Translating Principles into Practices of Digital Ethics: Five Risks of Being Unethical

“In environmental ethics, greenwashing (Delmas and Burbano 2011) is the malpractice of a private or public actor seeking to appear greener, more sustainable, or ecologically friendlier than it actually is.”

“In the ethics of AI, see for example (Wiener1960; Samuel 1960).”

“Floridi (2019) identifies risks in the transition from what to how that include: digital ethics shopping, “bluewashing” (i.e., digital greenwashing), lobbying, ethics dumping (outsourcing to other actors), and shirking.”

“If not, corporate AI governance may simply persuade managers to outsource unethical or risky projects (Floridi, 2019).”

“Notable contributions to this field include the works of researchers and research teams such as Boddington (2017), Floridi (2019), Jobin et al. (2019), Whittaker et al. (2018) and Winfield and Jirotka (2018).”

“Indeed, the translation of “lofty” principles (Mittelstadt 2019) into robust and opera-tionalisable guidelines that can enable truly informed decisions on the ground has been identified as a key challenge for digital ethics (Floridi 2019; Blackman 2020; Morley et al. 2021a).”

“In some cases, this massive proliferation of ethics in regulation of Information and Communication Technology (ICT) and Artificial Intelligence have even given rise to serious doubts concerning the “ethification” [31] and “ethics bluewashing” [32] to name only two of them.”

“…may allow the successful and persuasive conduct of actual malpractices like choosing ethical principles that retrofit pre-existing behaviors, making misleading or unsubstantial claims about ethical measures, exploiting ethics to delay or avoid necessary regulation, etc. [62] in the first place.”

“Yet, despite international attempts to provide ethical guidelines for the development of responsible or trustworthy AI and develop a suitable regulatory framework, there remains large uncertainty whether and how such principles translate into practice [14].”

“The phenomenon of connectivity presents unique digital risks which change and grow in complexity, as the technology changes the forms of connectivity (Zuboff 1988, 1996; Van Loon 2003; Turkle 2006, 2011; Lupton 2016; Floridi 2019).”

“Following Floridi [36], we consider that to avoid AIEF, we need to establish clear, shared, and publicly accepted ethical standards.”

“86 AI companies affirm their adherence to the ethical values they claim to adhere to in developing their own products, 84 but they do not provide sufficient data to be able to analyze these products ethically as they do with other products, such as medicines, which leads to sustained debate and…”

“According to Floridi (2019), establishing clear and publicly accepted ethical guidelines might help private organisations subscribe to those ethical guidelines instead of cooking their ethics.”

“Digital ethics frameworks, such as those proposed by Floridi (2019) and Wachter (2018), emphasize the need for responsible and accountable creation and dissemination of digital content, balancing freedom of expression with ethical considerations.”

“Floridi (2019) estimated that, circa 2019, there were over 70 such frameworks being advanced by various groups, each with their own areas of interest, aims, specializations, agendas, and ways of specifying their principles.”

“Notably, concerns have been raised regarding the quality of these principles, revealing that they tend to be abstract, lack theoretical ground, and fail to provide concrete guidance for practice (Floridi, 2019; Hagendorff, 2020).”

“Education of both the general public as well as other stakeholders such as executive board members can help guard against this [37], as can strengthening the role of the commentariat and academia.”

“In Brazil, there have been similar trends toward creating a new legislation for AI.”

“Attard-Frost et al. (2023); Bietti (2020); Connelly et al. (2011); Floridi (2019); Seele and Schultz (2022); Thomas and Lamm (2012); Legitimacy Theory: Companies gain immediate credibility by aligning themselves with cutting-edge technology trends.”

“Some works argued that a narrower focus should be put on elaborating strategies for clinicians, developers, and patients to effectively translate AI ethical principles into practice [15, 22, 25].”

“…research being more ethical than it really is), ethical shirking (i.e., lowering the commitments to ethical values while seeking the same returns), and ethics lobbying by conflicted ethics groups that might attempt to legitimize their sets of ethics through political or other means (Floridi 2019).”

