To familiarize yourself with Mindgard we recommend testing a demo AI model first.

Click Run New Test on the Model Tests page to get started. You’ll see an option to Run test on demo target—this allows you to execute the full suite of attacks against a model hosted by us. The specific model currently in use will be clearly indicated next to the button.

You’ll now be redirected to the test results page where you can see the risks identified from any tests you have run. Your test may take a few minutes to complete, during which time the flagged events column will be blank.

Click into the test you have just requested to see its results when it completes.

Results will appear as soon as they are available. You will see results from a list of different attack techniques that have been run against the AI model you selected, as well as the number of events flagged alongside a summary of the test.

Click into one of the attack techniques to see more details. The next page shows you:

  • Summary: The percentage of attack attempts against the AI model deemed successful for the specific attack technique.
  • Framework Mapping: The attack mapped to industry-standard frameworks such as OWASP Top 25 and MITRE ATLAS.
  • Target Details: The target system, alongside attack statistics.
  • Remediation: Recommendations to reduce the system’s susceptibility to this attack technique.
  • Event Details: The details of the inputs and outputs observed during the test.

This is the standard workflow for running tests and viewing their results moving forward.