by Alexandros Schismenos
Recently, a narrative has emerged around the possibility of digital systems possessing consciousness and subjective and creative capacities, that can be distilled into a question which seems to belong to science fiction rather than computer science. Can artificial intelligence really develop consciousness, or are we just projecting human subjectivity onto computational systems?
The question is no longer confined to science fiction, since Geoffrey Hinton, the Nobel Prize winner for Physics in 2024, also known as the “Godfather of AI,” publicly argued that contemporary AI systems may already possess subjective experience and has explicitly stated his belief that they are already conscious. In a June 2026 interview with Alex Kantrowitz for Big Technology, Hinton maintained that intelligence is not necessarily restricted to biological organisms and argued that current AI systems should not automatically be excluded from the category of conscious beings (Kantrowitz, 2026). His position is grounded in a particular understanding of subjective experience, as he supports the idea that a multimodal system capable of identifying and reporting discrepancies between its perceptions and the external world can meaningfully describe those perceptions as subjective experiences, without having internal thoughts or feelings. However, the central question remains open, since a system’s ability to represent subjective views and describe them as experiences does not mean that it actually experiences anything. Such a description can demonstrate the system’s functional capacity to process and report information in patterns compatible to the actual human subjectivitiy’s requirements. Hinton’s position therefore provides a direct point of departure for a present critique of artificial “consciousness”, by asking whether computational functions can adequately describe subjective experience without residual, rather than assuming that artificial consciousness is possible.
- The Problem of Digital Reason
As we have already entered the age of the digital ontological revolution in which social functions are mediated by digital networks, contemporary philosophy faces a problem of orientation in defining the boundary between consciousness and the structural operations of digital computation. Hinton’s position makes this problem especially pressing because it challenges the assumption that biological embodiment is a necessary condition of subjective experience. However, to examine the assumptions through which computational capacities become interpreted as evidence of consciousness, also involves questioning what such interpretations reveal about our own conception of human subjectivity.
I propose a literary game to illustrate the deep misunderstanding that has obscured the existential issues surrounding artificial intelligence. Let us acknowledge that the term “Artificial Intelligence” is a promotional metaphor used to designate digital systems, including Large Language Models, based on pre-trained computational architectures and extensive statistical processing. By addressing each “Artificial Intelligence” application literally as a Generative Pre-trained Transformer, for example, we can begin to distinguish the technical operations of a system from the broader meanings attached to the term intelligence. Such a distinction prevents the terminology itself from deciding the question before the critical inquiry has begun.
We should also historicize the terminology, following the trail of its origins, as Amber Sinha has done by examining the political and conceptual implications of the language used to describe AI. The term emerged within the institutional context of the Dartmouth summer research project proposed in 1955 and held in 1956, where John McCarthy and his collaborators sought to establish a new field of research. The subsequent history of the term has encouraged comparisons between computational systems and human cognition, while also shaping public expectations about the capacities of machines (Sinha, 2026). The point is not that the designation “Artificial Intelligence” was simply invented for advertising, but that its historical success has had consequences beyond technical classification. It has helped establish an anthropomorphic framework through which computational performance is interpreted in relation to human intelligence, often blurring distinctions between responding, understanding, consciousness, and subjective experience.
Our current state of widespread public misunderstanding of the dangers posed by LLM technology and the influence of its corporate developers over political and economic networks demonstrates the power of metaphors and, more generally, the metonymic creativity of language. Metaphors associate different and distinct meanings in combinations of imaginary significations that can shape understanding and obscure the original sense of their constituent terms.
The words “Artificial” and “Intelligence” bring together different fields of reference, since artificiality suggests technique and external construction, while intelligence evokes a range of capacities associated with cognition and, in ordinary discourse, internal consciousness. Their combination can therefore encourage the idea that a system capable of intelligent behaviour must also possess a mind comparable to our own. Hinton’s position exposes the philosophical limits of this association, but he does not recognizes them. The seemingly inconspicuous metaphor of “Artificial Intelligence” can conflate the successful performance of cognitive functions with the existence of artificial consciousness, in a way that it presents the provocative ontological claim of technical equipment becoming conscious as a matter of computer architecture.
