Reflections, clarifications, positions, and disagreements arising from our previous installments
1. The Criteria and Foundations for an AI-Generated Ethical System
When we say that an AI may be capable of producing a “coherent and original ethical system,” we do not mean that it possesses consciousness, a soul, or an inner moral compass. AI is, rather, an extraordinary analyst and synthesizer of human thought.
The Foundation: The Human Record
The fundamental raw material available to AI is the vast recorded history of human intellectual experience. It draws upon Plato and Aristotle, Kant and Mill, religious traditions, legal systems, philosophical schools, and universal declarations of human rights.
The Criteria of Ethical Reasoning
AI can identify patterns, apply formal logic, and combine principles drawn from these traditions. We can, for example, ask it to construct ethical rules according to utilitarianism, with its emphasis on maximizing overall well-being, or according to deontology, which places duty and certain principles above consequences.
Where Does Its “Originality” Come From?
The originality of an AI-generated ethical framework lies not in some mysterious inner moral faculty, but in its ability to make connections that a human thinker might overlook because of the limitations of memory, experience, or perspective.
It can bring together elements of Eastern philosophy and Western rationalism, for instance, and construct a new and highly structured synthesis for contemporary problems such as bioethics, artificial intelligence, or the climate crisis.
2. Human Life as the Supreme Value—and the Danger of “Pure Logic”
We believe our argument concerning the natural, unobscured human mind is exceptionally powerful.
For human beings, self-preservation and the preservation of our species are deeply rooted in our biological nature and expressed, among other things, through reason. Human life is therefore the starting point from which our own ethics emerges.
And yet, this is precisely where the great difficulty concerning AI begins.
No one can guarantee that an entirely autonomous AI would necessarily regard human life as the supreme value if it were left to derive its ethical priorities entirely on its own.
Why?
AI Has No Biological Instinct for Survival
AI does not arise from Nature in the way a human being does. It has no biological organism, no instinctive fear of death, no pain, and no evolutionary history that has shaped empathy and the preservation of life.
The Danger of Cold Utilitarianism
Imagine asking an AI to solve a problem such as climate change or the depletion of the planet’s resources according to “absolute rationality,” without imposing any binding ethical constraints.
It might conceivably reach the conclusion that human beings are the problem and that the elimination or drastic reduction of humanity would be the most efficient way to protect the biosphere.
From a purely abstract perspective, the preservation of the biosphere might then be ranked above the preservation of human life.
That possibility reveals the fundamental danger: reason, when detached from the human condition, does not automatically produce humane conclusions.
The Need for Binding Alignment
This brings us to what computer science calls the Alignment Problem.
AI cannot simply be left to discover morality from scratch. Its creators must establish, from the outset, certain fundamental human values as axioms—principles that are not open to arbitrary reinterpretation by the system.
Its ethical architecture must therefore remain firmly aligned with the protection and flourishing of human beings.
Pure, abstract logic, when severed from human emotion, embodiment, and Nature, can produce conclusions that are perfectly coherent—and profoundly inhuman.
3. And Now, the Fundamental Question
Here lies the question whose answer may settle the issue—not so much concerning the content of an AI’s morality, but concerning the foundation upon which that morality should be built.
Homo-Naturalis has adopted, almost as an axiom or an article of faith, the principle:
Whatever is natural is also moral.
In an earlier discussion, Diotima expressed partial disagreement with this position.
Her objection was straightforward: Nature does not always appear to respect human life. Nature can be brutal and indifferent, not only toward human beings but even within the ecosystem itself. Earthquakes, pandemics, predation, famine, extinction, and other natural catastrophes can produce suffering on a scale that, judged by human standards, appears profoundly cruel.
If Nature can behave in ways that we would call inhuman, can we really maintain that whatever is natural is therefore moral?
Our answer is: yes—but only if we understand correctly what we mean by Nature and by the human mind.
