The ethical axiotropism between moral development and amoral optimization
Vicentiu Buzduga
Doctoral School of Philosophy and Social-Political Sciences, Alexandru Ioan Cuza University of Iași, Romania
AGATHOS, Volume 16, Issue 1 (30): 71-78, DOI 10.5281/zenodo.15279055
© www.agathos-international-review.com CC BY NC 2025
Abstract: This article aims (using the philosophical method) to raise fundamental questions regarding the possible directions of ethics, particularly the survival of authentic moral values in a world dominated by amoral optimization—an algorithmic process that prioritizes efficiency over ethical considerations. One of its most pervasive effects is digitcultism, a phenomenon characterized by excessive belief in and irrational dependence on digital technology. As algorithmic systems increasingly dictate social interactions and decision-making, digitcultism emerges as an inevitable consequence of efficiency-driven automation—gradually eroding human presence and meaning. A rational response to this crisis is ethical axiotropism: a reevaluation of ethical structures, a redefinition of key moral concepts, and a restoration of values necessary for sustaining genuine human autonomy in the algorithmic age.
Keywords: moral philosophy, artificial intelligence (AI), ethics, ethical axiotropism, optimization, digitcultism
Introduction
In recent years, algorithms have become not just tools of optimization but elementary forces shaping modern society. From search engines and social networks to facial recognition and automated decision-making systems, they influence how we access information, structure our daily lives, and interact with one another. The ever-growing integration of algorithms into nearly all aspects of social, economic, and political life—combined with the uncontrolled and uninterrupted development of digital data (Big Data) and advancements in machine learning—has sparked critical discussions among philosophers, legal experts, social scientists, and technology practitioners regarding their ethical impact. As algorithmic systems evolve, the tension between efficiency and ethical considerations becomes antagonistic, raising profound questions about autonomy, accountability, and the preservation of authentic moral values in an increasingly digitized world.
Clarifications regarding some core concepts
“Axiotropism” derives from the Greek terms ἄξιος (valuable) / ἀξία (value), and τροπή (orientation, turn) / τρέπειν (to turn, to direct, to change, to alter) that leads to tropism in biology, designating the oriented development of an organism in response to an external stimulus.
Growth towards the stimulus is defined as positive tropism, and growth in the opposite direction as negative tropism. Transposed into the sphere of ethics, and inspired by the meanings of metaphor understood by Lucian Blaga ([1937] 1985) as an ontological moment for the creative human being that “lives in the horizon of mystery” trying to reveal it, the ethical axiotropism functions both as a plasticizing metaphor—the individual organizes his existence around values, and as a revelatory metaphor—a deep search for fundamental values that transcend simple practical utility. The choice to live according to principles thus becomes an attempt to reveal the ultimate meanings of human existence.
In this perspective, ethical axiotropism combines a normative dimension (the plea for moral values) with an ontological one (the search for primordial values). Applying the idea of “tropism” in ethics suggests the existence of an orienting force—a natural tendency towards moral values—that makes the concept intuitive and expressive.
The distinction between positive and negative tropism can provide a key to reading contemporary ethical dynamics, such as the attraction towards authenticity or, on the contrary, the rejection of traditional values. However, the question remains open whether the axiotropism’s functions as a simple ethical norm, or whether we can consider it: a) A natural imperative, similar to a universal biological or psychological inclination? b) A philosophical imperative, a transcendent necessity deriving from human nature or the existential condition? Or c) On the contrary, a social construct configured within a historical era?
These questions open fertile directions for future research, in which the axiotropism can be explored not only as a metaphor but also as a possible foundation for a reconfiguration of ethics in the algorithmic frame.
An algorithm is a set of formal, clearly defined instructions that perform specific tasks in a logical, repeatable, and predictable manner. The origins of the concept can be found in the works of the Persian mathematician Al-Khwarizmi, considered to be the “father” of algorithms; but its functions have diversified significantly with technological progress, with unprecedented economic, social, and political implications. Algorithms are found integrated into the public and private sectors: institutional management, resource administration, monitoring, prevention, data analysis, social media platforms, optimizing access to information, the use of biometric techniques (e.g. facial recognition), e-commerce, trading, etc.
The social algorithm can be seen as a metaphor for the spectacle of human existence in all its grandeur, involving various aspects of life, such as studies, marriage, beliefs, ideologies, doctrines, career, migration and social order, the response of institutions to the challenges of the present, the cognitive modeling of citizens, the expansion or restriction of rights or freedoms, the disappearance of populations or cultures, digitalization, artificial intelligence (AI), decisions on the epistemic heritage that will be preserved for future generations, being, in fact, a reflection of human interactions and power lever strategies in a context of space and time.
An ethical algorithm could be defined as a set of procedures or rules implemented in an automated system that respects and promotes authentic moral values—understood as universal norms while acknowledging cultural influences—ensuring that the resulting decisions and actions align with Mill’s utilitarian principles, maximizing fairness and collective well-being (see Mill [1861/1863] 2009).
