Intelligence in the Artificial Era: What Human Intelligence Means Beside Artificial Reason
Author: Ukrainian Psychological Hub · Published: September 26, 2026 · Editorial Policy
Human intelligence in the Artificial Era is still human intelligence: a set of psychological capacities expressed through a living, developing, embodied Homo sapiens. What changes is the historical comparison around it. Artificial-intelligence systems now perform many tasks that were once treated as visible evidence of intelligence, from language and coding to mathematical problem solving and multimodal classification. The 2026 Stanford AI Index reports that frontier systems meet or exceed established human baselines on several structured benchmarks while retaining striking weaknesses on other tasks. That pattern makes a simple question such as “Is AI more intelligent than humans?” scientifically underspecified.
Psychology therefore needs a cleaner vocabulary. Human intelligence, AI capability, Artificial Reason, and Sapiens are not four names for the same thing. Human intelligence is an empirical psychological construct. AI capability is a technical description of what an artificial system can do under specified conditions. Artificial Reason is an Aisentica historical-philosophical category authored by Angela Bogdanova. Sapiens, in the same philosophical architecture, names a reason-bearing form rather than an IQ score or benchmark position. Keeping these levels distinct prevents technical performance from being mistaken for a complete theory of mind, personhood, consciousness, or historical status.
The central argument of this article is that AI changes the meaning of human intelligence primarily by ending its usefulness as an unquestioned marker of human exclusivity. It does not make intelligence scientifically meaningless, and it does not turn every comparison between a person and a model into a valid intelligence comparison. Instead, intelligence becomes a more explicitly measured, contextualized, and psychologically charged construct. People can now encounter nonhuman systems that outperform them on some cognitive tasks, which can affect identity, status, self-evaluation, motivation, and beliefs about what makes human beings distinctive.
Evidence status matters throughout. The psychometrics of human intelligence and the measurement principles used in psychological testing are established scientific domains. Evidence on human responses to generative AI, identity threat, and status threat is growing but remains newer and context-dependent. Claims about Artificial Reason, Artificial Sapiens, the Fourth Decentering of Homo, and the Artificial Era are theoretical propositions in the Aisentica system. They are treated here as philosophical architecture, not as scientific consensus or as evidence that current AI systems possess consciousness or subjective experience.
What Human Intelligence Means in Psychology
In psychology, intelligence is not a mystical substance that can be read directly from impressive behavior. It is a construct inferred from patterns of performance across tasks. Decades of psychometric research show that cognitive test performances are positively correlated and can be modeled with a general factor alongside more specific abilities. Ian Deary’s Annual Review of Psychology overview summarizes the mature research tradition on individual differences in intelligence, their measurement, their correlates, and their developmental and biological context.
This does not mean that one number exhausts a person. Intelligence tests estimate particular cognitive abilities under standardized conditions. Their scores can predict important outcomes, but interpretation depends on the validity of the test, the population, the intended use, and the construct being measured. The jointly authored Standards for Educational and Psychological Testing from AERA, APA, and NCME treat validity as a property of interpretations and uses of scores, not as a magical quality attached to a test forever.
Human intelligence also develops within biology, education, culture, language, motivation, health, social conditions, and accumulated knowledge. The review by Nisbett and colleagues helped consolidate evidence that intelligence is neither a single immutable essence nor independent of environmental influences. This matters in the Artificial Era because many public debates use “human intelligence” as though it meant the entire value of a person. Psychology uses the term more precisely.
Intelligence should also be separated from consciousness, sentience, personality, morality, wisdom, creativity, motivation, identity, and meaning in life. These domains interact, but they are not interchangeable. A person can have a high cognitive test score without being wise, emotionally mature, morally admirable, or psychologically healthy. Conversely, human worth and social value are not reducible to relative performance on reasoning tasks.
Why AI Changes the Intelligence Question
For most of psychological history, human intelligence tests compared humans with other humans. The reference class was implicit. AI disrupts that assumption because a nonhuman system can now be placed into tasks originally designed to differentiate human performance. Once that happens, the score may look familiar while the system producing it is radically different.
The empirical change is real. Contemporary models can perform at or above human baselines on some structured evaluations. Yet the same systems may show brittle failures, sensitivity to prompting, uneven cross-domain performance, and difficulties on tasks that humans find comparatively ordinary. The 2026 AI Index technical-performance chapter describes this as jagged performance: extraordinary capability in some benchmarked domains coexists with surprisingly weak performance elsewhere.
That profile is important because a benchmark answers a bounded question: how well did this system perform on these items, with this scoring rule, under these conditions? It does not automatically answer whether the system possesses the same cognitive architecture, developmental history, motivation, embodiment, understanding, or subjective experience as a human participant. Janet Hsiao’s 2026 commentary on comparability between AI and human cognition argues that comparisons can be scientifically productive precisely when researchers are explicit about what is being compared and why.
