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).”

“According to Floridi [20], digital ethics is defined as a practice-oriented framework that addresses the integration of ethical ideals, such as human dignity and accountability, in the design and implementation of digital technologies, rather than merely emphasizing their technical performance.”

“Regulation is a product of the interaction among institutional, social, and economic dynamics in a very complex way, and government intervention is greatly decisive for the direction and outcome of such an interaction [22].”

“Additionally, theories of information justice and data ethics underscore the importance of fairness, inclusivity, and human rights in the development and use of AI technologies (Floridi, 2019; Taylor, 2017).”

“They have to deal with the structural limitations of ELS programs to go beyond problematic employment relationships that many interpret as a tactical move to prevent a stronger regulation of AI (Floridi 2019; Rességuier and Rodrigues 2020; Whittaker 2021).”

“Es decir, una de las principales limitaciones de la IA es su potencial para perpetuar sesgos existentes, ya que los algoritmos de IA dependen de los datos con los que son entrenados; por lo tanto, si los datos están sesgados, el algoritmo también lo estará (Floridi 2019).”

“…or a rare exercise in visionary science (as in Warren McCulloch’s (1956) blueprint for “some circuitry of ethical robots”) to a practical challenge in moral philosophy and a mainstream engineering concern (e.g., Metzinger, 2013; Dignum, 2018; Floridi, 2019; Jobin et al., 2019; Kuipers, 2020).”

“Concerns related to algorithmic bias, data privacy, and transparency are well documented, particularly in relation to vulnerable populations such as children (Floridi, 2019; Jobin et al., 2019).”

“…how difficult it can be to create workflows in corporate contexts that provide space for meaningful ethical deliberation but also points to a deeper concern about the potential for “ethics bluewashing” (Floridi, 2019; for more on digital ethicswashing more generally, see; Schultz et al., 2025).”

“…argue that such approaches risk reducing ethics to technical checklists, enabling ethics washing rather than fostering genuine accountability (Floridi 2019; Wagner 2019; Hao 2019), while many frameworks assume Western principles are universally applicable (Ghoshal et al. 2025), a stance…”

“From a practical perspective, the McKinsey survey “Tech for good” (2019, p. 2) shows that “the development and adoption of advanced technologies have the poten - tial not only to raise productivity and GDP growth, but also to improve wellbeing more broadly, including through healthier life and…”

“Over the past years, schol - arly literature has stressed the fundamental challenge of translating AI ethics principles into practice [ 35 , 52 , 88].”

“This positioning casts the AI not as a neutral tool, but as a “beneficial moral actor” exercising a digital duty of care [15].”

“Regulators under GDPR and related EU law, as well as U.S. federal/ state consumer protection regimes, raise both factors; deceptive design and opaque processing carry heightened legal exposure and enforcement likelihood [57], [58].”

“As Zhang [51] notes, this risks what Floridi [16] calls “ethics bluewashing”, i.e, creating ethical guidelines to only create a false image of social responsibility.”

“Ethics dumping represents the externalization of the same institutional weakness in transnational research contexts (Floridi 2019).”

“The fi rst approach focuses on direct regulation of judicial power, emphasizing judges ’ ethical norms and review obligations as primary safeguards (Ferreira and Gromova 2023; Floridi 2019).”

“The issue of ‘ethics washing’ emerges, where organisations endorse ethical principles without making tangible changes [58].”

“As a preliminary step toward problem validation, we took part in an exploratory research examining the phenomenon of AI ethics washing [34], broadly defined as the discrepancy between organizations’ stated ethical commitments and their actual practices [35], [36], [37].”

“These incremental intrusions mirror real-world strategies of persuasive design, a concept explored by digital theorists like Zuboff (2019) and Floridi (2019), where technology manipulates choice under the guise of enhancement.”

“In doing so, it aligns itself with recent scholarly debates on digital ethics (Burr et al., 2020; Floridi, 2019; Fuchs, 2023; Keith, 2014), such as the balance between technological innovation and personal freedom (Dorrestijn & Verbeek, 2013; Lempiälä & Vanharanta, 2018).”

“…artificial intelligence contributes processing speed, analytical capacity, and the generation of alternatives (Jiang et al., 2022; Kaplan & Haenlein, 2019; EC HLEG AI-European Commission, 2019), thus establishing a mutually supportive, performance-oriented relationship (Gignac & Szodorai, 2024).”

“As noted by Floridi (2019), this development marks a 'new conceptual era that places information as the foundation of human action' (p. 17).”

“In the contemporary technological landscape, however, the central issue may lie precisely in the emergence of a form of technological “wilderness,” often accompanied by attempts by large corporations to evade stronger forms of regulation by public institutions [86].”

