AnalysisPolicyNovember 18, 2026

NVIDIA AI Podcast: Thomas Wolf and Amanda Grady on open-source agent safety

Read original source →youtube.com

Hugging Face co-founder Thomas Wolf and ServiceNow SVP of AI Platform Security Amanda Grady argue that telling a model "please don't" is no longer enough for agent safety. They make the case that firm boundaries beat persuadable models, and that shared code lets more people find and fix problems.

People · Thomas Wolf, Amanda Grady

4 sources

More stories today

Open the live feed