Abstract
It has become a truism that the ethics of artificial intelligence (AI) is necessary and must help guide technological developments. Numerous ethical guidelines have emerged from academia, industry, government and civil society in recent years. While they provide a basis for discussion on appropriate regulation of AI, it is not always clear how these ethical guidelines were developed, and by whom. Using content analysis, we surveyed a sample of the major documents (n = 47) and analyzed the accessible information regarding their methodology and stakeholder engagement. Surprisingly, only 38% report some form of stakeholder engagement (with 9% involving citizens) and most do not report their methodology for developing normative insights (15%). Our results show that documents with stakeholder engagement develop more comprehensive ethical guidance with greater applicability, and that the private sector is least likely to engage stakeholders. We argue that the current trend for enunciating AI ethical guidance not only poses widely discussed challenges of applicability in practice, but also of transparent development (as it rather behaves as a black box) and of active engagement of diversified, independent and trustworthy stakeholders. While most of these documents consider people and the common good as central to their telos, engagement with the general public is significantly lacking. As AI ethics moves from the initial race for enunciating general principles to more sustainable, inclusive and practical guidance, stakeholder engagement and citizen involvement will need to be embedded into the framing of ethical and societal expectations towards this technology.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Data availability
See supplemental material.
Code availability
Not applicable.
Notes
While BKC indicated sampling 36 documents, two documents are reporting the exact same principles. The G20 Ministerial Statement on Trade and Digital Economy (G20 Trade Ministers and Digital Economy Ministers 2019) adopts the exact same principles as those of the OECD (Recommendation of the Council on Artificial Intelligence 2019). For the purpose of this study, we counted them as only one item and relied solely on the OECD document data.
To ensure consistency in the categorization of our sample, we followed Fjeld and colleagues’ classification of authoring bodies: government, private sector, multi-stakeholder, inter-governmental organization, and civil society.
These scales attempt to structure the degree of normative outputs’ completeness. Our scales differ from simpler assessments of AI ethics principles and guidance, such as Hagendorff’s scale (Hagendorff 2020) for evaluating the technical implementation of ethical goals and values, which is structured around three rather undefined levels: “yes”, “yes, but very few”, and “none”.
It may appear oxymoronic that multi-stakeholder and inter-organizational initiatives do not entail stakeholder engagement. This is because in documents that were classified as not engaging stakeholders, development and drafting of ethical guidelines was done only by internal committees. Therefore, apart from individuals associated with the authoring organizations, no external stakeholders appear to have been involved.
It would be possible to think that the year 2020 could be an outlier because of the health crisis. However, considering that the modalities of stakeholder engagement can be just as much carried out online (Dilhac et al. 2020) and that the year was a banner year in terms of ethical subjects blending normative considerations and AI uses (notably for contact tracing, epidemiological analysis, replacement of human functions deemed essential with automation, etc.), the year could also have been conducive to the enunciation of ethical benchmarks in a host of AI sectors that were affected by the crisis.
