=Coffee

2026-02-16 · 56 min · 7 entities

episode page audio

Asserted relationships

→ discusses Business company

0.40

evidence rules-v5
Feed category: Business
→ discusses Technology company

0.40

evidence rules-v5
Feed category: Technology
→ discusses True Crime podcast

0.40

evidence rules-v5
Feed category: True Crime

Entities found in this episode

companys 5

mentioned HiddenLayer company

0.62

evidence rules-v5
researcher at HiddenLayer
mentioned Business company

0.50

evidence rules-v5
Feed category: Business
mentioned Technology company

0.50

evidence rules-v5
Feed category: Technology
discusses Business company

0.40

evidence rules-v5
Feed category: Business
discusses Technology company

0.40

evidence rules-v5
Feed category: Technology

podcasts 2

mentioned True Crime podcast

0.50

evidence rules-v5
Feed category: True Crime
discusses True Crime podcast

0.40

evidence rules-v5
Feed category: True Crime

Episode description as stored

A lot of modern AI models have a kind of security guard layer that sits in front of them. Its job? A binary choice as to whether the prompt heading into the model is safe or not. Kasimir Schulz, a lead security researcher at HiddenLayer, has been researching how to trick these models. Their solution, a technique called "Echogram" involves words with such positive statistical sentiment — such overwhelming good vibes — that it flips that verdict. Learn more about your ad choices. Visit podcastchoices.com/adchoices