| Description |
1 online resource (1 video file (2 hr., 1 min.)) : sound, color. |
| Playing Time |
020100 |
| Description |
digital rdatr |
|
video file rdaft |
|
Instructional films lcgft |
| Performer |
Clint Bodungen, instructor. |
| Summary |
LLMs introduce new attack vectors that can compromise your AI systems. This intensive workshop equips you with hands-on skills to tackle the OWASP Top 10 most critical risks for securing enterprise-grade LLM applications. Led by cybersecurity expert Clint Bodungen, this masterclass focuses on fortifying your LLM stack against the OWASP Top 10 most critical risks. Diving deep into the attack vectors unique to these powerful generative models, you will learn hands-on techniques to safeguard your apps built on large language models. The workshop covers a wide range of practical methods to harden your LLM security posture. You will discover how to protect against supply chain attacks through vulnerable third-party code, libraries, models and plugins. The session outlines processes to prevent unauthorized data access, theft of proprietary data, and poisoning of your training dataset. Through interactive examples and sample code, you will grasp approaches to filter malicious user input, sanitize model outputs, and implement robust validation mechanisms. The workshop specially focuses on skill-building around prompt engineering as a powerful mechanism to keep generative models restricted within secure guardrails. What you will learn How to safeguard your LLM apps from supply chain vulnerabilities Ways to prevent data poisoning, unauthorized access, and theft Techniques to filter malicious user input and sanitize model output Methods to block jailbreaking and misuse of your LLMs Tools and frameworks to automate security mechanisms in your stack Audience Developers, data scientists, and security professionals seeking to fortify their enterprise-grade large language model (LLM) applications against cybersecurity threats. This workshop is designed for individuals interested in hands-on learning to secure LLMs and mitigate risks outlined in OWASP's Top 10. About the Author Clint Bodungen: Clint Bodungen is a globally recognized cybersecurity authority and brings over a quarter-century of experience to the table. A veteran of the United States Air Force and seasoned professional at notable cybersecurity firms like Symantec, Kaspersky Lab, and Booz Allen Hamilton, he is renowned for his innovative approaches in the field. Clint has contributed to the field as the author of two insightful books: 'Hacking Exposed: Industrial Control Systems' and 'ChatGPT for Cybersecurity Cookbook.' These works underscore his wide-ranging knowledge and expertise in cybersecurity, establishing him as a thought leader in this ever-evolving field. |
| Subject |
Computer networks -- Security measures.
|
|
Artificial intelligence -- Computer programs -- Security measures.
|
| Genre |
Instructional films.
|
|
Nonfiction films.
|
|
Internet videos.
|
| Added Author |
Bodungen, Clint E., presenter.
|
|
Packt Publishing, publisher.
|
| ISBN |
9781835880746 (electronic video) |
|
1835880746 (electronic video) |
|
More Resources