Nist Solicits Comments on Four Principles of Explainable Artificial Intelligence and Other Developments

• Jurisdiction: United States,Federal
• Publication year: 2021
• Citation: Vol. 4 No. 1

Lee J. Tiedrich, Sam Jungyun Choi, and James Yoon^*

The authors of this article discuss a National Institute of Standards and Technology white paper that seeks to define the principles that capture the fundamental properties of explainable artificial intelligence systems and related legislative developments.

The National Institute of Standards and Technology ("NIST") released the first draft of the Four Principles of Explainable Artificial Intelligence,¹ a white paper that seeks to define the principles that capture the fundamental properties of explainable artificial intelligence ("AI") systems. NIST accepted comments through October 15, 2020.

The White Paper

In February 2019, the Executive Order on Maintaining American Leadership in Artificial Intelligence² directed NIST to develop a plan that would, among other objectives, "ensure that technical standards minimize vulnerability to attacks from malicious actors and reflect Federal priorities for innovation, public trust, and public confidence in systems that use AI technologies; and develop international standards to promote and protect those priorities."

In response, NIST issued a plan³ in August 2019 for prioritizing federal agency engagement in the development of AI standards, identifying seven properties that characterize trustworthy AI—accuracy, explainability, resiliency, safety, reliability, objectivity, and security.

NIST's white paper focuses on explainability and identifies four principles underlying explainable AI.

- Explanation. AI systems must supply evidence, support, or reasoning for their outputs. Researchers have developed different models to explain AI systems, such as self-explainable models where the models themselves are the provided explanation.
- Meaningful. The recipient must understand the AI system's explanation. This principle is a contextual requirement—for example, different types of user groups may require different explanations, or a particular user's prior knowledge, experiences, and mental processes may affect meaningfulness. Hence, tailoring is necessary for effective communication.
- Explanation Accuracy. The explanation must correctly reflect the AI system's process for generating its output. In contrast to decision accuracy, explanation accuracy is not concerned with whether or not the system's judgment is correct. It is referencing how the system came to its conclusion. The principle is also contextual—there may be different explanation accuracy metrics for

...