IT-tillämpningar: allmänt

This document provides a high-level framework and guidance for the development and operation of conformity assessment schemes, including certification schemes, for artificial intelligence (AI) systems.

This standard proposal outlines artificial intelligence (AI) concepts, applications, and implications, as well as benefits, risks, governance framework, and scope for use by organizations within the context of project, programme, and portfolio management (P3M). This proposed standard is intended to supplement the other TC 258 standards and is in direct alignment with standards prepared by ISO/TC 258. This document provides guidance for organizations on what to consider when adopting or looking to improve on existing AI technologies that directly support P3M. This proposed standard is applicable to any organization, including public, private, and charitable, as well as to any type of project, regardless of purpose, delivery approaches, project or programme lifecycle model used, complexity, size, cost, or duration. This document also provides (P3M) practitioners (including top-level management, project managers, PMO’s, etc.) with AI-related guidance and consideration in the areas of security, privacy, transparency, and bias, as well as related data governance, ethical and risk considerations.

This document specifies the evaluation of natural language processing systems, in the sense of measuring the quality of a system’s results to assess its functional suitability. It provides a definition of evaluation methods for those systems, together with guidance on how to select, implement and interpret those evaluation methods. This document covers quantitative metrics as well as other evaluation methods. It includes requirements on the implementation of the described metrics, and further requirements on the technical resources involved in the evaluation process.

This document defines concepts, measures and requirements for assessment and treatment of bias in AI systems. This includes bias unwanted by the AI Provider and AI Deployer according to their specification of the AI system, in the context of the AI Act. This encompasses consideration of data bias including any data used to build or assess the AI system, but also system or model bias that can result from algorithmic factors, such as algorithm design choices.