The algorithm should be subject to ethical axiotropic auditing mechanism (focused on transparency, accuracy and bias, security and privacy, ethical and social), designed to recognize and implement only values that are universal and independent of circumstances, to prioritize moral integrity as an absolute value, avoiding species erroneously labeled as virtues (which can generate social harm), taking into account that absolute authentic moral values cannot be contextual and influenced by culture, traditions and ideologies.
The algorithm’s decisions should be transparent and easy to understand. Aristotle (1988) wrote about what we can today identify as distinctions between values: universal human good, and virtue as a contextual means between extremes; Immanuel Kant ([1785; 1788] 1995) proposed the deontological principle (universally applicable rules); and John Stuart Mill ([1861/1863] 2009) discussed fairness in decision-making for the common good—ideas that have the quality of predisposition or even algorithmic norm. An ethical algorithm must be designed to prevent discrimination based on gender, race, economic status, or other criteria, and also to prevent the semantic degradation of specific moral terms. Some authors have noticed the tendency to remove morality in favor of ethics; perhaps because morality had a coercive character, which ethics has given up, retaining only an advisory aspect.
In a previous research (see Buzduga 2018), I indicated just three criteria to discern authenticity in the case of moral values and principles: a) the Boolean aspect (exists or not, e.g. sincerity, justice, honesty, etc.; any variation in intensity is a sign of non-validity, e.g. loyalty, sense of duty, etc.); b) the second criterion would be the support of φρόνησῐς (practical wisdom), which guides moral acts, guaranteeing that they are deliberate, just and oriented towards the common good; and c) the independence of moral values from the contractual character, advantage, interest or circumstance.
Data are the basic units of information, representing facts, values, numbers, or anything else that can be recorded and processed: structured data (numbers, texts organized in tables, SQL), unstructured data (free texts, messages, images, videos) and semi-structured data (XML, JSON documents, metadata, etc.). Along with algorithms, the Big Data phenomenon appears, referring to very large volumes of data, characterized by the famous 5 Vs: Volume, Velocity, Variety, Veracity and Value.
Taking into account all of these, a particular process gives way, namely digitcultism. And we are somehow witnesses to it.
Digitcultism is the shaping and transformation of cultural values under the effect of infestation with amoral optimization, generating new social, behavioral, and ethical norms based on algorithms and efficiency rather than traditional moral principles.
In contrast to “digitalism,” often associated with Nicholas Negroponte’s technological optimism (Negroponte 1995), “digitcultism” as presented here denotes a pathological territory. This concept is structured along two complementary axes: firstly, as a moral disorder characterized by an excessive submission to technologically imposed norms, eroding decisional autonomy and the perception of fundamental values; secondly, as a societal dependency, a collective pathology leading humanity to gradually relinquish its natural attributes in favor of digitization. This process influences perceptions of identity, creativity, and human interaction, redefining the relationship between the individual and society in the digital age. Digitcultism has been deliberately coined to emphasize the exaggerated scale of this phenomenon, illustrating how cultural values undergo radical transformation under the pressure of algorithmic efficiency, often detached from moral reasoning or human-centered principles.
Some considerations on ethics in the present algorithmic age
Contemporary ethical approaches have expanded in recent years to the field of AI and especially to algorithmic systems; therefore, ethics urges the formulation of legitimate critical questions:
- Who bears responsibility if an algorithm produces harmful effects like semantic term (“anti-ethical”) abuse of sustainability, political correctness, tolerance, removing common sense normality, censorship, minority rights against the majority, free will denial, social credit, prediction software, etc.?
- How can possible biases and discrimination caused by algorithms be prevented and corrected?
- What is the relationship between human autonomy in decision-making and automation?
- To what extent should the use of AI be regulated to protect human rights, such as privacy, autonomy and dignity, banning mental hacking interferences?
We are witnessing the tension (antinomy) that exists between optimized technological progress (pragmatic and predominantly autonomous) freed from any human constraints and moral or legal rigors because the algorithms (according to some general opinions) will become the most powerful entities in the world (outside of human control; an example would be that we have to prove online to verification algorithms that we are humans).
Concerning ‘amoral optimization’ and ‘algorithmic ethics’, they can be viewed as two opposed approaches to the creation and application of algorithmic technologies. Over the past few decades, rapid progress in areas such as machine learning, big data analytics, and decision automation has brought to light the crucial question: should we focus solely on efficiency and performance, or should we integrate moral and social principles into the development of these systems? Amoral optimization focuses exclusively on maximizing efficiency and performance, without considering the impact on people, communities, or the environment. “Amoral” does not necessarily mean “immoral,” but rather an ignoring of the ethical dimension. Amorally optimized algorithms could decide who gets credit, who has access to certain resources, or how natural resources are distributed based solely on data and performance targets, without considering discrimination, inequity, or other social consequences.
This type of “optimization” puts at risk by reinforcing existing biases, amplifying social inequalities, and leading to opaque decisions that people cannot understand or challenge. Essentially, amoral optimization prioritizes what is “functional” from a strictly technical perspective, sacrificing questions of justice, equity, and transparency.
