurity and AI, equipping you with a strategic mindset to safeguard your innovative systems.

Geared towards developers, architects, product owners, and enterprise tech professionals, this engaging presentation explores the OWASP Top 10 for LLMs into three core security themes:

Input & Data Security: Understand how unvalidated inputs, weak boundaries, and data integrity issues can open the door to prompt injection, adversarial manipulation, and sensitive data exposure.

Asset Protection & Abuse Prevention: Explore how to defend against model misuse, API exploitation, and system compromise through effective access controls and protection strategies.

Trust, Governance & Transparency: Learn how resilient system design, strong observability, and governance structures foster responsible AI usage and organizational trust.

Rather than diving deep into technical implementations, this session provides a conceptual framework that transcends specific architectures, offering a strategic perspective that bridges traditional security paradigms with the emerging challenges of the AI era. Attendees will leave with a clear roadmap of where potential risks reside and actionable insights on how to defend against them, ensuring that their approach to LLM security remains both proactive and robust.