Common Vulnerabilities and Exposures for AI

Common Vulnerabilities and Exposures for AI

Reports analyzing open AI models using XAI and adversarial model analysis techniques.

How to add a model validation report?

You can simply submit your own report with AI model vulnerabilities.

This can be done in three steps.

  1. Make a fork of the repository https://github.com/ModelOriented/CVE-AI
  2. In the _publications directory you will find markdown files with report descriptions and in the files directory you will find the reports (currently they are mainly pdf files). Add your report here
  3. Submit the report via pull-request to the ModelOriented/CVE-AI repository

Make sure the report has explicit information about (1) what model is being analyzed, (2) what adversarial analysis techniques are being used, (3) and what the results of that analysis are.