References
Agar J (2020) What is science for? The Lighthill report on artificial intelligence reinterpreted. Br J History Sci 53:289–310. https://doi.org/10.1017/S0007087420000230
AI Ethics Lab (2020) Toolbox: dynamics of AI principles. https://aiethicslab.com/big-picture/
Alami H, Lehoux P, Denis J-L et al (2020) Organizational readiness for artificial intelligence in health care: insights for decision-making and practice. J Health Organiz Manag. https://doi.org/10.1108/JHOM-03-2020-0074
Algorithm Watch (2020) AI ethics guidelines global inventory. https://inventory.algorithmwatch.org
Anderson J, Rainie L, Luchsinger A (2018) Artificial intelligence and the future of humans. Pew Research Center, Washington, DC
Arnold DG, Oakley JL (2013) The politics and strategy of industry self-regulation: the pharmaceutical industry’s principles for ethical direct-to-consumer advertising as a deceptive blocking strategy. J Health Polit Policy Law 38:505–544. https://doi.org/10.1215/03616878-2079496
Artificial Intelligence Industry Alliance (2019) Artificial intelligence industry code of conduct. www.secrss.com/articles/11099
Australian Department of Industry, Science, Energy and Resources (2019) AI ethics principles Australia. Government of Australia, Canberra, Australia
Beijing Academy of Artificial Intelligence (2019) Beijing AI Principles. www.baai.ac.cn/blog/beijing-ai-principles?categoryId=394
Bélisle-Pipon J-C (2017) Les enjeux éthiques des communications directes aux patients par les compagnies pharmaceutiques. Université de Montréal
Bélisle-Pipon J-C, Del Grande C, Rouleau G (2019) “Engage patients in your research”, they say. Narrative Inquiry Bioethics 9:13–16. https://doi.org/10.1353/nib.2019.0007
Bélisle-Pipon J-C, Couture V, Roy M-C et al (2021) What makes artificial intelligence exceptional in health technology assessment? Front Artif Intell. https://doi.org/10.3389/frai.2021.736697
Berberich N, Nishida T, Suzuki S (2020) Harmonizing artificial intelligence for social good. Philos Technol 33:613–638. https://doi.org/10.1007/s13347-020-00421-8
Bezuidenhout L, Ratti E (2020) What does it mean to embed ethics in data science? An integrative approach based on the microethics and virtues. AI Soc. https://doi.org/10.1007/s00146-020-01112-w
Boddington P (2017) Towards a code of ethics for artificial intelligence. Springer International Publishing, Cham
Boisvert Y (2008) Leçon d’éthique publique. Santé Publique 20:313. https://doi.org/10.3917/spub.084.0313
Boni M (2021) The ethical dimension of human–artificial intelligence collaboration. Eur View. https://doi.org/10.1177/17816858211059249
Borenstein J, Howard A (2020) Emerging challenges in AI and the need for AI ethics education. AI Ethics. https://doi.org/10.1007/s43681-020-00002-7
Bostrom N, Yudkowsky E (2011) The ethics of artifical intelligence. In: Frankish K, Ramsey WM (eds) The cambridge handbook of artificial intelligence. Cambridge University Press, Cambridge, pp 316–335
Castro D, McLaughlin M (2019) Ten ways the precautionary principle undermines progress in artificial intelligence. Information Technology and Innovation Foundation, Washington, D.C.
Cave S, Dihal K (2020) The whiteness of AI. Philos Technol 33:685–703. https://doi.org/10.1007/s13347-020-00415-6
Cleek MA, Leonard SL (1998) Can corporate codes of ethics influence behavior? J Bus Ethics 17:619–630. https://doi.org/10.1023/A:1017969921581
Clemons EK (2019) New patterns of power and profit: a strategist’s guide to competitive advantage in the age of digital transformation. Palgrave Macmillan, London
Coeckelbergh M (2015) Artificial agents, good care, and modernity. Theor Med Bioeth 36:265–277. https://doi.org/10.1007/s11017-015-9331-y
Crawford K (2021) The atlas of AI. Yale University Press, New Haven
Crawford K, Dobbe R, Dryer T et al (2019) AI now 2019 report. New York University, New York
Crews Jr CW (2020) How the White House “Guidance for Regulation of Artificial Intelligence” Invites Overregulation. Forbes
Daly A, Hagendorff T, Li H et al (2019) Artificial intelligence, governance and ethics: global perspectives. Social Science Research Network, Rochester
de Fine LK, de Fine LJ (2020) Artificial intelligence, transparency, and public decision-making. AI Soc. https://doi.org/10.1007/s00146-020-00960-w
Dilhac M-A, Echaiz LF, Gomez Salazar FG et al (2020) The open dialogue on AI ethics. Algora Lab – Mila, Montréal
