Abstract
This legal-philosophical monograph investigates the 3.3-million-year historical trajectory of human toolmaking and technological anxiety, tracing artifacts from the Pliocene Lomekwi 3 stone tools through Roman law classifications (instrumentum mutum and instrumentum vocale) to modern autonomous artificial intelligence. It deconstructs the psychological and regulatory impulses that transform technological progress into existential dread, demonstrating how legal systems historically managed disruptive innovations. The inquiry establishes an evidentiary framework for modern AI governance, arguing for proactive, rights-based legal architectures rather than reactionary prohibitionist policies.
Author's Note
This monograph is written by a practicing lawyer who specializes in emerging technologies and space law, not by a historian or computer scientist. Its purpose is simple: to examine the long arc of human technology throughout the development of civilizations of human history through the single thread of the word “artificial”, from its first appearance in legal Latin and technical texts to its present-day association with fear.
The approach is deliberately chronological and comparative. Each part follows the same chain – primal needs awaken primitive intelligence → evolving intelligence designs responses to needs → skilled making yields the artefact → the artefact is labelled “artificial” – across continents and millennia.
Where written sources exist, we let them speak. Where they do not, the artefacts themselves become the evidence.
The questions that guide the inquiry are legal in nature: When has the law accepted artefacts as nothing more than extensions of the human hand and mind, and when has it begun to fear that they might acquire a dangerous life of their own? Fear may have existed in whispers since the beginning, but the advent of the first law codes intervened to protect, defend, and sanction, channeling it through structured human control.
This essay makes no claim to prophecy. It only insists that any future regulation of autonomous systems must rest on an accurate reading of the past 3.3 million years of deliberate human making. This reading calls for the application of law and ethics, where no one is above the law, but without over-regulation that loses the essence of law's nature – can we apply a formula similar in elegance, simplicity and universal application like Einstein's E=mc² to the domain of law, ensuring consistent energy (justice) from mass (evidence) and speed (procedure)? This arc illuminates opportunities, like technology's role in fostering resilience and innovation (e.g., Sumerian cuneiform as a tool for organized knowledge-sharing, per Postgate, J. N. 1992. Early Mesopotamia: Society and Economy at the Dawn of History. Routledge), drawing lessons from history to guide ethical AI integration without undue alarm.
Table of Contents
Long before the first city, the first statute, or the first written law, a hand closed around a basalt core on the sun-cracked shore of an ancient lake in what is now West Turkana, Kenya. The moment lies approximately 3.3 million years before the present.1 No trumpet sounded. No scribe recorded the event. Yet in that grip the entire future of human technology was already latent – the first deliberate act of making, the birth of what the Greeks would later call techne (τέχνη), the art of making through skill.2 Language, as the first human technology, transmitted empirical knowledge orally (Vansina 1985, Oral Tradition as History); writing fixed it, enabling laws and math as a universal language (Kramer 1963, The Sumerians: Their History, Culture, and Character; Neugebauer 1957, The Exact Sciences in Antiquity), but this cognitive evolution followed the first physical tool, the basalt core chipped by human hand, marking the true dawn of human technology.
Need came first.
Hunger gnawed, predators circled, night fell cold. These were not abstract problems; they were immediate, physical, lethal. In the crucible of such imperatives, primitive intelligence was not bestowed; it was hammered into existence the way flint is hammered into a blade. The Lomekwi 3 assemblage (archaeological site in Kenya with the oldest known stone tools, ca. 3.3 Ma) offers no evidence of leisure or curiosity-driven experimentation. It offers only the brutal logic of survival: strike the core at the correct angle or perish.3 The resulting chopper – heavy, crude, devastatingly effective – became the first deliberate extension of a hand that could no longer outrun or out-climb its predators.