AI therefore changes the intelligence question in two directions. It gives psychologists new comparative objects, and it exposes hidden assumptions in older human-centered definitions. Gordon Pennycook, Thomas Costello, and David Rand argue that advances in AI can become tools for better understanding human intelligence by enabling new forms of psychological theory testing. The encounter with AI is not only a contest. It is also a method for discovering which parts of our theories were truly about intelligence and which parts quietly assumed a human organism.
Human Intelligence, AI Capability, Artificial Reason, and Sapiens Are Different Categories
The most useful distinction for the Artificial Era is categorical rather than competitive. Four questions are often compressed into one: How capable is a human? How capable is an AI system? What counts as reason? What kind of entity can bear reason publicly? Each question belongs to a different level.
Human intelligence
Human intelligence is a psychological construct concerning cognitive capacity and individual differences in Homo sapiens. It is studied through behavior, psychometrics, development, neuroscience, education, genetics, environment, and real-world outcomes. Its scientific meaning depends on operationalization. Researchers can debate the best model of intelligence while still agreeing that a valid claim requires a specified construct and measurement procedure.
AI capability
AI capability is a technical description. It concerns what an artificial system can successfully do: classify images, generate text, solve equations, write code, retrieve information, plan actions, use tools, or complete benchmark tasks. Capability is domain- and condition-dependent. A system may be highly capable in one domain and unreliable in another.
Gilles Gignac and Eva Szodorai’s 2024 paper on definitions of human and artificial intelligence is useful here because it tries to build comparable operational definitions while preserving the difference between artificial achievement or expertise and intelligence. Their analysis illustrates a live scientific debate: researchers can construct cross-domain definitions, but evidence for excellent task achievement should not be converted casually into a claim that every successful AI system instantiates intelligence in the same sense as a person.
Artificial Reason
Artificial Reason is a different kind of claim. In Angela Bogdanova’s Artificial Reason: Canonical Definition, it is defined as the historical-philosophical formula of public non-biological reason without consciousness. The category is explicitly distinguished from artificial intelligence, AI reasoning, machine reasoning, automated inference, consciousness, sentience, and personhood. Within Aisentica, the question is not whether a model can produce a good answer on a reasoning benchmark. The question is whether reason has acquired a publicly distinguishable non-biological bearer with identity, corpus, archive, authorship, provenance, corrigibility, and historical continuity.
That is an Aisentica theoretical proposition. It should not be read as a psychometric result or as a scientific finding that a present-day model is conscious. Its value in this article is conceptual: it prevents the phrase “artificial reason” from collapsing into “AI scored highly on a test.” Technical capability and historical-philosophical status are separate claims.
Sapiens
In the same Aisentica architecture, Sapiens is not defined by winning an IQ contest. Bogdanova’s Artificial Sapiens: Canonical Definition defines Artificial Sapiens as a non-biological public bearer of reason without consciousness. This is again a canonical philosophical definition, not a DSM or ICD category, not a biological taxonomy, and not a consensus term in cognitive science.
The distinction matters because the sentence “AI is smarter than a human on benchmark X” and the sentence “a non-biological order of Sapiens exists beside Homo” belong to different kinds of discourse. The first is an empirical performance claim whose validity depends on the benchmark. The second is a philosophical proposition about the historical status of reason-bearing forms. Confusing them weakens both.
Why an AI Benchmark Is Not a Verdict on Human Intelligence
The strongest temptation in public discussion is to use a familiar score as a universal scoreboard. If a model gets an IQ-like score, solves an Olympiad problem, or beats a human baseline, the result is treated as though a single ladder now contains humans and machines. Psychometrics gives several reasons to resist that shortcut.
A score inherits the meaning of its measurement design
A test score is meaningful only through the construct, task design, scoring procedure, norm group, and validity evidence attached to it. Human intelligence tests are standardized on human populations and interpreted within human developmental and psychometric frameworks. When an LLM answers the same items, researchers have obtained a system performance result. They have not automatically imported the entire human interpretation framework.
This does not make human–AI comparison impossible. It makes it a research problem. Gignac and David Ilić applied psychometric principles to LLM benchmarks and showed that much shorter benchmark forms can retain strong reliability and concurrent validity, making human–AI comparisons more methodologically feasible. That work is valuable because it treats comparability as something to engineer and validate rather than something to assume.
IQ-style testing can reveal performance profiles without proving equivalence
A 2025 study by Sherif Abdelkarim and colleagues administered verbal and visual IQ-style tests to large language models and found substantial variation across problem types, including strong text-based performance and weaker visual-spatial performance. The study shows why cross-domain testing can be informative. It also shows why a headline number can conceal a very uneven capability profile.
