SIGNAL//SYNTH
Ai

Can Open Source Make AI Agents Safer?

aired Nov 18, 2026 · 29.0m
Signal
90.0/ 100
Essential
confidence 0.99
Orig54.2
Actn100.0
Dens100.0
Dpth100.0
Clty75.7
Summary

Let's talk about AI safety. I couldn't think of two better guests to discuss this with me today than Thomas Wolfe, physicist turned patent lawyer turned co-founder of Hugging Face, and Amanda Grady, the SVP of AI Platform Security at ServiceNow.

Why listen

It goes beyond the title with direct discussion of think, know, like, including: I couldn't think of two better guests to discuss this with me today than Thomas Wolfe, physicist turned patent lawyer turned co-founder of Hugging Face, and Amanda Grady, the SVP o.

Key takeaways
  1. 01You would tell a model, please don't do this thing that you're not supposed to do
  2. 02And I want to hear from you, Amanda, to start how you think about AI and agent safety and what you need around the model beyond just telling it what it shouldn't do
  3. 03we all in the security industry really firmly believe that we cannot rely solely on LLM-based controls because ultimately, you know, models can be persuaded, but firm boundaries ca
Best for
listeners looking for a practical AI episode debrief