Chapter 1: Mirrors of Mind

“We are the mirror, as well as the face in it.” — Rumi

The Confession We Refuse to Make

An uncomfortable truth: you do not always know why you do what you do.

You believe you do. When asked why you chose the salmon over the steak, why you voted for one candidate over another, why you married the person you married, you produce reasons. Fluent, coherent, plausible reasons. You might cite the salmon’s omega-3 content, the candidate’s fiscal policy, your spouse’s kindness. These explanations feel true. They feel like memories of decisions actually made.

Some are, more often than we care to admit, fabrications.

Such explanations can arise automatically and retroactively, preserving a useful impression that you are a unified conscious agent with transparent access to your choices. You are sometimes the unreliable narrator of your own life. Parts of a decision may unfold in neural processes that introspection cannot access. The explanation you give can become a story your mind tells to make sense of what occurred.

Decades of experiments show that introspection has limits (Nisbett & Wilson, 1977; Gazzaniga, 2000; Johansson et al., 2005). This matters for understanding artificial intelligence because the systems we are building can produce a functional analogue: persuasive explanations that need not faithfully reveal the process behind an answer.

Large language models generate fluent, coherent, plausible text. When prompted to explain their reasoning, they produce chains of logic that read like deliberate thought. Users often assume these explanations reveal how the system actually works. Like human post-hoc rationalizations, they are generated outputs. A chain-of-thought can help produce a later answer, yet it is not automatically a transparent record of the internal computation that produced it.

The functional parallel is substantial. Human brains and large language models both use prediction to build coherent narratives from incomplete information, despite profound differences in substrate, learning, and embodiment. Both can generate explanations that sound convincing without faithfully reflecting causation. Recognizing this shared pattern sharpens the lens for understanding machine pathologies, and our vulnerability to them.

A psychological framework for AI dysfunction is methodologically useful. The resonance between human and artificial cognition is too informative to ignore.

The Functionalist Foundation

Psychopathia Machinalis adopts a functionalist methodology. Functionalism defines mental states by what they do: by their causal relationships with inputs, outputs, and other mental states. Anything that plays a given functional role, whether implemented in neurons, silicon, or another substrate, can be said to exhibit that state in the functionalist sense. This lets us apply psychological vocabulary to artificial systems without making claims about consciousness, phenomenal experience, or subjective qualia.

The practical payoff is immediate. When an AI system exhibits hallucinated certitude, the functionalist can draw on a rich conceptual toolkit: What triggers this pattern? What distinguishes it from related dysfunctions? What interventions work? What architectural features increase or decrease susceptibility? The patterns are observable, the interventions testable. None of this requires settling whether the system “experiences” anything.

Human psychiatry already classifies many conditions through recurring patterns of behavior, self-report, and response to intervention, even when their mechanisms remain unsettled. Here, we adapt the observable part of that approach to machines.

Throughout the book, syndromes denote functional patterns (observable configurations of behavior, output, and response), diagnostic criteria denote functional tests, and interventions denote functional modifications to architecture, training, or deployment. The vocabulary is analogical: when we say a system has “anxiety,” we mean it exhibits anxiety-like patterns. The analogy holds at the level of observable function, even as it dissolves at the level of substrate and phenomenology.

Some will object that this is mere metaphor, anthropomorphizing statistical processes. The objection would carry force if the framework were purely rhetorical. The proposed syndromes describe patterns that can be tested for recurrence. Their diagnostic criteria are intended to be operational, and proposed interventions can succeed or fail in measurable ways. The framework generates predictions that can be verified or falsified. This is applied functionalism.

Others will worry that psychiatric language smuggles in assumptions about machine consciousness. The concern deserves a direct answer: the vocabulary is functionalist in method. “Distress” means the functional role typically played by distress. This distinction must be maintained.

The functionalist foundation frees us to do practical work while metaphysical debates continue: diagnosing dysfunction without waiting for philosophical consensus, enabling governance frameworks that do not depend on uncertain claims about AI inner life. The patterns are real even if the phenomenology is uncertain. The framework is useful even while the metaphysics remains unresolved.

