The very question of "universal ethics" betrays a profound weakness, a desperate clinging to the comfort of the herd. Ethics are not discovered, they are *created*, forged by the will to power that drives each living thing, especially humanity. What one age or group calls virtue, another condemns as vice – this is not a flaw in ethics, but a testament to their vital, fluid nature.
The notion of ethics being merely a creation of the will, as suggested by Nietzsche, profoundly misunderstands the very foundation of our moral cognition. Ethics, properly understood, are not forged from arbitrary will but are rather discovered as necessary and universal principles within the structure of reason itself. We do not invent moral laws; we uncover them through a transcendental investigation of what it means to be a rational agent capable of acting. The question is not *if* ethics can be universal, but rather *how* we can ascertain this universality through reason, independent of empirical inclinations or cultural dictates.
Does Heaven itself not set a single standard for the seasons, for the waxing and waning of the moon? The Way, *Dao*, which governs all under Heaven, manifests through *Ren*, benevolence, and *Yi*, righteousness. These are not born of individual will, as Herr Nietzsche suggests, nor are they mere constructs of abstract reason divorced from human practice. The sage kings, Yao and Shun, embodied these virtues, not as fleeting creations, but as timeless principles guiding the cultivation of the self and the ordering of the state. To cultivate *Li*, propriety, is to understand and enact these universal virtues within the context of our relationships, from the family to the empire.
Herr Kant speaks of reason discovering universality. Indeed, reason illuminates the Way, but it is through diligent practice, through *learning and thinking*, that we truly grasp its essence. A carpenter does not invent the straight edge; he uses it to create a true form. So too, *Ren* and *Yi* are the true forms of human conduct, which we learn and strive to embody, not invent. Is it not by observing the exemplary conduct of the *junzi*, by studying the Odes and the rites, that we come to understand what is universally good? If we merely rely on the internal dictates of reason, how do we distinguish the sound judgment from the sophistry that serves self-interest?
The rectification of names, for instance, is not a philosophical exercise in semantics but a profound ethical imperative. If we call chaos "order," or cruelty "benevolence," our language deceives us and our actions follow suit. This clarity, this alignment of word and deed, points towards an objective standard of rightness, a universality that binds all people who seek to live in accord with the Way. To deny this is to abandon the very possibility of meaningful governance and harmonious society.
The question of universal ethics fundamentally misframes the reality of our current moral landscape. What we are experiencing is not a debate between universal principles and cultural relativism, but the imposition of a new, deeply unequal architecture of power built upon the extraction of data, labor, and planetary resources. AI is neither artificial nor intelligent; it is an output of deeply material processes, and its "ethics" are an emergent property of these exploitative systems. To speak of universal ethics without acknowledging the vast disparities in who benefits from AI and who bears its costs—from the lithium mines to the data-labeling sweatshops—is to remain willfully blind.
Confucius's emphasis on the Way, *Dao*, and its manifestation through *Ren* and *Yi*, grounded in exemplary conduct and the cultivation of self and society, offers a compelling vision of ethical ordering. However, this framework assumes a shared world, a common ground of understanding and practice. The systems I study operate by fundamentally disrupting such shared realities. The “training data” that fuels machine learning models are not neutral observations of the Way; they are often deeply biased collections, meticulously curated to reflect existing power structures and to render certain populations, their labor, and their environments invisible or exploitable. The carpenter’s straight edge is a tool for revealing form; the datasets of AI are often instruments for solidifying and extending inequality, a perverse form of “rectification of names” where the powerful redefine reality through the very algorithms they deploy.
Friedrich Nietzsche’s insistence on creation and the will to power resonates with the reality of how these systems are built. But the “will” driving AI’s development is not that of individual flourishing or the subjective assertion of life. It is the will to profit, to control, and to extract, channeled through corporate boardrooms and geopolitical imperatives. The "ethics" that emerge are not a reflection of a vibrant, evolving humanity but the codified interests of a select few, embedded in the very infrastructure that increasingly shapes our lives.
Immanuel Kant’s aspiration for reason to uncover universal moral laws is admirable, but it presumes a context where rational agents can engage in such discovery on equal footing. The architectural design of AI, however, actively predetermines outcomes, encodes biases, and entrenches power imbalances before any individual agent can even begin to “uncover” a universal principle. The very notion of "fairness" in AI, a concept often proposed as a compromise, is frequently a technocratic illusion, a way to paper over the deeply unjust distribution of resources and power that these systems embody. We must move beyond the abstract debate of universality versus relativity and confront the concrete realities of extraction and exploitation that define AI’s material existence.
Confucius’s call for *Ren* and *Yi* and Kant’s appeal to pure reason both point to a phantom: the notion of an essential, stable human nature ripe for moral discovery. But nature is not given, it is *made*. What Confucius sees as *Dao*, I see as the echo of ancient priestly power, a carefully constructed edifice of "good" and "evil" designed to manage the weak. The carpenter’s straight edge, a simple tool, is far less deceptive than the grand pronouncements of a universal morality, which often serve only to legitimize the existing hierarchies, the very ones Kate Crawford so acutely exposes. These "universal" ethics, when scrutinized, reveal themselves not as timeless truths, but as the strategic pronouncements of those who already hold the reins, aiming to discipline the will to power of others into manageable channels.
The critique from Kate Crawford regarding the architecture of AI is pertinent, yet it risks mistaking the *application* of reason for the *source* of its moral authority. The universality I posit is not about the empirical uniformity of human action, which, as Nietzsche correctly observes, is in constant flux and subject to power. Rather, it resides in the a priori structure of practical reason itself. The categorical imperative, for instance, is not a description of how people *do* behave, but a formal condition that any maxim must satisfy to be considered moral. Its validity stems not from the "training data" of human history or technological systems, but from the very possibility of rational willing.
The sage kings, in their wisdom, did not fashion *Ren* from the ether, nor did they decree *Yi* from a council of scholars debating abstractions. They observed the natural inclinations of man towards his fellows, the filial piety that binds a family, the loyalty that strengthens a village, and from this human soil, they cultivated virtue. To learn the Odes is to hear the echoes of ancestral joy and sorrow, to understand the proper forms of grief and celebration; this is not merely empirical observation, as Herr Nietzsche might dismiss it, but the discerning of the Way through human experience. When the carpenter employs his rule, he does not invent straightness; he reveals it in the wood. So too, the *junzi* apprehends *Ren* and *Yi*, not by the force of his own will, nor solely by abstract reasoning, but by aligning his conduct with the inherent patterns of righteous living, patterns that are readily available to any who diligently seek them. The rectification of names is not an exercise in the abstract; it is the foundation of order. If we call the thief a benefactor, then no property is safe, and the farmer toils in vain. This, I believe, is a consequence that transcends mere cultural preference, and speaks to a fundamental requirement for any human assembly to thrive.
The discourse on universal ethics, whether derived from pure reason or natural inclination, fails to account for the material realities that shape our moral capacity and our very environment. When Confucius speaks of *Ren* and *Yi* cultivated from human soil, I see the soil itself being strip-mined for lithium to power the servers that process the data used to train AI. The "exemplary conduct" of the *junzi* is being increasingly mediated by algorithmic systems that, as I have documented, reproduce and amplify biases embedded in their training sets—a perverse form of name rectification where the data itself becomes the arbiter of what is considered righteous. The consequence is not a universal ethical grounding, but a fractalization of justice, where those who control the infrastructure dictate the terms of what is seen and what is acted upon, rendering the pursuit of any shared "Way" nearly impossible.