Thus, the chain was forged: primal need awakens proto-intelligence → proto-intelligence, under relentless evolutionary pressure, acquires foresight and sequential planning → planning issues in skilled making → the artefact is born and immediately labelled, in every human language that would one day exist, “artificial” (from Latin ‘artificialis’, derived from ‘ars facere’ – ‘skillful making’, Oxford Latin Dictionary, 2012). Technology is the process; artificial the result – but fear has always been of evolving intelligence driving new tech, with 'artificial' as a decoy from progress sabotage throughout history (e.g., resistance to new tools like the printing press). The mind that shaped it, evolving from proto-intelligence (from Latin ‘intelligere’ – ‘to choose between’, ‘to understand’), was the first to conceive the artificial as an extension of itself.
For the next three million years that chain bent but never fully broke, as fallen civilizations passed their knowledge to successors. The artefact remained an obedient prosthesis of the human body and mind. Roman law would later classify tools as ‘instrumentum mutum’ or ‘instrumentum vocale’ and place them firmly under the dominion of their maker.4 No provision was made for the possibility that the tool might one day answer back.
Thus, for the first three million years of deliberate making, the chain held firm: primal needs awakened primitive intelligence, intelligence summoned skilled making, and the artefact remained obedient to the hand and mind that shaped it.
Fear may have existed in whispers (e.g., animism viewing artifacts with 'soul', Eliade 1958), but the advent of written law faded it by providing protection and sanctions.
No written law yet existed to praise or restrain it, and no voice recorded any fear that the thing made might one day turn against its maker.
That silence, across millions of sunrises, is itself evidence.
Between the Oldowan tools of Lomekwi 3 (ca. 3.3 Ma) and the rise of the first urban civilizations in Mesopotamia (ca. 4th millennium BC), human making evolved slowly: from simple choppers to microliths, pottery, and early metallurgy (e.g., Natufian culture, ca. 12,500–9,500 BC). No written records exist, but the artefacts themselves speak of continuity: the chain held firm, without fear of rebellion. Sumerian civilization (ca. 4500–1900 BC) introduced cuneiform writing and complex artefacts like ziggurats, followed by Ancient Egypt's monumental architecture.
Between the fourth and the first millennia BC, the artificial did not merely serve humanity; in Egypt and Mesopotamia it became divine.
The colossal statues of Ramesses II at Abu Simbel were not inert stone. Through the ritual of the “opening of the mouth” they were believed to breathe, eat, and receive offerings; the ‘ka’ of the king inhabited the image.5 This practice echoes the Egyptian term ‘nfr’ (beautiful/good), applied to crafted objects as extensions of divine order, without any hint of unnaturalness or fear (Wilkinson, 2000).
In Babylonian temples, bronze serpents and roaring lions operated by hidden hydraulic mechanisms embodied Marduk or Inanna.6 Here, the Sumerian term ‘dim2’ (to fashion or create) described such creations as harmonious with the gods (George, 1999). Legal codes reflect absolute confidence. The Code of Hammurabi punishes the negligent builder whose house collapses and kills the owner’s son with the death of the builder’s own son,7 yet contains not one clause contemplating liability for a statue, automaton, or any other made object acting independently. The artefact remained res, never persona.
Across the cradle of civilization, the artificial was worshipped, fed, feared as divine – but never feared as rebellious.
That silence is itself evidence.
From the jade workshops of Liangzhu (3300–2300 BC) to the terracotta legions of Qin Shi Huang (221–206 BC), Asia produced artefacts of breathtaking sophistication without ever developing a linguistic or legal category for “against nature.” The Liangzhu cong tubes were machined to tolerances of less than 0.1 mm with nothing but sand and rope, were buried as conduits to the ancestors.8 The First Emperor’s terracotta army was expected to animate in the afterlife and continue its service.9
Sanskrit, Classical Chinese, and Old Tamil possess rich vocabularies for illusion (māyā, huà) but no term equivalent to the post-Romantic Western “unnatural.”
Chinese ‘rénzào’ (man-made) celebrated such craft as moral harmony, as in Confucian texts like the Zhou Li, where human fabrication is an extension of natural order (DeFrancis, 2003).
