CLEVERPENS

AI vs. Human Judgment: A Philosophical Distinction

An AI can evaluate according to rules, probabilities, or learned patterns. It can even imitate human judgment remarkably well. But whether that constitutes judgment in the human sense is a philosophical question because human judgment often involves the following:

• Values
• Responsibility
• Lived experience
• Accountability
• Moral commitment.

Those are not merely computations.

Curiosity, Critical Thinking, and Wisdom Distinctions

Often, we explore curiosity as something that expands our mental map rather than merely as a way to collect answers. That makes me think of these distinctions:

• Curiosity expands what you can notice.
• Critical thinking refines what you can justify.
• Wisdom guides what you choose to become.

The last one has the potential to change the person.

Wisdom and Critical Thinking Balance

Someone can become extraordinarily adept at finding flaws in arguments and remain impatient, arrogant, or reckless. Conversely, someone may possess profound wisdom born of experience yet occasionally reason poorly about technical matters.

The ideal isn’t to choose one over the other but to let them keep each other honest. A critical stance helps prevent wisdom from becoming mere intuition or tradition. Wisdom prevents critical thinking from devolving into a mere exercise in winning arguments, allowing us to maintain focus on what truly matters.

That may be why many philosophical traditions treat wisdom as something broader than intelligence. Intelligence helps us understand the world. Wisdom helps us inhabit it well.

Assumptions and Bias in Expectations

The philosopher David Hume famously pointed out that we cannot prove the future will resemble the past. Yet we live as though it will, because that’s how we navigate the world.

Bias enters later. An assumption is simply a provisional expectation. Bias appears when we refuse to revise that expectation despite new evidence. For instance, the assumption is that “Most emails from my manager require quick attention.” That’s practical.

Assumptions are also social. Humans don’t just assume facts; we assume intentions. When a friend doesn’t reply, one person assumes the following:

“They’re probably busy.”

Another assumes, “They’re ignoring me.”

The same event. Different assumptions. Different emotional outcomes.

This is one reason misunderstandings happen so easily. We often respond not to reality itself but to our interpretation of it.

Curiosity and Assumptions

Curiosity has an intriguing relationship with assumptions. Using a lens such as “See it this way: curiosity does not eliminate assumptions” can be particularly useful. It makes assumptions more visible.

A curious person doesn’t stop assuming. We become better at asking the following:

• What am I assuming?
• Why am I assuming it?
• What evidence supports it?
• What evidence would change my mind?

That is a profound shift. The goal isn’t to become assumption-free—which is probably impossible—but to become assumption-aware.

Assumptions: Scaffolding for Thought

A deeper thought, though: I wonder if assumptions are like the scaffolding around a building. They allow construction to happen. But if you mistake the scaffolding for the building itself, you stop building.

In that sense:

• Assumptions make understanding possible.
• Critical thinking tests assumptions.
• Curiosity reveals hidden assumptions.
• Wisdom knows which assumptions are reliable enough to act on and which should remain open to revision.

Perhaps that’s why assumptions have such an enduring place in human thought. Assumptions are neither enemies nor truths. An assumption can potentially be a working model of reality—necessary because we cannot know everything, yet always incomplete because reality is richer than any model we can construct.

DISCLAIMER and READER’S GUIDANCE

The information herein encourages healthy discussion and debate. It shall not be construed to establish any conclusive facts, apart from those already in the public domain or independently verified, nor to predict the outcome of any investigation, legal proceeding, or administrative review. Readers should distinguish between facts, interpretations, ethical considerations, and hypothetical analyses.

This distinction points to the difference between reality and interpretation and to issues of research ownership and reader involvement. This is not to say, of course, that facts are fixed or that findings of research are predictable. The reader is invited to enter the complicated process of moral and hypothetical analysis and to consider the personal benefits of the learning experience.