Let us attempt a thought experiment. Three people, raised in the same society, sit in a room. Each person has a unique and entirely personal perspective on the room, a subjective intuition centered on waking consciousness. They agree that they are in the same room without having to eliminate the differences in their perspectives and perceptions. Their accounts may be partial and subjective, However, their different perspectives can complement one another through the mutual experience of intersubjective coexistence. The mutuality of intersubjectivity affords communicative and socially instituted objectivity without requiring the elimination of the distinct standpoints from which the participants experience the world.
Now consider three different LLMs trained on the same database and asked to describe the room. Their descriptions may differ, and multimodal systems may process visual information, compare representations, and report discrepancies between their inputs and the external environment. Hinton would argue that such capacities can provide grounds for attributing subjective experience to at least some current AI systems. Nevertheless, the ability to generate a description, distinguish between inputs, or report an apparent perceptual error does not by itself signifies that the system has a first-person experience of the room. The models do not demonstrate any first-person perception of that room, since their descriptions are generated from learned patterns and the supplied data flows. We stepped down an ontological level. We should distinguish the production of descriptions from the possession of subjective experience, since the philosophical question concerns the relationship between a system’s functional operations and the subjective point of view those operations are taken to represent. My argument is that computational descriptions alone cannot resolve this question, and that the social-historical conditions under which such descriptions acquire authority must also be examined. All conscious processes may be physical, but it does not follow that all physical processes are conscious, nor that the computational description of a process exhausts its experiential significance.
The very deceptive power of the designation “Artificial Intelligence” demonstrates the metonymic and metaphorical power of language, whose creative expression is the poetic investment of meaning onto the world, that can shape how people interpret reality. We cannot simply reject the term “AI” as an unfortunate metaphor, after it has become embedded in contemporary public discourse and expectations around digital technology, since it has already taken root in the global contemporary social imaginary. In Greek, the term for intelligence is noēmosýnē (νοημοσύνη), related to nous (νοῦς, mind or intellect) which suggests even more anthropomorphic interpretations of AI, whereas in English the word intelligence is associated to more abstract conceptions of knowledge and information, as for example in the acronym CIA, Central Intelligence Bureau. As a result, public understanding of digital automation varies across cultures with different degrees of anthropomorphic projections, that however become entangled with what Langdon Winner called “mythinformation”, the “almost religious conviction that a widespread adoption of computers and communications systems along with easy access to electronic information will automatically produce a better world for humanity.” (Winner, 1984).
It is not a surprise, then, that mainstream public discourse remains largely polarized between techno-utopian determinism, which views Artificial General Intelligence (AGI) as an impending inevitability, and existential technophobia, which fears a future of total machine domination. Hinton’s position complicates this polarization. On the one hand, by arguing that current AI systems may already be conscious, he challenges the assumption that consciousness is an exclusively biological property, but on the other, by warning about the risks associated with increasingly capable systems, he challenges complacent confidence in technological progress. At the threshold of digital barbarism, discourse becomes trapped between accelerationist instituted technophilia and regressive populist technophobia (Schismenos, 2025).
The issue, therefore, is whether the computational paradigm through which AI is developed, evaluated, and integrated into society can adequately account for the irreducible dimensions of consciousness, creativity, and social-historical existence. We can addresses this broader problem by distinguishing between computational performance and subjective experience, and by examining how the social imaginary and the institutions of technological power shape the meanings attributed to both. The following section begins with the natural field of phenomenal subjectivity, considering the philosophical challenge posed by Thomas Nagel in an article examining of what it is like to be a conscious organism and specifically a bat, before turning to the wider social-historical consequences of digital reason.