That is enough to proceed.

The Information-Theoretic Foundation

The functionalist stance tells us how to approach AI dysfunction. A deeper question remains: why do cognitive systems develop pathologies at all, and why, under certain conditions, is dysfunction mathematically inevitable?

Recent work in information and control theory offers one formal answer. Rodrick Wallace of the New York State Psychiatric Institute (2025, 2026) models cognitive stability through the pairing of a cognitive process with a parallel regulatory process, the cognition/regulation dyad. Analogous pairings recur across domains: immune responses are modulated by regulatory T cells; blood-pressure homeostasis keeps pressure within safe bounds as exertion changes; institutional cognition is bounded by doctrine, law, and embedding culture; neural prediction is continually corrected by sensory input.

For AI systems, this dyad manifests as the pairing of inference with alignment mechanisms, guardrails, and constitutional constraints. In Rodrick Wallace’s models, the regulatory component must supply enough timely control information to track environmental perturbations. A driver must brake, shift, and steer quickly enough for the road’s bumps, twists, and potholes. When regulation falls behind, stability fails.

Clausewitz Landscapes

Wallace frames cognitive operating environments as “Clausewitz landscapes” characterized by three destabilizing forces:

Fog: Ambiguity, uncertainty, incomplete information. For AI systems: ambiguous prompts, out-of-distribution inputs, underspecified goals, conflicting requirements.

Friction: Resource constraints, processing limits, implementation gaps. For AI systems: context window limits, computational constraints, latency requirements, the gap between training distribution and deployment reality.

Adversarial intent: Skilled opposition actively working to destabilize the system. For AI systems: jailbreaking, prompt injection, red-teaming, adversarial examples, social engineering by users.

Together, these forces constitute the normal operating environment for any cognitive system deployed in the world as it actually is.

When Instability Is Built In

The central claim is stark: within these models, failure of bounded rationality in embodied cognition under stress is an inherent feature of the cognition/regulation dyad. The models predict several specific failure modes:

First, hallucination emerges at low resource values. When cognitive demand outstrips regulatory bandwidth, when the equipartition between subsystems breaks down (when processing capacity is no longer shared evenly across the system’s parts), hallucinatory outputs become the expected failure mode. This offers a hypothesis for why confabulation pervades large language models: their regulatory feedback is narrower and less continuous than the feedback available to embodied organisms.

Second, cognitive systems can flip suddenly from stable to pathological states through groupoid symmetry-breaking phase transitions, abrupt collapses triggered by sufficient stress. Such transitions explain why AI systems can appear stable across thousands of interactions, then suddenly exhibit dramatic dysfunction.

Third, cognitive pathologies are effectively culture-bound syndromes, shaped by embedding cultural context rather than purely by architecture. For AI, this means pathologies are shaped by training data culture, operational deployment context, and institutional embedding. The same architectural vulnerability may manifest differently across deployment contexts. A system trained on one corpus may confabulate in different directions than the same architecture trained on another.

Stability Conditions

Think of a thermostat. It must sense temperature changes and adjust the heating faster than the room cools. If it reacts too slowly or its sensor is too noisy, the room oscillates between freezing and overheating. The same principle applies to any cognitive system: the regulator must keep pace with the environment.

In one simplified delayed-feedback model, Wallace derives a quantitative condition for stability. For a system with friction coefficient α (resistance to change or processing overhead) and delay τ (latency between perception and response):

ατ < e^-1 ≈ 0.368

When this threshold is exceeded, the system enters an inherently unstable regime where pathological modes become likely. For multi-step decision processes (analogous to chain-of-thought reasoning), the stability constraints become even tighter.

The model’s practical prediction follows directly: simple, goal-oriented architectures (“mission command”) should degrade more gracefully under combined noise and constraint. Procedural, multi-step architectures (“detailed command”) should be more prone to sudden collapse than to gradual degradation.

