Name: Hacking Generative AI and Language Models with AI Red Teaming and Beyond
Start: 2024-06-20T18:00:00-07:00
End: 2024-06-20T20:00:00-07:00
Location: Loma Hall

Abstract:
Large language models have revolutionized natural language processing, but their expanding capabilities also raise concerns about vulnerabilities and potential attacks. In this presentation, we embark on a journey to explore the frontiers of large language models, unveiling attack strategies and discussing effective safeguards. We showcase real-world examples of adversarial attacks, highlighting their impact on model integrity and reliability. Moreover, we delve into state-of-the-art research and best practices for fortifying models against attacks. Ethical considerations and responsible AI practices are also addressed. Join us to gain valuable insights into the evolving landscape of large language models and ensure their responsible and secure use.

Speaker:
Gaspard Baye is a PhD candidate and a security AI scientist with over 5+ years of experience developing AI-driven defensive security applications. He has been recognized with several research publications in prestigious conferences such as NeurIPS, HASP and IEEE Access, accumulating 47+ citations. He is a recognized CVE holder with certifications, including OSCP, PNPT, Scrum, NSE1, NSE2 and CEH Practical. His work has been showcased at cybersecurity conferences such as DEFCON, BSides, and The Diana Initiative. Through a dedicated vulnerability disclosure program, he's identified and helped remediate over 20+ critical security vulnerabilities, earning Hall of Fame recognitions from Nokia and Ford Motors.

PopPC

George Callow

Open Web Application Security Project San Diego (OWASP-SD)

Movements & Politics

Web Security

New Technology

Network Security

OWASP

Information Security

Software Security

Computer Security

Cloud Security

Web Application Security

Every 3rd Thursday of the month until November 19, 2025

Hacking Generative AI and Language Models with AI Red Teaming and Beyond

Loma Hall

Share

Open Web Application Security Project San Diego (OWASP-SD)

Hacking Generative AI and Language Models with AI Red Teaming and Beyond

Open Web Application Security Project San Diego (OWASP-SD)

Details

Members are also interested in