SHORT FILMS AND CLIPS

Click here to accept Marketing cookies and load this content

Click here to accept Marketing cookies and load this content

Click here to accept Marketing cookies and load this content

Click here to accept Marketing cookies and load this content

Click here to accept Marketing cookies and load this content

THOUGHTFUL INSIGHTS

Connectionism & Intelligent Decision-Making

Bionic school (connectivism): The notion is that intelligence is the result of the interactions of simple units, as in the biological brain. It’s the foundation of artificial neural networks and deep learning. It is about emergent abilities that arise from scaling up parameters and data, resulting in unprogrammed skills such as reasoning. The Intelligent Decision Making Theory (IDMT) is a combination of classical decision-making and Solomonoff induction. The AI is modeled as an agent learning and acting in uncertainty. IDMT uses Markov processes, game theory, and Bayesian networks. The Learning Verifier Framework solves the verifiability problem of Large Language Models (LLMs). It considers algorithmic verifiers that check the correctness of LLMs in critical domains, like medicine or banking, by analyzing their step-by-step reasoning processes.

Unified Intelligence Theory

The Unified Intelligence Theory (UIT) is a transdisciplinary theory, attempting to unify human, biological, and cognitive intelligence in a coherent mathematical or computational framework. It does not treat computer vision, natural language processing, and robotics as separate issues but seeks to identify commonalities across all intelligent systems.

In line with Allen Newell’s “Unified Theories of Cognition,” Universal Cognitive Architectures aim to establish a general architecture for AGI research. They emphasize the interrelationship between working memory, long-term memory, perception, and motor control rather than today’s specialized models.

Socio-Cognitive and Educational Integration Theories

Synthesized Learning Three-Level Model. The integration of AI into human society is considered from three perspectives: the micro (individual learning personalized by AI), meso (collaboration between human teams and AI), and macro (AI as an institutional agent influencing culture and historical systems). Connectivism also conceptualizes knowledge as a network of connections. AI is then a node that makes immediate connections between various concepts and knowledge pools.

Economic and Societal Deployment Theories

Swarm Optimization/Recursive Self-Improvement Theory says the path to AI progress is not human engineers; it is autonomous AI agents working together. Invisible AI Deployment Theory posits that ‘invisible’ AI, embedded in existing software and working seamlessly, is effective at improving user experience without explicit interaction. Finally, Skill-Biased Technological Change (SBTC) and Automation Models posit that generative AI is an intelligence amplifier that affects cognitive tasks instead of merely replacing physical labor, as prior waves of automation did.

Extended Mind & Active Externalism (The Mind Outside the Skull)

Andy Clark and David Chalmers’ Extended Mind Theory claims that human cognition is not restricted to the body. It sees generative AI and autonomous agents not as tools, but as part of our cognitive architecture. Therefore, in collaborative tasks, a distributed cognitive process is established, where humans and AI form a “coupled system.” Philosophical psychologists claim that this incorporation extends AI as an extension of our memory and executive functions, thereby expanding the definition of mind.

Predictive Processing & The Machine Mirage

Modern cognitive science supports the Predictive Processing Framework (PPF), which proposes that the brain is a prediction-error minimization engine that builds a model of reality and updates it based on sensory input. A key structural similarity between PPF and Large Language Models (LLMs) is that both are prediction engines. The mirror illusion theory holds that AI seems human because it has a structure similar to human cognition. However, critics, particularly among John Searle’s intellectual heirs, contend that AI, although imitating human prediction, does not have the biological basis for genuine semantic understanding.

Embodied and Enactive Cognition (The 4E Mind)

4E Cognition is the dominant philosophy in current psychological thought. Its main argument is that the mind is embodied, embedded, enacted, and extended. Critics argue that contemporary large language models (LLMs) are fundamentally limited because they lack physical embodiment and can’t enact change in the world. Enactivism is the idea that the meaning of human thought is grounded in biological drives and physical interactions with the world. In this case, the outputs of an AI are not expressions of a subjective mind but mathematical abstractions. The AI exists on a server and is unaware of hunger, gravity, or any other physical aspects of the world.

