Mapping of AI Adversarial Threats to Robustness Tools

[divyasd]

Description:
We invite contributors to help map AI adversarial threats to existing robustness testing tools. This will help practitioners select appropriate tools for evaluating system resilience.

Problem Overview
Numerous tools exist for testing AI robustness, but they are not clearly linked to specific adversarial threats or GuardRail questions. This limits their strategic use in threat mitigation.

Contribution Scope

• Curate a list of AI adversarial robustness tools.
• Map tools to relevant threats and GuardRail questions.
• Suggest tool enhancements or new mappings.

Impact

• Enables more targeted use of robustness tools.
• Improves threat coverage and system resilience.