“…sociotécnicas de la IA, recurriendo a autores que analizan la no neutralidad de la tecnología, la deshumanización de los procesos educativos y la desprofesionalización docente (Selwyn, 2019; Luckin, 2018; Dignum, 2019; Feenberg, 1999; Braidotti, 2019; Noble, 2018; Floridi, 2019; Turkle, 2011).”

“Successful human-AI teams benefit from the efficiency of AI systems and the flexibility of humans [1, 30, 101].”

“Even highly autonomous agents cannot be considered responsible actors under current legal principles, meaning that identifiable individuals or organizations remain liable for their use [10], [13], and therefore, allow them to review, approve, or override system outputs when necessary.”

“Within AI governance, such willingness must be designed into institutional and organizational processes, for instance, through procedures that record contestation and require revision rather than optimization only for performance (Floridi 2019; Mittelstadt 2019).”

“Existing legal frameworks still assume humans as the ultimate source of liability [17], [18].”

“While XAI aims to make model decisions intelligible, reflexive AI extends this goal toward epistemic governance: the systematic management of how, why, and under what assumptions knowledge claims are produced by machines [28].”

“The final version of LAIS comprises 14 sections, each aligned with Floridi’s ethical AI principles [4].”

“The worst outcome would be a ‘market for principles’ where stakeholders may be tempted to ‘shop’ for the most appealing ones (Floridi 2019, 185-193).”

“60 61 We obtained ethical approvals for this study from the relevant university ethics committee, yet nevertheless carefully considered whether we should seek consent from content creators for the inclusion of their videos within our research.”

“As Floridi (2019a, 2019b) argues, building trustworthy AI requires translating ethical principles into concrete practices – a process that cannot be achieved overnight.”

“…on Responsible Artificial Intelligence (RAI) has grown rapidly over the last 5 years, with foundational works such as Jobin et al. [45], Floridi [35], and more recently Stahl [74] and Birkstedt et al. [14] articulating the core ethical principles behind RAI—fair-ness, accountability,…”

“XAI evaluation a) The ubiquitous lack of scientific rigor in XAI evaluation while claiming to have the explainability functionality benefits the technical sector as a form of ethics washing (Floridi, 2019).”

“Yan et al. (2023) emphasised the need for open access to large language models in education, aligning with Floridi (2019), who noted that while ethical principles are acknowledged, they often fail due to abstract formulations and insufficient regulation.”

“…substantive demand to improve their AI practices, by engaging instead in what Floridi has called ‘ethics shopping’—namely, ‘the malpractice of choosing, adapting, or revising… ethical principles… from a variety of available offers, in order to retrofit some pre-existing behaviours’ ([15], p. 186).”

“Floridi [27], for example, cautions that companies may engage in “ethics shopping” by choosing from among an abundance of AI guidelines according to which guideline best retroactively justifies the company’s actions, or “bluewashing” through the application of superficial ethical guidelines to…”

“However, this shift raises critical implications: while data-centric models enhance efficiency, they risk reducing complex, context-dependent care scenarios to quantifiable metrics, potentially overlooking relational nuances vital to quality care (Floridi, 2019).”

“However, many authors, including Floridi [8] and Taddeo & Floridi [46], acknowledged the challenges of balancing innovation and ethics in AI development, emphasizing the need for a careful and nuanced approach to integrating digital innovation and ethical AI in smart societies.”

“As principais categorias enquanto campos de investigação são Filosofia e Estudos Religiosos (74 artigos), Ética Aplicada (56 & Floridi, 2021; Milano, Taddeo & Floridi, 2020; Floridi, 2021).”

“Le risque que ces documents soient utilisés non pas comme des vecteurs de réel changement au sein des entreprises, mais plutôt comme stratégie de relation publique ( phénomène que l’on qualifie communément de « ethics washing ») est donc bien réel (Bietti, 2021; Floridi, 2019).”

“The Committee concludes, similarly to Floridi’s study (2019), that this multiplicity of principles and codes confuses the landscape and can undermine attempts to make any set of ethical principles authoritative; and that there is a lack of clarity in how these principles work together (p.”

“Yet critics argue that AI ethics may legitimise existing power asymmetries through ethics washing , where organisations adopt ethical language while resisting binding regulation ([76]; [25]; [5]).”

“In Europe, the recently adopted EU AI Act establishes a risk - based regulatory framework that classifies medical AI applications as “ high - risk, ” requiring stringent standards for safety, transparency, and post - market monitoring 42,43 .”

“This two-way interdependence can be used as a way to avoid effective accountability (Floridi, 2019), but coupled with a Responsible AI Adoption process, motivated by actual applications and aligned with the organizational culture, can be an effective path to actually implement and maintain…”