Contemporary debates on technology remain paralyzed between two symmetrical illusions: the instrumentalist alibi and the fatalist myth of autonomy.
The instrumentalist view, rooted in naive Cartesian common sense, insists that technological artifacts are value-neutral vessels entirely subordinate to the moral intentions of their users. In this telling, a rifle, an algorithmic credit-scoring system, or a generative neural network possess no inherent normative trajectory; only human will determines their virtue or vice. This view conveniently ignores how technical systems structure possibility spaces, foreclose alternatives, and enforce operational disciplines before a user ever engages them.
At the opposite extreme stands the fatalist dogma of systemic autonomy. Popularized in twentieth-century philosophy of technology by Jacques Ellul’s technological system and revived in Silicon Valley’s transhumanist accelerationism, this position claims that modern megatechnics has decoupled from human intentionality. Technology is depicted as an autonomous, self-augmenting leviathan that obeys only its internal imperatives of optimization, speed, and efficiency. Living beings are cast as passive conduits or obsolete biological friction forced to conform to machine rhythms.
Though ostensibly opposed, these two stances are functionally complicit. The instrumentalist alibi absolves technology of systemic critique; the fatalist myth absolves its human architects of moral, political, and epistemic responsibility.
The Abdication of the Architect
The apparent "autonomy" of modern technical systems is neither a metaphysical destiny nor a law of physical nature. It is a social and psychological projection: the alienated externalization of choices made by human designers, software architects, and institutional sponsors.
When an automated administrative pipeline denies a welfare claim without human explanation, when an algorithmic trading system crashes a financial market, or when a predictive model generates racialized risk scores, the standard apologetic recourse is systemic inevitability: "The algorithm reached the optimal statistical distribution; technology follows its own evolutionary course."
This rhetoric operates as an apparatus of moral absolution. It allows human engineers, corporate executives, and regulatory architects to offload accountability onto an impersonal computational matrix. Inanimate code does not possess desires, intentions, or systemic drives. What is experienced as the irresistible momentum of an "autonomous system" is the accumulated weight of concrete human decisions—metric selections, objective functions, optimization constraints, and funding vectors—carefully shielded from public deliberation and disguised as technical necessity.
To break this fatalism, philosophy must reclaim the origin of normativity: norms do not originate in silicon or steel; they originate in the living organism.
Canguilhem and the Vital Root of Technique
In The Normal and the Pathological (1943) and his seminal essay Machine and Organism (1952), Georges Canguilhem fundamentally overturned the Cartesian reduction of life to mechanics. Descartes explained the organism by comparing it to an automated clockwork or hydraulic fountain. Canguilhem demonstrated the historical and epistemological inversion of this model: machines are derived historical artifacts modeled on vital bodily functions, not the reverse.
For Canguilhem, life is not passive adaptation to an indifferent environment; life is normative activity (activité normative). An organism establishes positive and negative values, institutes physiological thresholds, and actively shapes its surroundings to secure its existence. A living being does not merely obey existing physical laws; it introduces novel norms into the physical universe.
Within this framework, technique is not an alien, demonic force hostile to life. Technique is a primary biological strategy—a prolongation and projection of the living organism’s normative impulse against environmental hostility. A flint blade, an optical lens, an engine, and an operating system are material extensions through which the living agent negotiates physical friction and extends its sensory-motor horizon.
When a technical system begins to impose rigid, pathological, and alienating norms upon human life—subordinating the worker to an assembly-line stopwatch or the citizen to an opaque automated metric—the problem is not that "technics has overcome biology." The problem is that one group of human norm-makers has weaponized a technical apparatus to enforce their particular economic or political norms upon other living beings, all while disclaiming responsibility through the fiction that the machine is autonomous.
The Critical Bifurcation: IA versus AI
This biological grounding demands a radical conceptual demarcation in contemporary technical design, best captured by the historical divergence between Douglas Engelbart and the pioneers of early artificial intelligence: the choice between Intelligence Augmentation (IA) and Artificial Intelligence (AI).
In his foundational 1962 framework Augmenting Human Intellect, Douglas Engelbart defined the purpose of computing not as the creation of artificial minds, but as the systematic expansion of human problem-solving capacity. IA views the technical artifact as an open-ended cognitive and material prosthesis. Like an artisan’s chisel or a mathematician’s chalkboard, an augmentative system preserves the living agent as the locus of judgment, exploration, and normative institution. It amplifies human perception, structures memory, and visualizes complex multidimensional relationships, while keeping meaning, verification, and moral consequence firmly anchored in the embodied thinker.
Autonomous "Artificial Intelligence," by contrast, has largely pursued the fantasy of automated substitution. Driven by the computationalist assumption that thinking is merely disembodied syntactic shuffling, the AI paradigm seeks to construct autonomous decision-making engines that displace human agency. In practice, this does not eliminate human judgment; it buries human judgment inside proprietary datasets, cost functions, and uninterpretable statistical matrices, creating the illusion of objective, authorless authority.
The distinction is not merely semantic; it is deeply political and ethical:
| Dimension | Intelligence Augmentation (IA) | Artificial Intelligence (AI) |
|---|---|---|
| Ontological Status | Vital prosthesis & cognitive scaffold | Simulated autonomous agent |
| Locus of Agency | The embodied organism in practice | The disembodied algorithmic pipeline |
| Locus of Meaning | Grounded in the human form of life | Projected onto statistical syntax |
| Accountability | Explicitly located in designers & operators | Diffused into the "autonomous system" |
| Normative Role | Amplifies human capacity to set new norms | Imposes static distributions from historical data |
Conclusion: Reclaiming the Normative Cut
Technical artifacts are neither innocent neutral instruments nor unstoppable metaphysical leviathans. They are crystallized human choices, material epistemic incisions, and bodily extensions of vital intent.
Overcoming technological fatalism does not mean smashing machines or retreating into bucolic illusions. It requires dismantling the myth of autonomy and demanding that every technical apparatus display its human lineage. Whenever an algorithmic system or megatechnical infrastructure presents its outputs as the inexorable mandate of objective reality, the critical task of Epistechnology is to ask:
What vital norms were selected? What friction was suppressed? And which human architect chose to hide behind the machine?