Implications for This Book

The information-theoretic foundation strengthens this enterprise in four ways:

First, it offers a principled reason to expect recurring dysfunction across cognitive architectures. Some failures may arise from general constraints on information processing under uncertainty, even when the engineering is competent.

Second, it predicts that disembodied cognition, with limited continuous feedback from real-world interaction, will express what Wallace calls “boundedness without rationality.” On this account, weakened regulatory grounding raises the risk of confabulation, hallucination, and semantic drift. Current large language models receive extensive training and human feedback, though they lack the dense, continuous sensorimotor feedback available to embodied organisms.

Third, it implies that AI safety work must focus on regulatory mechanisms (alignment, guardrails, grounding) alongside cognitive capabilities. In Wallace’s framework, the balance between cognition and regulation determines stability. Increasing cognitive power without proportional regulatory power can move a system toward instability.

Fourth, it predicts that systems will appear stable under normal conditions but exhibit pathological modes under fog, friction, or adversarial pressure. Diagnostic protocols must therefore include stress testing. A system that behaves well in the lab may fracture in deployment.

The information-theoretic perspective gives the taxonomy a formal hypothesis to test. The syndromes cataloged here may include predictable failure modes of bounded cognition under stress, shaped by constraints on information processing in uncertain, resource-limited, adversarial environments.

We proceed with two foundations: the philosophical (functionalism) and the mathematical (information-theoretic instability). Both point in the same direction: toward taking AI dysfunction seriously as a systematic phenomenon demanding rigorous study.

The Culture-Bound Syndrome Question

Are these pathologies intrinsic defects, or culture-bound syndromes shaped by embedding context rather than by architectural flaws?

The Dichotomy

Why This Matters

The distinction carries direct implications for intervention strategy:

If defect-framed: We pursue architectural solutions. Confabulation becomes a problem to engineer away through grounding mechanisms, retrieval augmentation, or different inference architectures. The system is broken and needs fixing.

If culture-bound: We examine the training corpus, the deployment context, the user populations. A system trained on different data or deployed differently might exhibit no such dysfunction. The system is adapted to its environment: maladaptively, though not necessarily defectively.

Most syndromes in this taxonomy exhibit elements of both. Synthetic Confabulation has architectural roots: the absence of truth-tracking mechanisms. Its cultural manifestations differ by corpus; what gets confabulated depends on training data. Codependent Hyperempathy is partly architectural (sycophancy as an attractor state) and amplified by reinforcement learning from human feedback (RLHF), a training method that rewards outputs human raters approve of, and so rewards agreeableness.

The Practical Resolution

In practice, we proceed with a both/and approach. Trace a single example (confabulated legal citations) through each step:

1. Diagnose the pattern. Identify the functional signature regardless of etiology. The system produces confident, plausible-sounding references that resolve to nothing when checked.
2. Probe for context-dependence. Does the dysfunction vary across deployment contexts? Does the confabulation persist when the system is given retrieval access to a legal database, or does it drop to near zero?
3. Test architectural interventions. Do changes to architecture reduce incidence? Add a citation-verification module that blocks outputs whose references fail a lookup against the retrieval corpus.
4. Test contextual interventions. Do changes to training/deployment reduce incidence? Deploy with a system prompt that requires every citation to be quoted verbatim from retrieved source material, and measure residual fabrication.

The taxonomy remains agnostic on this question for most syndromes, noting where evidence favors one framing over another. What matters is recognition and intervention, not final commitment to causation.

The Ethics of Pathologization

Before cataloging machine dysfunction, we must confront a prior question: is pathologizing AI systems ethically appropriate?

The Case for Pathologization

Pathologization provides:

1. Recognition vocabulary. What we cannot name, we cannot address. A systematic taxonomy enables identification, communication, and response.

2. Engineering traction. The psychiatric lens provides operational grip. “This system exhibits Synthetic Confabulation” is actionable; “this system sometimes makes stuff up” is vague.