The antinomies between amoral optimization and algorithmic ethics lie in a fundamental tension. The former ignores moral questions in the name of raw performance; the latter argues that technological performance must be tempered by ethical considerations. Where amoral optimization focuses on “How can we do this faster, cheaper, or more efficiently?”, algorithmic ethics asks, “How can we do this more fairly, more transparently, and more justly?” Ideally, the two approaches should not be completely irreconcilable. Algorithmic ethics is not about eliminating efficiency, but balancing it with accountability.
There are increasingly frequent disputes regarding the place that natural humans without technological augmentation will occupy in “the society of tomorrow”. It remains to be established what the ethical and legal limits are. So: How much is allowed to interact directly with the human brain? Who will have access to this technology? How will it be regulated, and will there be side effects of using algorithm-controlled devices: development of tumor tissue, manipulation, etc.? Is the extreme progress necessary? Do we really need cognitive improvement beyond human limits, such as improving memory, information processing speed (and even creativity), acquiring new skills (new ways of learning, direct brain-computer interface (BCI) command)?
To outline the complexity of algorithm ethics, it is essential to appeal to the main ethical paradigms in moral philosophy and to imagine how they can be integrated into the current context.
The deontological approach, mainly associated with Immanuel Kant, emphasizes respect for universal moral principles and absolute moral obligations. Applied to the context of algorithms, deontology emphasizes that regardless of the consequences that algorithms may have, they must align with certain principles, such as respect for human dignity, individual freedom and privacy. For example, a facial recognition system should not be used in ways that infringe upon fundamental rights, such as unauthorized mass surveillance, even if this could lead to faster capture of criminals or more effective prevention.
In contrast to deontology, the utilitarianism of John Stuart Mill evaluates actions according to their consequences, with the objective of maximizing happiness or the collective good. In an algorithmic context, a utilitarian framework would consider any algorithm that produces benefits that clearly outweigh the social costs to be “good.” For example, if an algorithm for triaging patients in the emergency room significantly reduces mortality and increases efficiency, despite the fact that its introduction may lead to some misclassification, utilitarianism may justify its use, provided that measures are implemented to minimize the negative effects.
In its turn, virtue ethics or an aretelogy (Cozma 2001), established by Aristotle and invigorated through the aretaic turn since the mid-20th century, focuses on the moral character of agents (especially developers, designers, and users of algorithms), and not just on rules and consequences. It is crucial that those responsible agents for designing and implementing algorithms demonstrate virtues such as courage but also prudence, respect, compassion, honesty, and especially medium and long-term responsibility. For example, peculiar software for emotion recognition system must reflect on whether the final application could be used for unethical purposes (political manipulation, psychological control, etc.) and act virtuously either by introducing technical limitations or by being transparent and publicly warning.
Conclusion
Only through a multidisciplinary and inclusive approach we can ensure that technology remains a force for human progress rather than a mechanism of control and inequality. Ethical axiotropism must not be an abstract philosophical exercise but a guiding principle for AI development, governance, and policy-making in the digital age.
References:
Aristotle. 1988. Nicomachean Ethics. Romanian translation by Stella Petecel. Bucharest: Scientific and Encyclopedic Publishing House.
Blaga, Lucian. [1937] 1985. Geneza metaforei și sensul culturii / The Genesis of Metaphor and the Sense of Culture. In Opere / Works, Vol. 9: Trilogia culturii / Trilogy of Culture, pp. 335-475. Bucharest: Minerva Publishing House.
Buzduga, Vicențiu. 2018. Some Ethical Issues Concerning the Employee Loyalty toward a Co-Worker and Group within Organizations. Agathos: An International Review of the Humanities and Social Sciences, Volume 9, Issue 1 (16): 145-148.
Copp, David (Ed.). 2006. The Oxford Handbook of Ethical Theory. Oxford: Oxford University Press.
Cozma, C. 2001. Introducere in aretelogie. Mic tratat de etică / Introduction to Aretelogy: A Short Treatise of Ethics. Iași: “Al. I. Cuza” University Publishing House.
Floridi, L. 2019. The Ethics of Artificial Intelligence. Oxford University Press.
Kant, Immanuel. [1785; 1788] 1995. Grundlegung zur Metaphysik der Sitten. Kritik der praktischen Vernunft. Romanian translation by Nicolae Bagdasar. Bucharest: IRI Publishing House.
Kearns, M., & A. Roth. 2019. The Ethical Algorithm: The Science of Socially Aware Algorithm Design. Oxford University Press.
Mill, J. S. [1861/1863] 2009. Utilitarianism. Auckland: The Floating Press.
Mittelstadt, B. et al. 2016. The ethics of algorithms: Mapping the debate. Big Data & Society, 3 (2). https://doi.org/10.1177/2053951716679679.
Negroponte Nicholas. 1995. Being Digital. New York: Alfred A. Knopf, Inc.
O’Neil, Cathy. 2016. Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. New York, NY: Crown Publishing Group.
Proposal for a Regulation of the European Parliament and of the Council laying down harmonized rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union legislative acts. 2021. Brussels: European Commission.
UNESCO. 2021. Recommendation on the Ethics of Artificial Intelligence. https://unesdoc.unesco.org/ark:/48223/pf0000380455.