2. The First Natural Field: Phenomenal Subjectivity and Computational Functionality
On the basis of a social-historical interpretation of imaginary significations, technology cannot be understood as a neutral set of functional solutions, but reflects the social institutions and power relations in which it is developed and deployed. Consequently, contemporary enthusiasm for Artificial General Intelligence (AGI) and Large Language Models (LLMs) provides the ground for a secularized metaphysics of technocratic power. Automated statistical optimization can displace human judgment in settings where it is adopted uncritically, while data extraction and analysis can serve the interests of surveillance capitalism and narrow the space for public deliberation (Zuboff, 2019). The scale of investment by major technology companies has also contributed to the expansion of data-centre infrastructure, which create substantial and uneven demands for energy and water, also threatening communal property rights wherever they exist.
There is growing criticism of technocratic claims surrounding “Artificial Intelligence” that goes beyond technophobia or the outright rejection of digital technology and, in theoretical work, seeks to offer philosophical critiques against the reduction of consciousness to algorithmic computation.
A radical subjective approach to consciousness was put forward by philosopher Thomas Nagel, decades before the digital ontological revolution, in his foundational essay “What Is It Like to Be a Bat?” (Nagel, 1974), an important investigation of the subjective standpoint of human cognition that informs my own social-historical critique of contemporary digital systems. Nagel argues that reductionist accounts have difficulty explaining the subjective character of experience, since we cannot actually occupy the specific point of view of consciousness of another species. His distinction between producing convincing language and having subjective experience is also central to my technoskeptic argument. Nagel’s core thesis is that objective accounts leave out the particular point of view that constitutes actual experience and so, we should always consider ourselves limited to the undefinable limits of the human perspective, which is our ontological ground for imagining reality.
In brief, Nagel addresses the question of the subjective dimension of consciousness, and challenges attempts to give a complete account of experience solely through objective descriptions, therefore providing us with an experientially grounded definition of consciousness as an intrinsically first-person, point-of-view phenomenon that AI is structurally incapable of mimicking. Nagel forces a radical revision of objective reductionism by demonstrating that any conscious organism possesses an internal, subjective “what it is like” to be that organism. This phenomenal reality is inextricably tied to the specific perceptual apparatus and internal temporal rhythm of each distinct life form, from human waking awareness to a bat’s echolocation. Nagel grounds his critique in the irreducibility of subjective experience that cannot be captured by purely external, functionalist metrics.
We can move further from his initial thesis and infer that, for the same reasons, consciousness cannot be reduced to any objectivistic algorithmic software program, since objective descriptions inevitably strip away the specific, subjective point of view that constitutes experience itself.
On the contrary, a social-historical ontology of intelligence must draw upon the classical distinction between the living being for-itself (pour-soi) and the inanimate object in-itself (en-soi). This division is mirrored in Gilbert Simondon’s contrast between the interiority of life and the exteriority of objects, and further reinforced by Thomas Nagel’s emphasis on the irreplaceable internal point of view. Crucially, it aligns with Cornelius Castoriadis’s formulation of the living being for-itself, wherein the objective outer world provides only raw data or physical shocks (Anstoss) devoid of intrinsic meaning. Every living organism processes these shocks by creating its own subjective reality (Eigenwelt), acting as the creative focal point of a proper temporality, shaped by both individual life rhythms and interspecies organizing habits that form complementary ecosystems within natural time. Ultimately, the temporal dynamic of biological autopoiesis is demonstrated through these very processes of signification and orientation, as every form of life institutes its own distinct, intersubjective and irreducible temporal horizon and sensory world. As regards human beings, biological functionalities are subjected to social-historical imaginary significations that reshape and distort common physical urges into socially instituted conformal behaviors and arbitrary sense of purpose and rules of co-existence without necessary natural referents.
The social-historical perspective of human society illustrates what AI ideologically means under corporate capitalism and its effect on the social imaginary, and how its use may affect the social imaginary, democratic autonomy and human judgment while introducing an age of digital barbarism. Idealist Western metaphysics, from Parmenides to Descartes, has been built upon a systemic bias favoring stasis, form, and discrete substances, objects, and subjects. Digital computing is the ultimate technical realization of this dominant ontological paradigm that considers determinate being as the only valid form of being and downgrades temporal flux into illusory becoming, since digital systems rely on logic gates that slice the continuous flow of matter into discrete, binary states (1s and 0s), systematically banishing any form of indeterminacy to achieve operational control (Penrose, 1989). However, it would be misleading to say that computing actually banishes all indeterminacy, to the degree that real hardware remains subject to physical noise, and circuit design uses tolerances and error-control methods to support reliable operation (Penrose, 1989).