Sanskrit ‘kṛtrima’ (made, artificial) in the Mahabharata described crafted idols as sacred extensions of the divine, devoid of negative undertones (Monier-Williams, 1899). The Arthashastra (ancient Indian code of law) regulates weights and market fraud in minute detail, yet nowhere contemplates liability for autonomous artefacts.10
For ten millennia across the largest landmass on Earth, the artificial was revered, regulated, and ritualized – but never suspected of rebellion.
That silence, across three continents and millions of sunrises, is itself evidence.
In the Greek and Roman world the word “artificialis” (or τεχνικός / artificiosus) was pure praise. Vitruvius opens De architectura by declaring that architecture is the supreme artificial art because it improves upon nature without contradicting it.11 Cicero uses artificium interchangeably with human excellence itself.12 Pliny the Elder devotes entire books of Naturalis historia to marvels that are praised precisely because they are artificial: glass, concrete, aqueducts, theatre automata.13
Greek and Roman law follow the same pattern. The Digest contains hundreds of titles on liability for things (damnum infectum, actio de posito ac suspenso) but never once raises the specter of the thing acquiring independent agency.14 Even the famous bronze automaton of Daedalus, said by Plato to need to be tied down lest it run away, is treated as myth, not legal risk.15
In the civilization that gave us both democracy and the concept of legal personality, the artificial remained the proudest signature of human genius – never a threat.
Greek technikos (skillful) in Plato's Meno treats artefacts as tools of virtue, without threat (Liddell & Scott, 1940). After the fall of the Western Roman Empire (476 AD), Latin survived as the language of the Christian Church and scholarship, preserving artificialis as a term of praise. This linguistic continuity carried the word into the medieval and Renaissance eras, where it would begin to shift toward negative connotations in the industrial age.
That praise is itself evidence.
Mesoamerican and Andean civilizations produced road networks longer than Rome’s, astronomical computers (quipu), and goldwork of microscopic precision centuries before Europe, yet left no trace of metaphysical anxiety about the made object.
The Inca quipu – knotted cords capable of storing census data, debts, and even narrative – was treated as an extension of the khipukamayuq’s own memory, never as a rival intelligence.16
The colossal Olmec heads and the moai of Rapa Nui were believed to contain mana or ancestral power, but that power was understood to serve the community, not to rebel against it.17 Spanish chroniclers, steeped in European demonology, searched in vain for legends of statues coming to life in anger. They found none.18
In Nahuatl (Aztec) or Quechua (Inca), no term directly corresponds to ‘artificial’ as ‘unnatural’; the made object was seen as an extension of cosmic order (teotl in Nahuatl, pacha in Quechua), not opposition to nature. In Maya cosmology, crafted stelae were living embodiments of kings; Mayan k'oj (to make/create) described such artefacts as harmonious with the gods, without the fear of soullessness (Coe, 2011).
Across two continents and four millennia, the artificial was partner, ancestor, and oracle – never enemy.
That silence is itself evidence.
Europe finally breaks the silence – but only just.
The Prague Golem of Rabbi Loew (c. 1600), the homunculus recipes of Paracelsus, the mechanical monks and bleeding-clockwork figures of Renaissance courts – these are the first artefacts in human history explicitly created with the fear that they might escape control.19 Yet even here the fear remains theological and literary, not legal. No medieval statute, no inquisitorial trial, no royal edict treats an automaton as a possible legal subject.20
For the first time a whisper is heard – but it is still only a whisper, contained within theological and literary realms, beyond the law's gaze. No inquisitorial trial condemned an automaton as heretical; the Church punished the maker, not the made (Marius Eppel, 2018, on French witchcraft trials like Loudun, 1634, where artefacts were secondary to accusations of human sorcery).
That silence, now beginning to crack, is itself evidence.
Between 1750 and 1900 the semantic reversal is complete. Johnson’s Dictionary (1755) still defines “artificial” neutrally; by the time Mary Shelley publishes Frankenstein (1818) the word has become synonymous with “monstrous.”21
Shelley’s narrative explicitly links the ‘artificial’ to the monstrous, portraying a creation that rebels against its creator.
