# 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](/topics/generative-ai-llms) topic. These are the other terms in the same group:

- **[Guardrails AI](/glossary/guardrails-ai)**: A framework that checks model responses against validators for structure, safety and content rules.
- **[NeMo Guardrails](/glossary/nemo-guardrails)**: A toolkit that adds topic, safety and security rails to LLM applications using a dedicated rule language.
- **[Presidio](/glossary/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](/tools/red-teaming)** explains how it works, when to use it and how it compares with the alternatives, with examples. The [glossary](/glossary) lists every term on the site from A to Z.

Previous term: [Recurrent neural networks](/glossary/recurrent-neural-networks) · Next term: [Redis](/glossary/redis)