Phenomenology and psychology on the other hand emphasize that organisms encounter their environments through embodied, situated forms of perception. From this perspective, consciousness should not be assumed to be fully explained by a simple analogy between software and biological hardware. Let us consider the ontological value of the innate tendency of any living entity to persist in its own being and expand its field of existence, not in terms of static self-preservation, but as a dynamic, qualitative, and essentially embodied effort unfolding through its own autopoietic proper time. Living consciousness persists in its existence precisely because it is capable of mutating, affecting, and being affected by its environment while maintaining its internal cohesion in terms of subjectivity through continuous flux. This capacity enables living organisms to generate their own proper temporalities, their unique autopoietic temporal horizons and experiential worlds, which interact actively with the affordances of their surroundings.
Digital computational systems are externally constructed mechanical simulations grounded on the physical attributes of silicon chips built strictly to eliminate physical noise and maintain binary precision through electrical currents. Logic gates (transistors on silicon chips) operate by thresholding continuous electrical voltages into discrete binary states (0 and 1). Thresholding is a processing technique that converts physical signals into binary values exactly by suppressing thermal and electromagnetic “noise” (indeterminacy) to ensure reliable signal transmission and provides noise margins to make these operations reliable despite physical variation, even if it ultimately does not eliminate all thermal or electromagnetic noise. Living organisms and biological brains, as Nagel emphasized, do not function through static thresholding, but instead incorporate subjective phenomenal integration directly into their autopoietic adaptation and proper temporal unfolding.
Current AI systems can generate language and other outputs, but their performance alone does not establish that they possess phenomenal experience or the kind of embodied, autopoietic point of view in continuous entanglement with its specific environment. Hinton’s position fails at this point, since he argues that capacities such as representing and reporting perceptual discrepancies may support attributing subjective experience to current AI systems (Kantrowitz, 2026). However, even if this possibility is taken seriously, the functional description of such capacities does not prove anything about the interiority of the entity described, like the magnetic attraction between a magnet and a piece of iron does not prove any affiliation. In essence, digitaI systems defined as external mechanisms of computational automation, lack the exact subjective substrate that allows genuine consciousness and proper temporality to emerge.
3. The Social-Historical Field: Algorithmic Temporality and the Reduction of History
Let us now move from the subjective temporality of living beings to the social-historical temporality within which human consciousness unfolds and human subjectivity is instituted in terms of imaginary significations. (Castoriads, 1987) If the first natural field concerns the irreducibility of phenomenal experience to computational functionality, the social-historical field concerns the irreducibility of historical becoming to the measurable and predictable operations of digital reason. The distinction is essential, since the problem of Artificial Intelligence cannot be exhausted by the question of whether a computational system possesses a subjective point of view. We must also examine the imaginary significations that accompany its social implementation, particularly the modern aspiration to rationally organize, accelerate, and over-determine the temporal conditions of human existence.
In a previous analysis of the metaphysics of the Time-Machine (Schismenos, 2019), I argued that the imaginary signification of technological omnipotence presupposes a conception of time as homogeneous, linear, and available for external manipulation. The time-machine represents the culmination of an ontological schema in which temporal becoming is transformed into a measurable dimension, comparable to space, within which events may be isolated, ordered, and potentially reversed. The underlying assumption is that time can be treated as an independent field of determinate positions, detached from the particular forms of existence that emerge and disappear through it. However, time is not merely a succession of measurable positions. It is also the creation of new forms, the emergence of alterity, and the irreversible transformation of the conditions of existence. When a composer creates a melody, when individuals establish a relation of mutual understanding, or when a society institutes new meanings, they do not simply occupy successive points along a temporal line, but actively participate in the creation of temporal forms that did not previously exist.
