In this episode Michael and Sam are talking to malware researcher Dr. Marcus Botacin. Dr. Botacin discusses his journey from early work in sandboxing to advanced malware detection through machine learning, his recent efforts combining LLMs with GANs to create an iterative malware generation system that enhances evasion capabilities, and the importance of scaling defensive technologies to match the generative power of AI in offensive use cases.
Things Mentioned:
- With 'TPUXtract,' Attackers Can Steal Orgs' AI Models - https://www.darkreading.com/vulnerabilities-threats/tpuxtract-attackers-steal-ai-models
- GPThreats-3: Is Automatic Malware Generation a Threat? - https://marcusbotacin.github.io/publication/2023-05-01-paper-gpt-number-27
- SoK: All You Need to Know About On-Device ML Model Extraction - The Gap Between Research and Practice - https://marcusbotacin.github.io/publication/2024-05-01-paper-model-extraction-32
- Prospective Students and Research Ideas - https://marcusbotacin.github.io/research/
- HOU.SEC.CON. 2024 Talk: https://youtu.be/5lk_xklzcMg?si=oYyqMMyBUhY0ZbsZ
Do you have a question for the hosts? Reach out to us at podcast@houstonseccon.com
Keep up with HOU.SEC.CON.:
Check out our other show:
Check out our Conferences and Events:
Support or apply to our Scholarship Program:
Subscribe to:
In this episode:
· Host: Michael Farnum
· Host: Sam Van Ryder
· Guest: Marcus Botacin
· Production and editing: Lauren Andrus
· Music by: August Honey