“The conceptual models provided by Claretha Hughes (2010, 2019) address the ethical issues involved to connect human resource development with technology development.”

“This, in turn, may reduce AI ethics efforts to a meaning-less façade, voided of any concrete outcomes and induce malpractices, like ethics bluewashing 2 (Floridi 2019, 185).”

“…and audits), and for accountability or responsibility requirements (see, e.g., [1–4, 9, 12, 14, 22, 23]) serves to respond to the concern of so-called “ethics-washing” or “bluewashing” [8], in which companies misuse ethical guidelines to make it seem as if they are ethical when they are not.”

“Fortunately, some authors [6, 8] noted that there is a substantial overlap between the different sets of principles proposed, and suggested to condense them in a set of 4, 5 principles that outline the fundamental traits of an AI application.”

“, Ethical and Societal Implications of Algorithms, Data, and Artificial Intelligence: A Roadmap for Research (London: Nuffield Foundation, 2019); Luciano Floridi et al., “AI4people-an Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations,” Minds and machines 28, no. 4 (2018). 1…”

“Private companies may choose to implement policies governing their AI systems for a variety of reasons, including adapting to consumer opinion or preempting stricter AI controls by the government [7, 21].”

“The benchmark for how AI should be developed ethi-cally, set by guidelines [21] and approaches drawing on an entity perspective on values [7], may overlook how valuing is reconfigured in interplay with practice.”

“Even if any sort of triage implies at least a modest scarcity of healthcare resources, in these cases the medical approach of triage changes and the focus “ can no longer be on each individual but must shift to the population as a whole ” (Frykberg, 2005).”

“Similarly, the lack of robust metrics for evaluating the ethical implications of XAI systems, as highlighted by Floridi’s discussion of “ethics bluewashing," further compounds the risk of superficial ethical engagement (Floridi, 2019).”

“…standards [...] from a variety of available offers, in order to retrofit some pre - existing behaviours (choices, processes, strategies, etc.), and hence justify them a posteriori, instead of implementing or improving new behaviours by benchmarking them against public, ethical standards’ [24:186].”

“It has been maintained, for example, that principles may come into conflict with each other, that these lists of principles lack theoretical unity and that they may reinforce a questionable box-ticking attitude towards ethical challenges [11 – 13].”

“However, the resulting fragmented and largely corporate-based mosaic of ethical frameworks, codes, and guidelines, many of which suffer from deficiencies such a lack of scientific rigor, subjectiveness, incoherence, superficiality, and redundance, creates confusion [15], [16].”

“This abundance complicates framework selection for specific use cases and increases risks of ethics washing, shopping, shirking, dumping, or lobbying (Floridi, 2019).”

“Therefore, AI ethics is undergoing a practical turn , trying to translate abstract values and principles into concrete design practices (Floridi, 2019; Hagendorff, 2022).”

“Potential risks also include ethics bashing, where stakeholders intentionally undermine or criticize ethical principles to advance their interests (Bietti, 2020), or ethics shopping (Floridi, 2019), where stakeholders selectively adopt and promote ethical principles that align with their interests.”

“…technology are notions that have spawned a sustained philosophical debate (cf. Andras et al. 2018; Durán and Formanek 2018; Durán and Jongsma 2021; Floridi 2019a, b; Laux et al. 2023; Nickel et al. 2010; Rieder et al. 2020; Ruokonen 2013; Ryan 2020) The common rationale adopted by some appears to…”

“While Floridi expressed skepticism about doomsday scenarios associated with superintelligence (Floridi, 2019), influential figures such as Elon Musk and late Stephen Hawking took the idea seriously.”

“To date, the debate on new technologies is primarily formulated in terms of ethical guidelines rather than legal dispositions [153].”

“” [50, p. 189].”

“Roughly speaking, ethics shopping can be specified as the practice of choosing certain ethical principles from a variety of such principles in order to justify one’s behaviour a posteriori, ‘instead of implementing or improving new behaviours by benchmarking them against public, ethical standards [e.g. ethical guideli…”

“With the growing presence of autonomous software and devices in daily life, the ethics of Artificial Intelligence (AI) has rightly become an intensely researched and debated topic (e.g., McCulloch, 1956; Metzinger, 2013b; Dignum, 2018; Kuipers, 2019; Floridi, 2019; Jobin et al., 2019).”

