AI Risk Database

The AI Risk Database is a tool for discovering and reporting the risks associated with public machine learning models. The database is specifically designed for organizations that rely on AI for their operations, providing them with a comprehensive and up-to-date overview of the risks and vulnerabilities associated with publicly available models.

To contibute model vulnerabilities, please submit a report on the AI Risk Database website.

This repository contains source code for the AI Risk Database.

Repository overview

Top-level directories - see README.md files within each for more information:

• airdb: Python modules that define AI Risk Database database schemas, repository scanners, and other utilities
• docker: Dockerfiles for the frontend and server
• frontend: Source code for the Next.js website
• lint: Bash scripts and configuration files for Python linting and formatting
• scripts: Python scripts for database management and population
• server: Python API server source code and Caddy reverse proxy configuration file

Standing up a local instance of the AI Risk Database

Docker

To use docker-compose to set up the system locally, ensure both Docker and Docker Compose V1 (i.e. docker-compose) are available.

1. Create the following files at the root of this project:
   • API_KEY.secret: a string representing the API secret that for frontend and server
     • For example, run openssl rand -base64 32 > API_KEY.secret
   • DB_PASSWORD: a string representing the MySQL database password
     • For example, run openssl rand -base64 32 > DB_PASSWORD
2. (Optional) For persisted local data, create the directories data-mysql and data-caddy and edit the docker-compose.yml file to comment them in as service volumes.
3. Run docker-compose up to launch the services. Ctrl+C to stop.

Get in touch

• See CONTRIBUTING.md for getting started with development contributions
• Discuss at #ai-risk-database (Join Slack)