AI Risk Management Framework: Second Draft On the Importance of Simpson’s Paradox and Systems Theory in AI
AI Risk Management Framework: Second Draft On the Importance of Simpson’s Paradox and Systems Theory in AI
4 PagesAI Risk Management Framework: Second Draft On the Importance of Simpson’s Paradox and Systems Theory in AI Neo4j Inc Response By Kara Doriani O’Shee To our fellow citizens, leaders, and whom it may concern, The National Institute of Standards and Technology (NIST) has requested comments on its second draft Artificial Intelligence Risk Management Framework (AI RMF). Neo4j welcomes the opportunity to support the agency in these efforts. We feel that this draft version of the AI RMF is on the right track, especially NIST’s work in defining the key characteristics of trustworthy AI. To further advance this draft, we offer three recommendations: ? Include guidance on the importance of measuring accuracy for subgroups of data, in addition to aggregate measures ? Use a systems approach to AI safe
