Passing the Test: Breaking Down the When, Where and How of Testing AI Systems

AI testing has emerged as a best practice (and requirement, in some jurisdictions) but key questions remain about how to implement testing in an AI governance program, such as who is responsible for testing, when you should test, how you should test, which safeguards you should implement to protect confidential information while testing, and how to assess and/or disclose test results.

What you will learn:

• How to design testing metrics that assess the most significant issues affecting performance and vulnerability.
• Identifying safeguards to implement when conducting testing to protect production environments and business and customer data.
• Identifying resources practitioners can use to enable development of effecting testing of AI models and accompanying governance measures.