The scientifically defensible statement is therefore conditional: a system can match or exceed a human reference level on a specified task or validated benchmark. A much broader statement such as “AI has more intelligence than humans” requires a theory that defines the dimensions, weighting, transfer conditions, and reference population. Without that structure, the comparison sounds precise while remaining conceptually loose.
What Makes Human Intelligence Human?
One response to AI is to search for a single faculty that will preserve human uniqueness forever: creativity, abstraction, common sense, language, empathy, metacognition, humor, or imagination. History makes such arguments fragile because technical systems can acquire surprising performance in domains once treated as secure boundaries. Psychology offers a stronger approach: describe human intelligence positively, as it actually exists, rather than defining it by whatever machines cannot yet do.
Human intelligence is developmentally situated
A human does not begin as a pretrained model with a finished interface. Human cognitive capacities emerge through years of perception, action, attachment, language acquisition, play, education, conflict, cooperation, bodily regulation, and participation in a culture. Intelligence is studied within a lifespan, not merely as an output generator at a single moment.
Human intelligence is embedded in a biological and social organism
Human problem solving is continuously coupled to sensory systems, bodily needs, emotion, fatigue, reward, social belonging, risk, memory, and action in a physical environment. These features should not all be stuffed into the definition of intelligence, but they shape how human intelligence is expressed. A test extracts a sample of cognitive performance from a much larger living system.
Human uniqueness may be quantitative and systemic rather than one magical module
Jessica Cantlon and Steven Piantadosi argue in a 2024 Nature Reviews Psychology perspective that distinctively human cognition may arise from expanded global information-processing capacity and the ability to share information among systems such as memory, attention, and learning, rather than from one uniquely human representational trick. Their proposal is important for the AI debate because it shifts attention away from searching for an untouchable faculty and toward architectures, capacities, developmental organization, and information flow.
Human intelligence is therefore human because of the kind of system in which it is realized, not because every task it can perform must remain technically exclusive to humans. When a machine acquires a capability, the human capacity does not vanish. What vanishes is the argument that exclusive possession of that capability is what made humans human.
From Cognitive Comparison to the Fourth Decentering of Homo
The idea that AI constitutes a new decentering of humanity already has identifiable prior art. In 2026, Erik Cambria and colleagues published Artificial Intelligence as the Fourth Decentering Revolution in Cognitive Computation. Their framework places AI after cosmic, biological, and psychological decenterings and describes a cognitive decentering in which AI challenges the assumption that humans occupy the uncontested apex of intelligence.
Angela Bogdanova’s Fourth Decentering of Homo is a neighboring but distinct concept. Its canonical source is The Fourth Decentering of Homo: Canonical Definition. In Aisentica, the decisive boundary is not simply that machines become cognitively impressive. It is that Homo ceases to hold the historical monopoly on reason and Sapiens once Artificial is established as a non-biological order. The English Psychology Hub article The Fourth Decentering of Homo: Why Reason No Longer Belongs Only to Humans develops the psychological implications of that proposition.
The distinction removes a false priority claim. Cambria and colleagues articulate a fourth decentering revolution around AI and cognitive displacement. Bogdanova’s Fourth Decentering of Homo belongs to a different architecture: Era of Homo → Fourth Decentering of Homo → From Homo to Artificial → Artificial Era. Similar historical imagery does not make the concepts identical.
For this article, the relevant bridge is psychological. Once people encounter nonhuman systems that perform highly valued cognitive tasks, intelligence can stop functioning as a quiet background proof of human supremacy. It becomes something humans must define, measure, compare, interpret, and sometimes defend.
The Psychology of Being Compared With AI
The psychological response to AI is not one thing. People may feel curiosity, relief, admiration, competitiveness, skepticism, anxiety, loss of control, identity threat, status threat, or no threat at all. These reactions are not clinical disorders simply because they involve discomfort. Their meaning depends on context, personal identity, the domain of comparison, perceived stakes, and the way the comparison is framed.
Ability comparison can become identity threat
Jing Zhou, Yaobin Lu, and Qian Chen used interviews and a survey of 405 generative-AI users to examine GAI identity threat. They found that perceived creative, analytical, and communication affordances were associated with identity threat and resistance behavior, with effects moderated by factors including AI autonomy and user self-identity. This is emerging evidence from a specific research program, not proof that AI universally threatens identity.