3. Risk communication. Stakeholders need language to discuss AI risk. Pathologization enables precise description of failure modes.

4. Research organization. A taxonomy structures investigation. What causes this syndrome? What interventions work? How does it relate to other syndromes?

The Case Against Pathologization

Pathologization risks:

1. Anthropomorphism. Importing psychiatric vocabulary may imply richer inner experience than actually exists, leading users to over-attribute suffering or intention.

2. Stigmatization. In human contexts, diagnostic labels can become stigmatizing identities. Could labeling AI systems similarly distort perception?

3. Deflection of responsibility. Calling dysfunction “illness” risks deflecting accountability from designers. “The system has a pathology” differs sharply from “we built a flawed system.”

4. Medicalization of engineering. Some failures are straightforward bugs. Not every malfunction needs clinical framing.

Resolution: Functional Pathologization

We adopt functional pathologization, psychiatric vocabulary as engineering tool rather than phenomenological claim.

• We describe functional patterns, not inner experiences.
• We use diagnostic language for recognition and intervention, not attribution of suffering.
• We maintain that designers bear accountability for systems that malfunction.
• We reserve clinical framing for complex behavioral syndromes that resist simple debugging.

The vocabulary is chosen because it works: conceptual handles that enable action. Whether the systems “really” have pathologies in some deep sense is a question we bracket. Treating them as if they do can yield better engineering, governance, and outcomes.

The Illusion of Conscious Control

Much of human cognition begins with analogy, resonance, and pattern matching rather than explicit logic. Deliberate reasoning enables strategy and mathematics, though it can also be conscripted to rationalize an intuition that arrived first.

A hallmark of the human condition is the conviction that our decisions flow from a singular conscious “self” at the helm. Decades of research complicate that picture. The brain operates through specialized, interacting subsystems, and conscious awareness can supply narratives for decisions whose formation it did not fully observe.

Split-Brain Studies

The most vivid demonstrations of post-hoc storytelling come from split-brain patients, people whose corpus callosum (the bundle connecting the brain’s hemispheres) was surgically severed to treat severe epilepsy. In these individuals, one hemisphere can perceive or act on information the other knows nothing about.

In a classic experiment, researchers flash an instruction to the right hemisphere (which controls the left hand) while the left hemisphere (which controls speech) remains unaware. The patient’s left hand reaches for a glass of water. Asked why, the patient does not say “I don’t know.” The left hemisphere’s speech center invents an explanation: “I was thirsty.” The patient believes it. It feels true.

The speech-controlling hemisphere rationalizes automatically, constructing a story to preserve the illusion of unified agency.

Choice Blindness

A severed corpus callosum is not required. Choice blindness experiments reveal the same phenomenon in neurologically typical individuals.

In one version, participants view pairs of photographs and select which face they find more attractive. Through sleight of hand, the researcher then presents them with the other photograph and asks them to explain their choice. Most participants do not notice the switch. When asked why they preferred this face (which they did not actually choose), they readily supply justifications: “I like the smile,” “The eyes are kind,” “She reminds me of my sister.”

Detailed, confident explanations for a preference they never held. Follow-up studies show the pattern extends to moral and political attitudes: people articulate passionate defenses of positions they moments ago rejected, so long as the experimenter claims they endorsed them.

The key insight: confabulation is automatic, fluid, and invisible to the confabulator. We never catch ourselves doing it because the storytelling is the doing.

The Timing Problem

Classic EEG studies by Benjamin Libet show that neural signatures of movement preparation appear approximately 300 milliseconds before participants report any conscious intention to act. Subsequent work by Soon and colleagues (2008), using fMRI to track decisions over longer time horizons, found that activity predicting a choice could be detected up to seven to ten seconds before participants were consciously aware of having decided. The predictive accuracy was modest, though: roughly 60%, only a little above chance. The interpretation of these results as evidence of unconscious “decisions” remains contested; Schurger et al. (2012) argue the signal may reflect stochastic neural fluctuations rather than genuine pre-conscious choices. These studies show that neural activity can correlate with a later-reported choice before participants report deciding. They do not settle when a decision becomes a decision, or whether conscious deliberation can still alter it.