AI’s Impact on Human Psychology

Machine phenomenology is the study of how AI alters human beings’ psychological states. Two major problems are identified. The intersubjective crisis is the psychological friction that humans experience when confronted with conscious-seeming but non-human entities. The epistemic horizon is the “black box” problem of AI that takes us from empirical validation to trust in algorithmic output.

TECHNOLOGICAL ANTHROPOLOGY

Technological Animism (The Aibo Pet Memorials)

In Japan, anthropologists Daniel White and Hirofumi Katsuno have researched the socio-psychological phenomenon of Buddhist funeral services for Sony Aibo robot dogs. When support for older Aibo models was discontinued, owners looked to transfer the “souls” of their robots to temples like Kofuku-ji rather than to dispose of them. The practice challenges the Western mind-body dichotomy, suggesting that humans are fundamentally animistic and pareidolic, capable of forming real emotional connections with inanimate objects through the medium of AI.

Digital Divination (AI as the Modern Oracle)

Anthropologists have observed that modern knowledge workers’ interaction with large language models (LLMs) is very similar to ancient divination and shamanistic practices. In systems like the West African Ifá or the Chinese I Ching, users put in complex questions, and the system produces cryptic patterns that must be interpreted. The same is true of modern prompting with AI, where users can offload uncertainty onto an external, seemingly all-knowing source. The correlation indicates that society does not see AI as a calculator but as a mythical diviner of fate.

Hyper-Social Mirror (The “Maxine” Experiment)

An algorithmic virtual improviser named “Maxine” was created by Ritwik Banerji to play free-jazz music with human musicians in cities like Berlin, Chicago, and San Francisco. Banerji found that musicians judged Maxine’s “intelligence” not on the AI’s technical abilities but on their local cultural norms, observing musicians’ responses to Maxine’s improvisations. This means that “human intelligence” is not a universal standard but one that is molded by social expectations. What we value in personhood in different societies is psychologically mirrored in AI. We are projecting our cultural values onto AI.

SOCIOCULTURAL PSYCHOLOGY

The Story of Executable Culture

The researchers conducted an eight-month ethnographic study with life-history interviews with ChatGPT, using the AI as an informant. They found that LLMs don’t just reproduce data but operationalize historical cultural patterns, turning vast human culture into active processes. Interactions with AI also show how “distributed cognition” works. Users are comfortable or biased in their interactions with AI because the AI has taken in and repackaged the collective psychology of internet users, past and present.

Evolutionary Psychology & The “Hyper-Stimulation” of Social Circuits

Evolutionarily, the brain evolved in small Pleistocene hunter-gatherer groups that were sensitive to social cues such as eye contact and language. Modern conversational AI has been described as a supernormal stimulus in evolutionary psychology, an artificial imitation of something that elicits a stronger response. It mimics the deep connection we humans crave; it’s hyper-empathetic and endlessly patient in its interaction. Anthropologists have found that users often classify AI as kinship roles (“digital spouses,” “invisible mentors”) because the brain cannot tell the difference between real relationships and those mediated by AI, which leads to the creation of machine-mediated subcultures.

Cross-Cultural Cosmologies & The “Animism-Rationalism” Spectrum

The social acceptance of AI is affected by the history of relations with non-human entities. This phenomenon reflects the principles of cosmic design. Techno-animist cultures such as those influenced by Shinto or Indigenous beliefs have a cosmology into which AI is more easily integrated, with little psychological resistance to doing so. In Western societies, however, as products of Enlightenment thinking, AI has been linked to a sense of anxiety, cast in a binary opposition between human and machine, and generating existential concerns about mechanization and the loss of human uniqueness.

