717: Overcoming Adversaries with A.I. for Cybersecurity, with Dr. Dan Shiebler
Super Data Science: ML & AI Podcast with Jon Krohn
00:00
Harnessing Large Language Models in Cybersecurity
The chapter explores how large language models can be optimized for specific cybersecurity tasks through parameter efficient fine tuning, focusing on their role in message classification for identifying attacks in a phishing mailbox product. It stresses the importance of balancing false positives and false negatives in training machine learning models to mitigate serious consequences in cybersecurity tasks. The discussion also delves into the evolving threat landscape with attackers utilizing generative AI tooling for personalized attacks, raising concerns about the malicious use of open-source AI models like GPT-5 or GPT-6 in planning sophisticated cyber attacks.
Transcript
Play full episode