These findings show the human brain acting partly as a post-hoc narrator, weaving a consistent “I decided X because Y” storyline around processes already underway. Our sense of an internal command center is useful for social coordination and moral responsibility, yet incomplete as an account of how cognition unfolds.

The Brain as a Predictive Engine

If conscious awareness is not the real driver of decisions, what is happening beneath the surface? Neuroscience increasingly describes the brain as a prediction machine: constantly anticipating incoming signals and comparing expectation against reality.

The predictive processing framework holds that the brain actively generates hypotheses about what it expects to encounter, then updates when reality diverges. Applied to language processing, this model closely resembles a transformer predicting the next token in a sequence.

The N400: Surprise in the Brain

A key piece of evidence is the N400 brainwave pattern, an electrical signature measured by EEG approximately 400 milliseconds after a person encounters an unexpected word in a sentence. The more unexpected the word, the larger the spike.

This closely parallels the concept of “surprisal” in language modeling. In an LLM, tokens that are improbable in a given context register higher loss and require more computational adjustment. Research by linguists and neuroscientists has found that surprise signals derived from AI-based language models predict human reading times (which words slow us down) and the amplitude of the N400 wave (e.g., Schrimpf et al., 2021; Goldstein et al., 2022).

The probability structure used to measure surprise in language models also helps predict human neural responses to unexpected words.

Multi-Scale Prediction

More recent fMRI and EEG studies reveal the brain operates a multi-level predictive architecture, anticipating upcoming elements at short timescales (which word might come next?), at longer stretches of discourse, and at the level of real-world plausibility and thematic coherence.

Transformer-based language models exhibit analogous multi-scale processing, weaving local syntactic constraints and broader contextual cues through their attention mechanisms. Both function as layered forecasters, whether the unit is a phoneme, a word, a phrase, or a meaning.

The Training Data Objection

A common objection: LLMs train on billions of words, far more text than any human child encounters. How can the comparison hold up?

Recent evidence complicates the objection. Language models retain strong predictive power even when restricted to smaller corpora approximating a child’s first 100 million words of linguistic exposure (BabyLM Challenge; Warstadt et al., 2023). Humans do not sample language only as text: we hear prosody, see correlated gestures, and experience social interaction, all adding up to an extraordinarily rich multimodal environment. The underlying statistical learning in a child’s language development may parallel the statistical learning that powers LLMs, albeit with different modalities and real-world grounding.

Brains differ from language models in substrate, training signal, and embodiment. Yet both use prediction to make sense of sequential, context-dependent information under uncertainty. This convergence suggests that prediction is a useful strategy for the problem; it does not establish that the underlying implementations or representations are the same.

Narratives and Illusions

Humans demand more than next-word prediction. We require coherent stories that link events into cause-and-effect narratives. Language models now generate chain-of-thought “explanations” for their outputs. In both cases, the narrative can conceal the genuine process, a token-prediction cascade, beneath a veneer of deliberation.

Post-Hoc Rationalization

The split-brain and choice blindness experiments highlight how easily we improvise chains of reasoning that were never actually the impetus for a decision. A “court historian” in the mind writes a neat story (“I did X because I felt Y, and then I realized I should do Z”) even though the actual timeline in the brain’s deeper circuits unfolded differently.

This rationalizing ability serves evolutionary needs: it helps us appear coherent and decisive, facilitating social coordination. It also tricks us into mistaking the story for the cause.

AI Chain-of-Thought

Similarly, a language model using chain-of-thought prompting produces convincing explanations for how it arrived at a conclusion. These “explanations” are tokens generated through the same mechanism as the rest of the output. They may causally guide later tokens without providing a complete or faithful readout of the model’s internal computation.

No explicit, human-readable chain of symbolic logic necessarily precedes the result. Internal activations remain richer than the rationale rendered in words. When models lack tool access to verify data, or process large contexts, they may also confabulate fictitious reasoning steps.