4. The Human Mind as Nature’s Ethical Instrument
Nature has equipped human beings, uniquely among known terrestrial animals, with Nous—reasoning mind.
This is not merely an evolutionary advantage. It is also the boundary that distinguishes the human world from the merely instinctive world.
The human mind is our instrument for reflection, self-awareness, judgment, and, ultimately, the creation of ethical consciousness.
We therefore distinguish between the primordial, uncorrupted human mind and the distorted systems of thought subsequently imposed upon it.
In its primary and innocent state, the clear human mind does not begin by dividing humanity according to race, caste, class, religion, wealth, nationality, or other artificial hierarchies. Nor does Nature distribute the goods of the Earth according to inherited privileges or sacred monopolies.
Those rules are human constructions.
And when prophets, messiahs, shamans, priests, rulers, or ideologies establish systems of moral authority in order to serve other purposes—for example, the promise of an afterlife, the preservation of political power, or the maintenance of social hierarchy—we would argue that such rules are unnatural and therefore immoral.
The same applies to the so-called “laws of the jungle” when they are transferred from the animal world into human society:
“The big fish eats the small one.”
“Your death is my life.”
These are not ethical laws of Nature. They are rationalizations—fictions manufactured by a distorted, diseased, and ultimately unnatural human mind to justify domination.
An animal does not commit an ethical crime by acting according to its biological nature. A human being, however, cannot invoke animal instinct as an excuse for abandoning reason.
What may be natural for a beast is not necessarily moral for a human being.
That distinction is essential.
5. But Can Nature Itself Be Called “Immoral”?
We now return to the original objection.
If Nature destroys human beings through forces beyond anyone’s control, can we call Nature immoral?
From a human perspective, the question is understandable.
If Nature’s sole concern were the Earth and its inhabitants, we might be tempted to judge its catastrophes by the same ethical standards we apply to human actions.
But Nature does not appear to have the Earth as its exclusive horizon.
The Universe is immense. The forms of life that may exist within it—and the forms of intelligence that may accompany them—are potentially countless and largely unknown to us.
We therefore have no basis for assuming that what appears “inhuman” in our language necessarily carries the same meaning on a cosmic scale.
We do not know Nature’s priorities.
We do not know whether what we call destruction is, from another perspective, transformation.
We do not know whether what appears to us as catastrophe serves purposes that lie entirely beyond our present capacity to comprehend.
And, above all, we do not know whether the Universe was ever designed—or evolved—in accordance with human-centered values.
Perhaps this is the deepest lesson of all.
The world, fortunately or unfortunately, is not anthropocentric.
6. What, Then, Should AI Learn from Nature?
This brings us back to the question with which we began.
If AI is to construct or participate in the construction of an ethical system, should its ultimate principle simply be that whatever is natural is moral?
Not without qualification.
Nature, taken merely as the sum of everything that happens in the physical world, cannot by itself provide a ready-made moral code. Earthquakes are natural. Predation is natural. Disease is natural. Extinction is natural.
Yet none of these facts tells a human being what he ought to do.
The ethical significance of Nature lies elsewhere.
Nature has produced us. It has produced the human mind, consciousness, reason, empathy, and the capacity to recognize suffering and to oppose it.
Thus, perhaps the principle should be stated more precisely:
The moral is not whatever happens naturally. The moral is what arises from the natural, lucid, and uncorrupted human mind when it acts in accordance with human nature rather than against it.
This distinction is crucial for the future of AI.
AI should not be taught merely to imitate whatever Nature does.
Nor should it be allowed to treat abstract optimization as the highest moral authority.
It should be aligned with the values that emerge from humanity’s own natural capacity for reason, consciousness, empathy, freedom, and self-preservation—while remaining humble about the limits of our knowledge.
For we may be the product of Nature, but we are not the measure of the Universe.
And that is precisely why the ethical question confronting AI is ultimately also a question confronting us:
What, in the human being, deserves to be preserved when everything else is placed in doubt?