Sign inSign in
ConceptAI & ML

Jailbreaking

A tool or method used to bypass safety guardrails in commercial language models.
as Rongzhe Wei defines itGoogle Research & TechTalks
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Language model safety1Large language models1

Lecturers

2

Each in their own terms, from the easiest level.

  • Matthias Scheutz

    Tufts University · Graduate

    Circumventing safety guards to make models output restricted information.
    • 37:57AI in Science and Engineering: Virtues and Vices of Robot Foundation Models

Connections

3

The lectures that define it or assume it, and the concepts next to it.

2 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.