🛡️ Breaking Agent Backbones: Evaluating LLM Security in AI Agents

episode

Build Wiz AI Show not yet transcribed

▲ 0

Transcript

This episode hasn't been transcribed yet Upvotes decide what gets transcribed next — 0 so far. Sign in to upvote

Related

recent transcribed audio

Your partner wants profit. You want freedom. Now what? 22 min · 16 Sep 2026 · The Optometry Success Podcast

Board of County Supervisors Meeting - Work Session - Sep 15, 2026 3h 17m · 15 Sep 2026 · Prince William County Board of Supervisors

Finance and Budget Committee - Sep 15, 2026 1h 43m · 15 Sep 2026 · Prince William County Finance and Budget Committee

The Art of True Happiness | Dr. Arthur Brooks 2h 49m · 14 Sep 2026 · Huberman Lab

The Chris Wolfe Podcast: The Power of Quality Supplements 48 min · 14 Sep 2026 · EyeCode Media

NPR News: 09-13-2026 6PM EDT 4 min · 13 Sep 2026 · NPR News Now

Report any issue

Source

Recorded31 Oct 2025

Publisheranchor.fm ↗

Handlerec:12512865

Show notes

Breaking Agent Backbones: AI agents are being deployed at scale, but their security is challenged by non-deterministic behavior and novel vulnerabilities. This episode introduces the "threat snapshot" framework and the new b3 benchmark, which systematically isolate and evaluate security risks stemming from the backbone LLM. We reveal crucial findings: enhanced reasoning capabilities generally improve security, yet model size does not correlate with lower vulnerability scores.

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Sign in