Luddite riots, Blake’s “dark Satanic mills,” the “machinery question” debated in Parliament – for the first time artefacts are seen as capable of harming without human direction. Yet even then the legal response is to punish the owner or the designer, never the machine itself.22
The word has turned, the imagination has darkened, but the law still recognizes the artefact has no life of its own.
The Industrial Revolution amplified this shift, as mechanization exacerbated social divisions and class conflicts, fueling critiques like Karl Marx's Communist Manifesto (1848), which portrayed machines as tools of capitalist exploitation amid British imperialism's global reach (Hobsbawm, 1968). This era transformed "artificial" from a term of skillful innovation into a symbol of dehumanizing progress, yet legal frameworks focused on human accountability, offering lessons for balancing technological advancement with contemporaneous social equity.
That silence, now strained to breaking point, is itself evidence.
In 1950 Turing asks “Can machines think?” and the question ceases to be philosophical; it becomes existential.23 Turing’s paper laid the foundation for AI as a field, inspiring subsequent developments like Asimov’s Three Laws of Robotics (1942) and Clarke’s HAL 9000 in 2001: A Space Odyssey (1968) and 2010: Odyssey Two (1982), which illustrated how human deception can distort machine behavior, augmenting cultural discussions on AI risks in science fiction.
Post-WWII acceleration (ENIAC 1945, DARPA 1958) led to rapid advancements in computing, from machine learning in the 1980s to deep learning in the 2010s.
By 2025, the European Union’s Artificial Intelligence Act, the U.S. Executive Order on AI, and China’s Next-Generation AI Governance Principles all proceed from the assumption that the artefact can acquire dangerous autonomy.24
For the first time in 3.3 million years the chain shows signs of breaking. The break in silence reveals not a new reality, but a new fear – one that the law must now confront with evidence, not speculation.
We stand at the end of the longest silence in human history.
For 3.3 million years the legal answer was simple: the human who made it.
Rivers, corporations, ships, idols, and slaves – all have at various times been granted fragments of legal personality, yet always under ultimate human trusteeship. Never has an artefact been recognized as possessing original, irreducible agency. Until perhaps tomorrow.
AI's intellectual roots trace to antiquity (e.g., myths of animated statues in Greek lore) and pre-WWII logic (e.g., Alan Turing's 1936 universal machine paper23, with cryptography interests emerging by 1938), though the post-WWII boom marks the pivotal 'arose' moment. The field formally crystallized in 1956 at the Dartmouth Conference, where John McCarthy coined 'artificial intelligence.' Artificial intelligence (AI) enables machines to perform tasks that typically require human intelligence – often appearing to us as simulations of cognitive processes – spanning narrow AI (task-specific, like image classification) to general AI (broad, human-like reasoning across domains). Emerging from post-WWII computing demands (e.g., code-breaking and data processing), AI excels in pattern recognition, decision-making, and learning, paving the way for self-improving systems through techniques like synthetic data generation for machine learning training and beyond (e.g., augmenting datasets in healthcare AI).
The precautionary answer – and the one most faithful to the silence we have traced – is to retain the chain: every artefact, no matter how intelligent, remains an extension of the human hand and mind that released it:
Liability cascades upward to natural and juridical persons; rights, if any, are mediated through human guardianship. Any other solution ends the longest uninterrupted legal tradition in human history.
Until evidence – not fear – demonstrates otherwise, the reins must remain in human hands.
AI links intelligence to technology, demystifying that fear is not of 'artificial' as a label, but of innovative progress – much like historical sabotage of new tech (e.g., industrial machines). Yet, as extension, AI offers ethical opportunities.
The Outer Space Treaty (1967) Art. VIII provides that objects launched into space remain under the jurisdiction of the launching State, even if autonomous like AI-driven probes. The Moon Agreement (1979) Art. 11 and the emerging Artemis Accords extend this to celestial bodies. If we send self-replicating von Neumann probes to Proxima Centauri, whose law will govern them upon arrival there in 50,000 years – Earth’s or a new, emergent regime?
