Plain-AyI
← All projects
AI testing & safety

Adversarial Robustness Toolbox (ART)

A Python toolkit for testing how AI models hold up against attacks, and for defending them.

New this week

What it does

ART is a library for machine-learning security. It helps developers and researchers check how AI models can be fooled or misused, and add defences.

It covers four kinds of threat: tricking a model with altered inputs, corrupting its training data, copying a model, and working out private information from it. It includes both attack methods for testing and defence methods for protection, for so-called red and blue teams.

It works with the main machine-learning frameworks and with images, tables, audio and video. It is hosted by the Linux Foundation's AI & Data group.

Who it's for

AI security researchers and teams who need to test and harden their machine-learning models.

Can I use it without coding?

Developers only. It is a Python library used in code. New to this? Start here.

Can I run it?

Cost
FreeNothing to pay or sign up for.
Privacy
Runs on your own device: your data stays with you
API key
Not needed

Try it

Using Claude Code, Cursor or another AI coding assistant? Paste the prompt and it will walk you through the setup.New to this?

Before you install

  • Only download from the official GitHub page. Copies with similar names exist for many popular projects.
  • License (MIT): Very permissive: free to use, change and share, including for business.
  • More on staying safe

Where people are talking about it