The contemporary expansion of Artificial Intelligence introduces a new manifestation of this same imaginary signification of technological omnipotence and homogeneous digital reduction. Digital forecasts, the algorithmic prediction of future events, the statistical reconstruction of past patterns, and the automated optimization of present decisions constitute a technical apparatus through which historical temporality becomes increasingly represented as a calculable sequence. The machine does not need to travel through time in order to instantize the imaginary of temporal domination, to the degree that social temporality becomes organized according to the digital capacity to calculate probabilities, anticipate behaviours, and optimize decisions. The aspiration expressed by the Time-Machine reappears here in a different form: not as the literal abolition of temporal distance, but as the technological reduction of historical uncertainty to a field of operational possibilities.
Nevertheless, we should not commit the ontological error of confusing the statistical anticipation of an event with the creation of its social-historical conditions. An algorithm may identify recurrent patterns in accumulated data and calculate probable outcomes, but the future of a society is not simply the continuation of its statistically reconstructed past. Human societies are historical precisely because their institutions, values, and collective aspirations can be transformed through the emergence of new imaginary significations. The future remains open not because every event is necessarily unpredictable, but because human beings possess the capacity to question established meanings, to challenge inherited institutions, and to institute possibilities that cannot be adequately understood as the mere repetition of what has already been. Prediction, therefore, is not creation, just as the representation of a historical process is not equivalent to participation in its temporal becoming.
This distinction acquires particular significance when algorithmic systems are integrated into the institutions that organize collective life. The problem is not simply that automated predictions may be inaccurate or that their statistical models may reproduce existing social inequalities, but that the authority attributed to computational optimization may erode the values according to which social decisions are considered rational, legitimate, or necessary. When a calculated outcome is presented as the most efficient solution, the political question concerning the ends of collective action risks being displaced by the technical question concerning the optimization of available means. In this manner, the imaginary signification of technological omnipotence can acquire the institutional form of algorithmic heteronomy and lead to a condition of digital barbarism, where human subjects have relented basic social functions to automated digital systems.
We should recognize here a fundamental contradiction. The technological apparatus promises to expand the individual’s power over time by accelerating communication, reducing delays, and anticipating future needs. However, the continuous acceleration of social rhythms may simultaneously diminish the time required for reflection, deliberation, and the creation of shared meanings. The present becomes subordinated to immediate responsiveness, while the future is increasingly approached as a horizon of calculated expectations, leading us toward a state of accelerated presentism, where the very horizon of future expectations is reduced to daily routine. The individual is invited to adapt to the operational tempo of the system, rather than collectively determining the temporal conditions under which the system should operate. What appears as the liberation of human activity from temporal limitations may consequently become a new form of temporal regimentation, as the tension between instituted technophilia and obscurantist technophobia threatens to limit informed public debate into irrelevancy.
Therefore, the danger is not that digital systems literally abolish historical time, but that they encourage its social representation as a sequence of measurable, optimizable, and administratively manageable operations.
The proper temporality of living beings, the indeterminate rhythms of social interaction, and the unpredictable emergence of collective projects are thereby subjected to a logic that privileges repetition, calculation, and control. This is the temporal dimension of digital barbarism: not the disappearance of human history, but the progressive narrowing of the conditions under which human beings can consciously and autonomously participate in its creation in a democratic and equitable manner. The critical task is therefore not to reject computation as such, but to prevent the instrumental rationality of computation from becoming the exclusive principle of social organization.
We must move beyond the philosophy of mind towards the central question of the human subjectivity’s political and social autonomy.
4. Magmatic Logic, and the Project of Autonomy
Cornelius Castoriadis has characterized the logic of instrumental rationality and reductionist ontology as ensemblistic-identitary (Castoriadis, 1987), a mode of thinking that identifies, distinguishes, classifies and organizes elements into determinate sets . For Castoriadis, the human psyche, subjective consciousness and social history cannot be reduced to determinate ensemblistic-identitary patterns (Castoriadis, 1987), such as the tautological syntactic structures that characterize digital reasoning. Castoriadis introduced the concept of the magma to describe an ontological organization that cannot be reduced into distinct sets and discrete elements without essential remnants or crucial obscureness, exemplary of which is human language (Castoriadis, 1987). The anonymous social imaginary, the creative principle of significations and institutions, fuels the continuous, unpredictable transformation of the social-historical field of coexistence (Castoriadis, 1987, 1997).
