Glossary · Generative AI and LLMs
What is Red-teaming? Definition and meaning
Systematically testing a model or application with adversarial inputs to find unsafe or unintended behaviour.
Definition
Red-teaming: Systematically testing a model or application with adversarial inputs to find unsafe or unintended behaviour.
In short: Probing a system for failures like an attacker would.
When it is used
Finding safety and security weaknesses before release.
Where it fits
Red-teaming belongs to LLM safety in the Generative AI and LLMs topic. These are the other terms in the same group:
- Guardrails AI: A framework that checks model responses against validators for structure, safety and content rules.
- NeMo Guardrails: A toolkit that adds topic, safety and security rails to LLM applications using a dedicated rule language.
- Presidio: Microsoft's open-source toolkit for finding and anonymizing personally identifiable information in text and images.
Learn more
The full guide to Red-teaming explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Recurrent neural networks · Next term: Redis