• MCP framed as conventional containerized API with familiar risks
• Offensive AI Con format, focus, and key trends in automation
• LLMs for code triage, bug patterning, and limits on complex targets
• Reinforcement learning environments as scalable, shareable training
• Hallucination risks, verification loops, and provenance checks
• Garak architecture, probes for XSS and Jinja template injection
• CWE mapping for actionable remediation and controls
• Jailbreak limits, alignment realities, and output safeguards
• Career resilience: depth in niches, broad literacy, measured skepticism
Find me by searching “Eric Galinkin.” Garak: github.com/NVIDIA/garak — Discord link in the repo. It’s also on PyPI: pip install garak. Our docs are okay and getting better
Chapters
00:00 Introduction and Context Setting
02:43 Exploring Offensive AI Conference
05:16 AI in Red Teaming and Security
08:10 The Role of AI Models in Security
10:41 Challenges in AI and Security Integration
13:29 Garak: A New Vulnerability Scanner
16:00 Jailbreaking and Security Implications
18:44 Future of AI in Security and Job Market
21:20 Skills for the Future in Cybersecurity
Affiliates
➡️ OffGrid Faraday Bags: https://offgrid.co/?ref=gabzvajh
➡️ OffGrid Coupon Code: JOE
➡️ Unplugged Phone: https://unplugged.com/
Unplugged's UP Phone - The performance you expect, with the privacy you deserve. Meet the alternative. Use Code UNFILTERED at checkout
*See terms and conditions at affiliated webpages. Offers are subject to change. These are affiliated/paid promotions.
Tesla Referral Code: https://ts.la/joseph675128