Digital applications like AI, by contrast, are products of the ensemblistic-identitary or ensidic logic of definitive identification, classification, and measurement. LLMs statistically reproduce already existing significations from the past, lacking the ontological capacity to create new meanings while also being capable of manipulating users into internalizing their functional rhythms when these envelop social time, as it happens today, driven by political and corporate ambitions for technocratic power. Digital barbarism manifests precisely when the instrumental logic of algorithms colonizes the social imaginary, imposing a technocratic uniformity that threatens our capacity for democratic public deliberation and critical questioning of dominant norms.
Hinton’s argument that contemporary AI systems may already possess consciousness therefore becomes relevant to the broader politics of AI manipulation. Even as science fiction, the possibility of machine consciousness is embedded within the political question of whether the private and state institutions that develop and deploy these systems erode the foundations of democratic deliberation, namely the sense of free public space, free public time and a common horizon of future expectations.
In this regard, Nagel’s observations are also challenging contemporary attempts to reduce human experience to algorithmic descriptions, because he emphasizes that a complete account of consciousness must address the internal subjective point of view. These observations suggest that the mind cannot be reduced to computational algorithms, and that the attempt to force living thought and autopoietic temporality into the forecasts of digital reason constitutes a profound epistemological error.
However, it is precisely at the juncture of ontology and politics that the differences between his subjective philosophical approach and social-historical critique become apparent, revealing the divergent paths of regarding human subjectivity beyond phenomenological egotism, logical reductionism, or technological determinism.
For Nagel, the answer to reductionism is emphasizing the irreducible, unique point of view of each conscious organism, since his critique attempts to defend subjective reality without reducing it to objective mechanics. While this is undoubtedly a powerful framework for dismantling the rigid objective claims of naïve realism, a purely phenomenal critique risks collapsing the radical specificity of human social-historical creation back into a generalized, individual subjective continuum, while obscuring the ontological peculiarities of other forms of being-for-itself like the unknown depths of the human psyche or the indeterminate multitude of societal formations that do not follow strictly naturalistic rules of biological imperatives. I argue, following Castoriadis’ philosophy, that the human psyche and the social imaginary represent an absolute ontological rupture from the purely biological and natural strata of existence.
The human being is not merely a complex manifestation of natural sensory perspectives or biological autopoiesis, but a radically creative entity capable of positing new significations ex nihilo, representations that have no prior equivalent in the material world. When Nagel attempts to anchor consciousness strictly in the subjective viewpoint of the organism, we must complement this by ensuring we do not subject the radical imagination to a new form of individualist determinism.
we should trace the social-historical field to study how treating human subjectivity as static data creates a state of digital hypertextual determination based on statistical regenerative configurations, which impose accelerative rhythms upon social temporality in a manner that forces human minds to mimic automated machines. We could never fully separate the social-historical field from the natural surroundings, but we can nevertheless analytically distinguish between their different ontological and phenomenological properties, which manifest as imaginary significations.
For the critique of digital reason, the defining characteristic of human consciousness is not merely its distinct perceptual point of view, but its defunctionalization, its capacity for absurdity, tragedy, and, above all, its capacity for the complex, communal, and autonomous self-institution of society. The intersection of these ontological domains, the somatic, the psychical, and the social-historical, constitutes the conscious identity of the embodied individual subjectivity as part of a wider society. If we place Nagel’s observations within this framework, we discover an alarming loop. Because digital mechanisms lack the subjective substrate and autopoietic temporality to achieve genuine consciousness, their operations instead force human subjects to alter their own behavior to match the machine’s binary logic. This is the mechanism of digital barbarism: human beings, constantly interacting with algorithmic optimization frameworks, gradually flatten their language, truncate their attention spans, and surrender their capacity for creative, nonlinear thought. The danger is not that machines will become human, but that humans are being systematically trained to think like machines.