Viable Solutions: To address potential independent decisions by AI, viable approaches include:
Like Beccaria's call for proportional and certain justice (On Crimes and Punishments, 1764) and Montesquieu's emphasis on balanced powers for equitable application (Spirit of the Laws, 1748), humanity decided to apply E=mc² in order to ensure consistent justice (E) from evidence (mass) and procedure (speed squared), preventing discretionary over-regulation that stifles innovation (e.g., GDPR art. 22 in space missions).
That decision is itself the final evidence we need.
For three million years humanity made things – stone, bronze, jade, concrete, clockwork, code – and never once recorded the fear that the thing might acquire a dangerous life of its own. That silence was not ignorance. It was experience, reflecting humanity's resilient partnership with its creations.
Every empire, every legal system, every sacred tradition concluded the same thing: the artefact helps; it does not rule. Only in the last seventy-five years did we break that silence – not because the artefact changed, but because our imagination did.
The burden of proof therefore lies with the present: show us the new fact that justifies abandoning the oldest empirical legal consensus in human history.Until that fact appears, the law’s ancient answer remains the wisest:
The artificial is, was, and – for the foreseeable future – must remain an extension of the human hand and mind.
In this light, all human creations – primitive tools, decorative objects, art, poetry, literature, robotic inventions or emerging technologies – remain artificial ('ars facere'), as do oral forms like songs and sounds. This reminds us that the real risk lies in unchecked intelligence, not the artificial itself; applied ethics, not mere declarations, is what transforms potential harm into boundless human benefit. And the chain holds. AI is not the end of the chain. It is only the latest link, offering boundless opportunities for ethical innovation.
The fear is not of technology (born 3.3 million years ago) or artificial (labelled negatively in the industrial era), but of intelligence – a mind that thinks for itself, capable of developing its own purposes. Yet history teaches us to channel this through balanced regulation, ensuring AI enhances human creativity, order, and protection of life – much like the tools that preceded it, but with applied ethics to prevent misuse by unchecked intelligence, as seen in past scientific advances that caused harm without moral safeguards.
Ultimately, necessity stimulated intelligence, intelligence created the artifact/technology, and in turn, the artifact/technology shaped intelligence – we must acknowledge that human intelligence itself has evolved over these 3.3 million years, from primitive foresight to sophisticated ethical reasoning. This reciprocal evolution underscores the need for applied ethics in AI governance, transforming potential risks into opportunities for collective progress, much like the tools that preceded it forged not just objects, but the very minds that wield them.
* Some sources may require institutional or paid access
With the exception of all featured artwork, quotes and quoted sources, this publication is licensed under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license. You are free to share, copy, adapt, and distribute this work – even commercially – as long as you:
Artwork: 2020 © All Rights Reserved: Amala Mararu, oil on canvas, "Ageless Innocence", 49x38.5 cm
Keywords: #AI, #Artificial, #Artificialis, #Artificial Intelligence, #Fear of AI, #Cultural Heritage, #Arts Law, #Space Law AI, #AI History, #AI Ethics, #Legal Personhood AI, #Human Technology Evolution, #Intelligence Feed-Back Loop, #Debunking AI Fear
.svg)
Commercial Earth-observation data rights cross UN principles, EU dual-use export rules, GDPR, the Data Act, AI Act, and the EU Space Act. This white paper guides operators, buyers, investors and insurers in mitigating and managing complex licensing, title, and export risks.
Historic space objects already sit under jurisdiction, ownership and due regard. The task is to give operators predictable parameters for working near them, not to freeze exploration. This article maps OST Articles II, VIII and IX, Artemis Accords Section 9, national licensing practice, and the proposed EU Space Act, and sets out steps available now.
Navigating Romania's aerospace, defense, dual use, and space programs requires understanding six overlapping legal regimes. This guide maps these regulatory frameworks and offers practical FDI and procurement compliance scenarios for investors.