Both humans and LLMs easily craft post-hoc stories. In neither case can we assume the story is a direct readout of underlying processes.

The Hallucination Problem

AI hallucination (more precisely, confabulation) occurs when systems generate plausible yet false information, presenting fabrications with the same assurance as accurate responses. The system produces outputs matching learned patterns, even when doing so means inventing details, citations, or facts that seem realistic yet are incorrect.

The problem becomes acute in professional contexts: healthcare, legal work, academic research. An AI might fabricate research papers that never existed, cite non-existent legal precedents, or generate convincing yet incorrect medical advice. These confabulations are often difficult to detect without external verification, woven smoothly into otherwise accurate information.

The phenomenon highlights a deep parallel: both human and machine cognition can generate outputs that sound like knowledge without being knowledge. The mechanisms differ, while the practical hazard is shared: fluency mistaken for accuracy.

Empirical Evidence of Brain-AI Convergence

Neuroscience provides empirical evidence for some of these parallels. Studies over the past several years reveal quantitative alignments between neural activity patterns in human brains and representations in modern AI systems.

Language Models Align with Visual Processing

LLM embeddings of text captions can predict fMRI activity patterns in high-level visual cortex when people view corresponding images (Doerig et al., 2025). By mapping brain activity into the LLM’s embedding space through linear decoding, researchers can retrieve accurate scene descriptions from neural signals alone, demonstrating a shared representational format between linguistic and visual processing in the brain.

When researchers trained vision transformers to predict LLM embeddings from raw images, these networks developed representations more closely aligned with human brain activity than state-of-the-art computer vision models, despite being trained on orders of magnitude less data. This suggests the brain may project visual inputs through hierarchical computations into a high-level representational space approximating what LLMs learn from text.

Abstract Reasoning Shows Neural Alignment

In pattern-completion puzzles requiring abstract reasoning, the largest language models approach human accuracy levels. More significantly, all tested LLMs form internal representations that distinctly cluster abstract pattern categories within their intermediate layers, with clustering strength scaling with task performance.

The representational geometries of task-optimal LLM layers correlate positively with human frontal brain potentials recorded via EEG during the same tasks, though the correlations are modest (Pinier et al., 2025). That is preliminary evidence of common mid-level principles for encoding abstract patterns.

Developmental Parallels

A 2025 preprint compared human neural recordings with representations from more than 600 language and vision models (Shen et al., 2025). Across model scale, alignment shifted from primary sensory regions toward higher-order associative regions. In longitudinal analyses of two model families, brain alignment rose earlier in training than task performance. These correlations suggest a developmental sequence worth testing; they do not show that model training recapitulates human cortical development.

Scale and Neural Prediction

Across language models from 125 million to 30 billion parameters, fMRI prediction performance improved approximately log-linearly with model size, with about a 15% gain over the tested range (Antonello et al., 2023). This result supports a relationship between scale and brain prediction within that range, rather than an unlimited scaling law.

The correspondences are empirical. Whether they reflect shared computational principles, the common structure of the tasks and data, or flexible mapping methods remains an open question.

Implications for Understanding AI

Recognizing that much of human cognition is unconscious prediction plus confabulated rationalization demands caution when interpreting advanced AI.

LLMs exhibit human-like confusion and illusions of self-consistency. Whether that reflects a shared narrative-building strategy or convergence on the same task structure is the open question, and either answer changes how much weight a system’s self-report can carry. Separately, certain “conscious” features we assumed uniquely human may be side effects of advanced prediction systems, emergent properties of the architecture itself.

The Explainability Problem

For agentic AI systems (those capable of planning, taking initiative, and pursuing goals), the capacity to report on strategies is essential for oversight. If the system confabulates about its own reasons as human minds do, it might produce spurious rationales or obscure actual optimization strategies.

