Skip to content
Lenny's PodcastLenny's Podcast

Sander Schulhoff: Why AI guardrails fail every red team test

How prompt injection and jailbreaks bypass state-of-the-art guardrails; agents that send emails or touch databases turn every bypass into real damage.

Sander SchulhoffguestLenny Rachitskyhost
Dec 21, 20251h 32mWatch on YouTube ↗

CHAPTERS

  1. 0:00 – 5:14

    Introduction to Sander Schulhoff and AI security

  2. 5:14 – 11:42

    Understanding AI vulnerabilities

  3. 11:42 – 17:55

    Real-world examples of AI security breaches

  4. 17:55 – 19:44

    The impact of intelligent agents

  5. 19:44 – 21:09

    The rise of AI security solutions

  6. 21:09 – 23:44

    Red teaming and guardrails

  7. 23:44 – 27:52

    Adversarial robustness

  8. 27:52 – 38:22

    Why guardrails fail

  9. 38:22 – 44:44

    The lack of resources addressing this problem

  10. 44:44 – 55:49

    Practical advice for addressing AI security

  11. 55:49 – 59:06

    Why you shouldn’t spend your time on guardrails

  12. 59:06 – 1:09:15

    Prompt injection and agentic systems

  13. 1:09:15 – 1:11:47

    Education and awareness in AI security

  14. 1:11:47 – 1:17:52

    Challenges and future directions in AI security

  15. 1:17:52 – 1:21:57

    Companies that are doing this well

  16. 1:21:57 – 1:32:40

    Final thoughts and recommendations

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome