A/78/310 as well as the significance and the envisaged consequences of such processing for the data subject. 47 55. The National Institute of Standards and Technology summarizes the scope of this principle in the following chart: 48 56. The following table explains the most relevant aspects of each principle according to the National Institute of Standards and Technology document: 49 Principle Meaning or scope Explanation Evidence, support or reasoning related to outputs and/or processes of an [artificial intelligence] system. Meaningful Explanation in terms that are understandable to the intended consumer. In other words, this principle seeks to make the explanation comp rehensible for a given audience. Many factors affect a good explanation, which is why the target audience or audience to which the explanation is addressed must be taken into account. Explanation Accuracy This principle requires the technical explanation to be rigorous, accurate and comprehensive. Knowledge Limits Identifying and declaring the limits of knowledge implies making it clear that the system is neither perfect nor infallible because the [artificial intelligence] operates within certain limits and constraints within which it has been programmed. They also depend on the quality and quantity of information processed, among other factors. 57. It has been argued that the explanation must: (a) “be understandable and convincing to the user”; (b) “accurately reflect the system’s reasoning”; (c) “be comprehensive”, and (d) “be specific in the sense that different users with different __________________ 47 48 49 23-15851 Ibid., art. 15 (1). National Institute of Standards and Technology, Four Principles of Explainable Artificial Intelligence, - NISTIR 8312 (2021), p. 3. Available at https://doi.org/10.1017/bhj.2023.10. The explanation in the table is an adaptation and summary of the original English text cited and available at https://doi.org/10.6028/NIST.IR.8312. 17/20