Eitel-Porter R (2020) Beyond the promise: implementing ethical AI. AI Ethics. https://doi.org/10.1007/s43681-020-00011-6
Estrada D (2020) Ideal theory in AI ethics. http://arxiv.org/abs/201102279
Etzioni A, Etzioni O (2017) Incorporating ethics into artificial intelligence. J Ethics 21:403–418. https://doi.org/10.1007/s10892-017-9252-2
European Commission’s High-Level Expert Group on Artificial Intelligence (2018) Ethics guidelines for trustworthy AI: high level expert group on artificial intelligence
Fiske A, Tigard D, Müller R et al (2020) Embedded ethics could help implement the pipeline model framework for machine learning healthcare applications. Am J Bioeth 20:32–35. https://doi.org/10.1080/15265161.2020.1820101
Fjeld J, Hilligoss H, Achten N et al (2020) Principled artificial intelligence: a map of ethical and rights-based approaches. Berkman Klein Center for Internet and Society, Cambridge
Floridi L (2019) Translating principles into practices of digital ethics: five risks of being unethical. Philos Technol 32:185–193. https://doi.org/10.1007/s13347-019-00354-x
Floridi L, Cowls J, Beltrametti M et al (2018) AI4People-an ethical framework for a good AI society: opportunities, risks, principles, and recommendations. Minds Mach 28:689–707. https://doi.org/10.1007/s11023-018-9482-5
G20 Trade Ministers and Digital Economy Ministers (2019) G20 Ministerial Statement on Trade and Digital Economy
Gagnon M-A (2020) Ghost management as a central feature of accumulation in corporate capitalism: the case of the global pharmaceutical sector. In: Benquet M, Bourgeron T (eds) Accumulating capital today: contemporary strategies of profit and dispossessive policies, 1st edn. Routledge, pp 163–177
Geis JR, Brady AP, Wu CC et al (2019) Ethics of artificial intelligence in radiology: summary of the joint European and North American multisociety statement. Radiology 293:436–440. https://doi.org/10.1148/radiol.2019191586
Goldsmith J, Burton E (2017) Why Teaching Ethics to AI Practitioners is Important. Proc AAAI Conf Artif Intell. https://doi.org/10.5555/3297863.3297908
Gordon K, Miyake M (1999) Deciphering codes of corporate conduct: a review of their contents. Organisation for Economic Co-operation and Development, Paris
Government of Canada (2020) Responsible use of artificial intelligence. Canada, Ottawa
Greene D, Hoffmann AL, Stark L (2019) Better, nicer, clearer, fairer: a critical assessment of the movement for ethical artificial intelligence and machine learning. In: Proceedings of the 52nd Hawaii International Conference on System Sciences
Hagendorff T (2020) The ethics of AI ethics: an evaluation of guidelines. Mind Mach. https://doi.org/10.1007/s11023-020-09517-8
Hagerty A, Rubinov I (2019) Global AI ethics: a review of the social impacts and ethical implications of artificial intelligence. http://arxiv.org/abs/190707892
Häußler H (2021) The underlying values of data ethics frameworks: a critical analysis of discourses and power structures. Libri 71:307–319. https://doi.org/10.1515/libri-2021-0095
Hickok M (2020) Lessons learned from AI ethics principles for future actions. AI Ethics. https://doi.org/10.1007/s43681-020-00008-1
Jobin A, Ienca M, Vayena E (2019) The global landscape of AI ethics guidelines. Nat Mach Intell 1:389–399. https://doi.org/10.1038/s42256-019-0088-2
Keller P, Drake A (2020) Exclusivity and paternalism in the public governance of explainable AI. Comput Law Secur Rev. https://doi.org/10.1016/j.clsr.2020.105490
Lacroix A, Marchildon A (2013) Quelle éthique pour la finance? Portrait et analyse de la finance socialement responsable. Presses de l’Université du Québec, Québec
LaGrandeur K (2020) How safe is our reliance on AI, and should we regulate it? AI Ethics. https://doi.org/10.1007/s43681-020-00010-7
Latham A, Goltz S (2019) A survey of the general public’s views on the ethics of using AI in education. In: Isotani S, Millán E, Ogan A et al (eds) Artificial intelligence in education. Springer International Publishing, Cham, pp 194–206
Lauer D (2020) You cannot have AI ethics without ethics. AI Ethics. https://doi.org/10.1007/s43681-020-00013-4
Metzinger T (2019) Ethics washing made in Europe. Der Tagesspiegel
Mittelstadt B (2019a) Principles alone cannot guarantee ethical AI. Nat Mach Intell 1:501–507. https://doi.org/10.1038/s42256-019-0114-4