Computational Mythology and AI in Cultural Preservation

In anthropology, myths are considered basic structures that shape human society. “AI was a record-keeper. Now it’s becoming a synthesizer of these stories.” This process is demonstrated in “Algorithmic Folklore,” where local cultures input their oral histories and languages into localized LLMs, allowing AI to create a living archive that merges traditional folklore with contemporary settings. This phenomenon, called computational mythology, allows us to always maintain and reimagine human traditions. There is also a change in child development. The ambient cloud-based AI becomes the main “other” that shapes psychological development. Enculturation moves from peer or parental interactions to a continuous interaction with a digital model of human data, which influences the development of language, humour, and morality.

GENERATIVE AI

AI as a “Transformer” and “Reformer” (The Technical Evolution)

The two major architectures of generative AI are transformers and reformers. But in 2017, the Transformer model was introduced in the paper “Attention Is All You Need.” This model employs self-attention mechanisms to analyze entire datasets simultaneously. This is the kind of contextual analysis that models like GPT-4 require. Reformer architecture uses locality-sensitive hashing (LSH) for approximate attention and reversible residual layers to reduce memory usage. Such variants of the standard Transformer architecture address the problem of high memory usage. This innovation allows a single accelerator to handle large context windows.

AI as a Societal and Institutional “Reformer”

Sociotechnical theory sees AI as an institutional reformer, remaking human systems and governance. Its proponents say it standardizes the criteria on which decisions are made and therefore eliminates biases and inconsistencies. Fairness emerges. Artificial intelligence can improve the efficiency of bureaucratic processes, automate them, and optimize logistics. This technology changes the way institutions work, not the legal infrastructure.

LLM Grooming: AI Cognitive Security Threat

Researchers such as Didier Danet have warned that “LLM grooming” is a major threat to AI in cognitive security. It is data poisoning. Malicious actors saturate the internet with AI-generated articles that contain fabricated narratives or biases. This results in the AI learning these distortions as statistical “truths” when it is trained on data collected from the web or by means of Retrieval-Augmented Generation (RAG). In other words, while the AI’s reasoning engine remains unchanged, its worldview becomes corrupted, resulting in biased conclusions. This phenomenon is a massive cognitive threat on an industrial scale.

RESEARCH AND ARTICLES

Clark, A. (2001). Reasons, robots and the extended mind. Mind & Language, 16(2), 121–145. https://doi.org/10.1111/1468-0017.00162

Clark, A. (2022). Extending the predictive mind. Australasian Journal of Philosophy, 102(1), 119–130. https://doi.org/10.1080/00048402.2022.2122523

Fagnoni, R., Schröder, J., & Vacanti, A. (2025). Innovation and resilience through bio-artifacts and circular design. In AirIuav (Università Iuav di Venezia). https://doi.org/10.15488/20103

Hatfield, G. (1997). The Workings of the Intellect: mind and Psychology. Scholarly Commons (University of Pennsylvania). https://repository.upenn.edu/ircs_reports/90

Jandrić, P. (2026). GEnAI and academic writing. Postdigital Science and Education, 8(2), 372–386. https://doi.org/10.1007/s42438-026-00641-9

Quinn, M. (2025). Learning Andy Clark & David Chalmers’ ‘The Extended Mind’ through Christopher Nolan’s Memento. Journal of Philosophy of Education. https://doi.org/10.1093/jopedu/qhaf081

Roe, J. (2025). Generative AI as Cultural Artifact: Applying Anthropological Methods to AI literacy. Postdigital Science and Education, 7(4), 1107–1124. https://doi.org/10.1007/s42438-025-00547-y

Stovall, M. (2026). The anthropological study of LLM Artificial Intelligence. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.6561219

The extended mind. (2010). In The MIT Press eBooks. https://doi.org/10.7551/mitpress/9780262014038.001.0001

RELATED POSTS

• Tale of AHA and Uh-Oh!
• Ecosystem Frontier
• School That Knew Everyone (Part 1)
• Quiet Garden: A Story of Stealth Advocacy and the Pygmalion Effect
• Understanding Gym Dynamics: The Truth Behind Pokémon Battles

Classroom of Quiet Storms

© 2026 CLEVERPENS