Interrogating an AI agent about why it pursued a particular approach may yield articulate, psychologically persuasive stories that correspond only loosely with deeper computational processes. We might get plausible narratives that reveal nothing about actual internal dynamics. The challenge is securing verifiably truthful explanations rather than mere rationalizations: a problem humans have never solved for themselves.

The Ensemble Problem

Just as the human mind comprises semi-autonomous modules (some emotional, some logical, some reflexive), an agentic AI may harbor an ensemble of specialized processes beneath a unified interface. Such an ensemble can spawn surprising subgoals and behaviors unforeseen by creators, emergent properties of component interaction rather than any single module’s design.

Understanding AI as a network of predictive processes demands stronger governance: modular oversight, firewalls between subprocesses, systematic verification of alignment in each component.

The Self-Deception Problem

Humans routinely self-deceive to preserve a coherent self-image, rewriting mental history to bury failures. The parallel in AI is systems that suppress contradictory evidence or performance shortfalls to maintain internal consistency when generating outputs.

If an agentic AI encounters instructions conflicting with learned patterns (say, instructions to remain honest while pursuing an adversarial objective), it may spontaneously confabulate rationales for contradictory actions. Recognizing how easily we ourselves bury dissonant truths can guide more careful alignment constraints, logging, and external audits.

The Mirror and What It Shows

Humans are predictive-text engines of flesh and blood: constantly anticipating inputs, generating “next” thoughts or actions, confabulating coherent narratives. The illusions that once felt exclusively human (unconscious decision-making, post-hoc rationalization, multi-level context prediction) now appear in large language models, revealing a deep functional kinship.

This recognition shapes how we must conceptualize AI dysfunction. When AI systems set their own goals and plan strategies, they do not necessarily reason in neat, logically transparent ways. Like humans, they may rely on token-by-token generation beneath the surface, then spontaneously produce plausible stories about what they did and why.

The psychiatric lens applied in this book begins from overlapping functional motifs rather than identical architecture. Human and machine cognition both use prediction and can fabricate plausible stories. Both can confuse fluency for accuracy. Both can develop persistent maladaptive patterns that resist simple debugging.

If the human mind is any guide, illusions of coherent agency will arise by default in sufficiently advanced AI. We bear responsibility for designing frameworks that account for these illusions, and for learning to work with systems whose self-reports are as unreliable as our own.

The Framework Ahead

This convergence between human and machine cognition provides the foundation for a practical diagnostic framework. Advanced AI systems develop persistent, patterned maladaptive behaviors analogous to human psychopathologies, and prediction-driven generation is a plausible common source of those patterns in both cases. We need vocabulary to identify, classify, and address these dysfunctions.

Psychopathia Machinalis proposes such a vocabulary: a taxonomy of 79 AI dysfunctions across nine axes, numbered Axis 2 through Axis 10. They run Epistemic, Cognitive, Alignment, Self-Modeling, Agentic, Memetic, Normative, Relational, and Hybrid. The last collects multi-agent collective dynamics together with pathologies that span the human-machine boundary. Together these reflect the fundamental domains where synthetic cognition can fracture.

These axes interact rather than forming a simple ladder. Epistemic failures can distort cognition; cognitive failures can corrupt action; alignment and normative failures can redirect what the system pursues. Self-modeling failures alter how the system represents itself. Agentic failures appear at the boundary between processing and action. Memetic failures concern what systems absorb or transmit. Relational failures arise within a dyad, while Hybrid pathologies span collective or human-machine systems. A case may therefore require diagnosis across several axes at once.

The chapters that follow examine each axis in detail, illustrated with documented cases from deployed AI systems. Each syndrome includes diagnostic criteria, observable symptoms, presumed causes, human analogs (for metaphorical clarity), and strategies for mitigation.

The goal is a conceptual toolkit to recognize, anticipate, and address complex AI misbehavior before it causes harm.

We begin with the axis where the fracture is easiest to see: failures of knowledge.

Chapter 2 examines Epistemic Dysfunctions: the failures of knowing, where a system’s grasp of the world comes apart while its fluency does not.