Mittelstadt B (2019b) AI ethics–too principled to fail? SSRN J. https://doi.org/10.2139/ssrn.3391293
Morley J, Floridi L, Kinsey L, Elhalal A (2020) From what to how: an initial review of publicly available AI ethics tools, methods and research to translate principles into practices. Sci Eng Ethics 26:2141–2168. https://doi.org/10.1007/s11948-019-00165-5
Norman W (2011) Business ethics as self-regulation: why principles that ground regulations should be used to ground beyond-compliance norms as well. J Bus Ethics 102:43–57. https://doi.org/10.1007/s10551-011-1193-2
ÓhÉigeartaigh SS, Whittlestone J, Liu Y et al (2020) Overcoming barriers to cross-cultural cooperation in AI ethics and governance. Philos Technol 33:571–593. https://doi.org/10.1007/s13347-020-00402-x
Organisation for Economic Co-operation and Development (2019) Recommendation of the council on artificial intelligence
Partnership on AI (2016) Tenets. https://www.partnershiponai.org/tenets/
Pasquale F (2015) The black box society. Harvard University Press, Cambridge
Rees C (2020) The ethics of artificial intelligence. In: Strous L, Johnson R, Grier DA, Swade D (eds) Unimagined futures–ICT opportunities and challenges. Springer International Publishing, Cham, pp 55–69
Ryan M, Stahl BC (2020) Artificial intelligence ethics guidelines for developers and users: clarifying their content and normative implications. J Inf Commun Ethics Soc. https://doi.org/10.1108/JICES-12-2019-0138
Samuel G, Derrick G (2020) Defining ethical standards for the application of digital tools to population health research. Bull World Health Organ 98:239–244. https://doi.org/10.2471/BLT.19.237370
Select Committee on Artificial Intelligence (2018) AI in the UK: ready, willing and able. UK House of Lords, London
Sloane M (2018) Making artificial intelligence socially just: why the current focus on ethics is not enough. In: British Politics and Policy at LSE. https://blogs.lse.ac.uk/politicsandpolicy/artificial-intelligence-and-society-ethics/. Accessed 16 Dec 2020
Sloane M, Moss E, Awomolo O, Forlano L (2020) Participation is not a design fix for machine learning. http://arxiv.org/abs/200702423
Stahl BC (2021) Artificial intelligence for a better future: an ecosystem perspective on the ethics of AI and emerging digital technologies. Springer International Publishing, Cham
Taddeo M, Floridi L (2018) How AI can be a force for good. Science 361:751–752. https://doi.org/10.1126/science.aat5991
Tencent Institute (2017) Six Principles of AI. http://www.kejilie.com/iyiou/article/ZRZFn2.html
Université de Montréal (2018) Montreal declaration for a responsible development of artificial intelligence. https://www.montrealdeclaration-responsibleai.com/thedeclaration
Webster G (2019) Translation: Chinese AI Alliance Drafts Self-Discipline “Joint Pledge.” In: New America. https://www.newamerica.org/cybersecurity-initiative/digichina/blog/translation-chinese-ai-alliance-drafts-self-discipline-joint-pledge/
Whittlestone J, Nyrup R, Alexandrova A, Cave S (2019) The role and limits of principles in ai ethics: towards a focus on tensions. https://doi.org/10.1145/3306618.3314289
Wilson C (2021) Public engagement and AI: a values analysis of national strategies. Gov Inf Q. https://doi.org/10.1016/j.giq.2021.101652
Zeng Y, Lu E, Huangfu C (2018) Linking artificial intelligence principles. arXiv. https://arxiv.org/abs/1812.04814v1
Funding
This project has been funded by a Partnership Engage Grant from the Social Sciences and Humanities Research Council (SSHRC) of Canada.
Author information
Authors and Affiliations
Contributions
JCBP designed the study and wrote the first draft. EM and JCBP read all the documents and extracted all pertinent information for content analysis. MCR and VC helped to mitigate the divergent interpretation. MCR, VC and EM critically revised the paper.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they do not have any conflict regarding this project.
Ethical approval
This study does not involve human participants.
Consent for publication
This study has not been submitted elsewhere and the author consents to publication in the journal.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Bélisle-Pipon, JC., Monteferrante, E., Roy, MC. et al. Artificial intelligence ethics has a black box problem. AI & Soc 38, 1507–1522 (2023). https://doi.org/10.1007/s00146-021-01380-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00146-021-01380-0