A Taxonomy of Trustworthiness for Artificial Intelligence
Summary
Improving the trustworthiness of artificial intelligence (AI) systems is a shared priority for the private and public sectors, as indicated by prolific research and guidelines in recent years. However, the meaning of trustworthy AI, and potential approaches to attain it, remain contested. Stakeholders often lack a shared vocabulary or set of questions to consider, and guidance that speaks both to AI developers and policymakers is rare. Existing frameworks for trustworthy AI have additionally tended to focus on a relatively narrow set of AI models and applications that directly interact with people.
Since 2021, the National Institute of Standards and Technology (NIST), in collaboration with diverse stakeholders, has been developing an AI Risk Management Framework (RMF) intended to promote trustworthy AI. In this paper, we analyze the landscape of trustworthy AI and introduce a taxonomy of trustworthiness for artificial intelligence that is intended to complement and support the use of the NIST AI RMF.
The taxonomy introduced in this paper includes 150 properties of trustworthiness for AI. Each property builds upon a relevant “characteristic of trustworthiness” as defined by NIST in the AI RMF. NIST’s characteristics of trustworthiness include: valid and reliable, safe, secure and resilient, accountable and transparent, explainable and interpretable, privacy-enhanced, and fair
with harmful bias managed.