Towards a Code of Ethics for Artificial Intelligence

Abstract

the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The global landscape of AI ethics guidelines

“…Fears that AI might jeopardize jobs for human workers 7 , be misused by malevolent actors 8 , elude accountability or inadvertently disseminate bias and thereby undermine fairness 9 have been at the forefront of the recent scientific literature and media coverage. Several studies have discussed the topic of ethical AI [10][11][12][13] , notably in metaassessments [14][15][16] or in relation to systemic risks 17,18 and unintended negative consequences like algorithmic bias or discrimination [19][20][21] .…”

Economic patterns in a world with artificial intelligence

“…(Naughton 2015). Accountable algorithms have become a dedicated field of interest in AI (Kroll et al 2016;Boddington 2017). This and the sheer fact that individual AI agents have no "skin in the game (Taleb 2018) suggest that normative individualism needs to be critically reviewed as a foundation for economic and social institutions.…”

AI Ethics – Too Principled to Fail?

“…22 Formal professions are defined by trust between clients and practitioners, mediated by common goals and values within the profession, and enforced through sanctions and self-governance. 17,23,24 The fiduciary duties derived from the client-practitioner relationship separate 'professions' from other vocations, 25 and facilitate a principled approach to ethical decision-making by requiring practitioners to promote their clients' best interests. AI development is not a formal profession.…”

“…Likewise, AI development includes practitioners from varied disciplines and professions such as software engineering and data science, which have incongruous histories, cultures, incentive structures, and ethics. 14,23,24 Software engineering, which is the closest analogue, has historically not been legally recognised as a profession with fiduciary duties to the public 14,32 due to a perceived lack of licensure schemes and a well-defined 'standard of care' for the profession. 33 Reflecting this, a comparably rich account of what it means to be a 'good' AI developer or software engineer does not exist; while the IEEE and ACM, two of the field's largest professional associations, have published codes of ethics, these documents are comparatively short and relatively lacking in grounded advice and specific behavioural norms.…”

“…Self-regulatory frameworks must be embedded and highly visible in organisational culture to be effective. 24 Binding accountability structures as well as clear implementation and review processes are needed at a sectoral and organisational level. While select AI companies have created (and disbanded) ethics committees, 68 their remit, independence, and decision-making power is rarely clear.…”