Assessing AI's risks and impacts: A conversation with NIST's Reva Schwartz
The Privacy Advisor Podcast
00:00
Evaluating Trust in AI
This chapter explores a structured initiative by a government agency to evaluate the societal impacts of artificial intelligence through realistic testing scenarios. It focuses on a three-level evaluation program that includes model testing, risk examination, and real-world field testing, aiming to ensure AI applications are trustworthy and responsible. The discussion highlights the complexities of AI evaluations, the importance of risk management, and the potential biases involved in AI models relying on synthetic data.
Transcript
Play full episode