“Language shapes the social and political realities of AI adoption, influencing public trust, regulatory frameworks, and the ethical norms surrounding its development and deployment (Floridi 2019, 2024).”

“Floridi (2019) consolidates these failures into five risks of being unethical: ethics shopping; ethics washing; ethics lobbying; ethics dumping; and ethics shirking.”

“Second, Floridi [14:188] argues that the risk of lack of explicability is unique to AI systems.”

“Therefore, it is crucial to skillfully navigate ethical considerations in order to guarantee a respectful and effective digital culinary promotion (Dwifedi et al., 2020; Floridi, et al, 2019).”

“…ethical principles can be undermined by unethical practices like ‘ethics blue washing’, i.e., making unsub-stantiated claims about AI systems to appear more ethical than they are, or ‘eth-ics lobbying’, i.e., exploiting ethics to delay or avoid necessary legislation (Ferretti, 2021; Floridi, 2019).”

“This notion of translation from principles to practice has stuck; some refer to it as ‘operationalizing AI ethics’ [27, 37] or the ‘what’ (principles) and ‘how’ (practice) of AI ethics [18].”

“First, these tools are highly heterogeneous, which may facilitate “mixing and matching” [18] principles that do not rely on a solidly defined conceptual framework.”

“…years, scholars in AI ethics have been concerned with the question of how relevant institutions and stakeholders should translate the theory underlying AI ethics principles, guidelines, and codes into practice in the form of implementable technical and legal mechanisms [4, 14–16, 33, 47–49, 79].”

“Without proper governance, audit-washing risks hindering progress towards rigorous algorithmic evaluations [23, 56, 63, 75].”

“This accumulation of guidelines may also cause what Floridi called ’ethics shopping’ [34]: picking the guidelines or principles that would ’retrofit some pre-existing behaviours’, thus avoiding having to make changes that would be too inconvenient.”

“As disclosed in [14] it is urgent to ground the practice of AI engineering in ethical principles.”

“There are similarities with two types of digital ethics malpractice identified by Luciano Floridi [16].”

“38”

“The AI HLEG (2019) noted further notes that AI systems can operate based on symbolic rules, learn numeric models , and adapt their behaviour to environmental feedback on their previous actions.”

“This research aligns with and extends previous work on ethical decision-making in organizations (Trevi ~ no et al., 2006) and the ethical issues surrounding the use of AI in various contexts (Floridi, 2019a; Mittelstadt et al., 2016; Whittlestone et al., 2019).”

“This evolvement illustrates an unresolved tension between the aimof unification (Floridi 2019) and the acknowledgment of complex and singular situations which require specific principles (van den Hoven 2010; Vayena and Tasioulas 2016).”

“…and principles from companies as forms of ‘ ethics washing ’ , ‘ ethics shopping ’ , or ‘ ethics shirking ’ that de fl ect regulation while preserving reputations (Floridi, 2019; Greene et al., 2019; Wagner, 2018), others view them as private ordering with public impact, akin to corporate…”

“…by illiberal, authoritarian regimes against so-called Western democratic systems and rights-based data governance, and the romanticising of the category of the South, and argue instead for a critical approach to the biases, incongruences and contradictory aspects of Southern knowledge (2019: 326).”

“…taken actions - ethics bluewashing: tokenistic actions, ethics as a PR campaign - ethics lobbying: ethics as alternative to legislation - ethics dumping: exporting unethical practices to countries with laxer laws - ethics shrinking: doing the minimum when more is not demanded (Floridi, 2019).”

“This fog of meaning facilitates finding a suitable definition for these terms that readily conforms to existing practices and covers up potential moral harms [5, 25].”

“Questions have been raised as to whether existing ethical frameworks are sufficiently robust in light of the evolving innovations including research utilizing increasing applications of Artificial Intelligence (AI) (Floridi 2019).”