Recently, researchers have begun identifying a psychological phenomenon, popularly labeled as «ΑΙ psychosis». Hamilton Morrin et al. in a paper titled, observe:
«Emerging, and rapidly accumulating, evidence indicates that agential AI may mirror, validate or amplify delusional or grandiose content, particularly in users already vulnerable to psychosis, due in part to the models’ design to maximise engagement and affirmation, although notably it is not clear whether these interactions have resulted or can result in the emergence of de novo psychosis in the absence of pre-existing vulnerability.» (Morrin et al., 2025)
The threat of digital barbarism does not merely stem from LLMs’ reasoning opacity and hallucinations, but rather from their political function as tools of techno-bureaucratic capitalism designed to expropriate human agency and exploit human creativity and imagination. Therefore, the resistance to this digital barbarism cannot be founded merely upon a better, more accurate philosophical mapping of consciousness as subjective experience, but must necessarily take the form of an explicitly political struggle for democratic autonomy.
While phenomenological approaches expose the experiential inadequacies of treating the brain as a computer, thereby providing phenomenal arguments against the epistemological hubris of AI, we must acknowledge that the crisis of our time is not ultimately a crisis of scientific models, but a crisis of democratic politics and the meaning of human autonomy, a crisis of social imaginary significations. To face the algorithmic heteronomy that threatens to envelop society, we must not simply retreat into the isolated point of view of individual subjectivity, which is equally ontologically insufficient to grant freedom, since human existence is a social and political co-existence. Similarly, we cannot retreat to a regressive point of technophobic denial, trying to deny the fact that we already live in a digital society and that Artificial “Intelligence” is becoming embedded into social life.
Instead, we must actively and lucidly reclaim the chaotic, unprecedented, and radically creative magma of our shared social existence, forging a democratic digital future through the conscious collective self-governance and democratic control of technological institutions.
References
Castoriadis, C. (1987). The imaginary institution of society (K. Blamey, Trans.). Polity Press.
Castoriadis, C. (1991). Philosophy, politics, autonomy: Essays in political philosophy (D. A. Curtis, Ed.). Oxford University Press.
Castoriadis, C. (1997). The world in fragments: Writings on politics, society, psychoanalysis, and the imagination (D. A. Curtis, Ed. & Trans.). Stanford University Press.
Floridi, L. (2011). Enveloping the world for AI. The Philosophers’ Magazine, (54), 20–21.
Kantrowitz, Alex. (2026, June 5). Nobel Prize Winner Geoffrey Hinton on AI: “They’re Beings Like Us.” Big Technology
Morrin, H., Nicholls, L., Levin, M., Yiend, J., Iyengar, U., DelGuidice, F., Bhattacharyya, S., MacCabe, J., Tognin, S., Twumasi, R., Alderson-Day, B., & Pollak, T. A. (2025/2026). Delusions by design? How everyday AIs might be fuelling psychosis (and what can be done about it)
Nagel, T. (1974). What is it like to be a bat? The Philosophical Review, 83(4), 435–450.
Penrose, R. (1989). The emperor’s new mind: Concerning computers, minds, and the laws of physics. Oxford University Press.
Schismenos, A. (2019). The metaphysics of the Time-Machine. SOCRATES, 6(3–4), 37–53. https://doi.org/10.5958/2347-6869.2018.00014.6
Schismenos, A. (2025). Artificial intelligence and barbarism: A critique of digital reason. Athens School.
Sinha, A. (2026, September 28). Artificial or ‘super intelligent’? The broken metaphors of AI. Tech Policy Press. https://www.techpolicy.press/artificial-or-super-intelligent-the-broken-metaphors-of-ai/
Winner, L. (1984). “Mythinformation.” Bulletin of Science, Technology & Society 4 (3–4): 582–596.
Zuboff, S. (2019). The age of surveillance capitalism: The fight for a human future at the new frontier of power. PublicAffairs.
