🔥🔥 Don’t Miss Out on Our End of Autumn Sale. If you have any questions, feel free to click the bottom right corner to contact our customer service..

Concat us[email protected]
Shopping Cart

No products in the cart.

🔍
BSI PD ISO/IEC TR 24028:2020 2022

BSI PD ISO/IEC TR 24028:2020 2022

$198.66

Information technology. Artificial intelligence. Overview of trustworthiness in artificial intelligence

Published By Publication Date Number of Pages
BSI 2022 52
PDF Format Searchable Printable Preview Now
Guaranteed Safe Checkout
Category:

If you have any questions, feel free to reach out to our online customer service team by clicking on the bottom right corner. We’re here to assist you 24/7.
Email:[email protected]

This document surveys topics related to trustworthiness in AI systems, including the following:

  1. approaches to establish trust in AI systems through transparency, explainability, controllability, etc.;

  2. engineering pitfalls and typical associated threats and risks to AI systems, along with possible mitigation techniques and methods; and

  3. approaches to assess and achieve availability, resiliency, reliability, accuracy, safety, security and privacy of AI systems.

The specification of levels of trustworthiness for AI systems is out of the scope of this document.

PDF Catalog

PDF Pages PDF Title
2 National foreword
7 Foreword
8 Introduction
9 1 Scope
2 Normative references
3 Terms and definitions
15 4 Overview
5 Existing frameworks applicable to trustworthiness
5.1 Background
16 5.2 Recognition of layers of trust
5.3 Application of software and data quality standards
18 5.4 Application of risk management
5.5 Hardware-assisted approaches
19 6 Stakeholders
6.1 General concepts
20 6.2 Types
6.3 Assets
21 6.4 Values
7 Recognition of high-level concerns
7.1 Responsibility, accountability and governance
22 7.2 Safety
8 Vulnerabilities, threats and challenges
8.1 General
23 8.2 AI specific security threats
8.2.1 General
8.2.2 Data poisoning
8.2.3 Adversarial attacks
24 8.2.4 Model stealing
8.2.5 Hardware-focused threats to confidentiality and integrity
8.3 AI specific privacy threats
8.3.1 General
8.3.2 Data acquisition
25 8.3.3 Data pre-processing and modelling
8.3.4 Model query
8.4 Bias
8.5 Unpredictability
26 8.6 Opaqueness
8.7 Challenges related to the specification of AI systems
27 8.8 Challenges related to the implementation of AI systems
8.8.1 Data acquisition and preparation
8.8.2 Modelling
29 8.8.3 Model updates
8.8.4 Software defects
8.9 Challenges related to the use of AI systems
8.9.1 Human-computer interaction (HCI) factors
30 8.9.2 Misapplication of AI systems that demonstrate realistic human behaviour
8.10 System hardware faults
31 9 Mitigation measures
9.1 General
9.2 Transparency
32 9.3 Explainability
9.3.1 General
9.3.2 Aims of explanation
9.3.3 Ex-ante vs ex-post explanation
33 9.3.4 Approaches to explainability
9.3.5 Modes of ex-post explanation
34 9.3.6 Levels of explainability
35 9.3.7 Evaluation of the explanations
9.4 Controllability
9.4.1 General
36 9.4.2 Human-in-the-loop control points
9.5 Strategies for reducing bias
9.6 Privacy
9.7 Reliability, resilience and robustness
37 9.8 Mitigating system hardware faults
9.9 Functional safety
38 9.10 Testing and evaluation
9.10.1 General
9.10.2 Software validation and verification methods
40 9.10.3 Robustness considerations
41 9.10.4 Privacy-related considerations
9.10.5 System predictability considerations
42 9.11 Use and applicability
9.11.1 Compliance
9.11.2 Managing expectations
9.11.3 Product labelling
9.11.4 Cognitive science research
10 Conclusions
44 Annex A (informative) Related work on societal issues
45 Bibliography

Related products

BSI PD ISO/IEC TR 24028:2020 2022
$198.66