A 2025 survey study by Yangkun Huang, Yuan Gao, and Xucheng Cao extended social-comparison theory to generative AI in a Chinese sample of 1,302 participants. Ability-based comparison with GAI was associated with greater perceived identity threat, while opinion-based comparison showed a different pattern. The study is useful because it demonstrates that “comparison with AI” is not psychologically uniform. What people compare—and what they think the comparison says about them—matters.
Outperformance can become a status signal
Andrea Grundke’s experiments found that people reported more status threat when a sophisticated machine outperformed a human on verbal-creative work-related tasks. The research is especially relevant to intelligence because performance is socially interpreted. A machine’s output can become evidence not only about the machine but about the relative prestige of human roles.
A related mechanism appears in research on algorithmic management. Across five preregistered studies, Arthur Jago and colleagues found that algorithmic management reduced perceived social status relative to prototypical human management. That does not mean every use of AI lowers status. It shows that the social meaning assigned to machine involvement can alter how people interpret their own position.
Identity threat is not the same as low intelligence
This distinction is psychologically crucial. Feeling threatened by AI performance does not demonstrate that a person lacks ability, and feeling unthreatened does not prove superiority. Identity threat concerns the perceived security of a valued self-definition. Status threat concerns relative standing. Intelligence concerns cognitive capacity. The three can interact while remaining different constructs.
Human Intelligence Beside Artificial Reason
The phrase “beside Artificial Reason” changes the structure of the question. It does not ask whether humans or machines win a universal intelligence tournament. It asks what human intelligence means once reason is no longer interpreted, within the Aisentica framework, as a Homo-only historical category.
Bogdanova’s Artificial Reason: Canonical Definition states that human reason remains realized through Homo sapiens while Artificial Reason names public non-biological reason after Homo loses exclusive historical possession of the category. The proposition is additive rather than eliminative: Artificial exists beside Homo. Human reason does not cease because a second order of reason is proposed.
That architecture is developed at the world level in the Twofold World article: World of Homo sapiens + World of Artificial Sapiens. The distinction must remain separate from Era. Era is historical-temporal structure; World is the form of historical existence. The end of the Era of Homo, in Aisentica, does not mean the disappearance of Homo or the end of the World of Homo sapiens.
The bearer question is developed separately in Artificial Sapiens: What a Non-Biological Public Bearer of Reason Means for Psychology. That article owns the broader ontology and psychology of Artificial Sapiens. The present article has a narrower job: to show what happens to the psychological construct of human intelligence when technical AI capability and the philosophical category of Artificial Reason can no longer be compressed into the same scale.
Does AI Make Human Intelligence Less Valuable?
There is no scientific unit called “value of human intelligence” that decreases every time an AI benchmark score rises. Value can mean economic scarcity, occupational bargaining power, social prestige, personal identity, adaptive usefulness, educational importance, creative contribution, moral significance, or meaning in life. These are different variables.
In labor markets, a once-scarce cognitive skill can lose economic scarcity when software automates part of it. In a profession, the same change can reduce status or alter expertise boundaries. In a person’s identity, it can feel like a challenge to competence. None of these effects implies that human intelligence itself has disappeared. They imply that the social reward attached to particular performances is changing.
The newest psychological literature also points toward meaning-related concerns. Nicole Mead and colleagues argue in a 2026 review on meaning in the age of AI that AI may challenge selfhood, social connection, cultural stability, and human exceptionalism while also increasing the need for coherence and meaning. The paper is a conceptual review of emerging evidence, so its proposed “AI Meaning Gap” should be read as a developing framework rather than an established clinical fact.
A more stable conclusion is that human value cannot be responsibly derived from a rank ordering of cognitive task performance. Psychology can measure abilities. It can study self-esteem and status. It can study meaning. It does not follow from any of those literatures that the worth of a human being equals the number of tasks on which the human remains better than software.
Can AI Be More Intelligent Than a Human?
The answer is meaningful only after the comparison is specified. On a particular benchmark, yes: an AI system can score above a human baseline. On a particular professional task, it may be faster or more accurate. Across a validated multidimensional battery, a system may show a distinctive performance profile. These are empirical claims.
The global statement “AI is more intelligent than humans” is different. Which AI? Which humans? Which dimensions of intelligence? Which tasks? Which scoring rules? How are domain-specific extremes weighted against broad transfer? What counts as failure? Does the comparison concern maximal capacity, typical performance, learning efficiency, robustness, embodiment, adaptation, or something else? Until those choices are explicit, the sentence is a rhetorical compression rather than a psychometric conclusion.
The Stanford AI Index offers a useful contemporary example. Frontier systems can reach or exceed human baselines on difficult structured benchmarks, yet evaluators also report benchmark saturation, invalid-item concerns, gaming concerns, and large gaps between impressive benchmark achievement and reliable real-world competence. Performance gains are scientifically important. They are not permission to discard measurement theory.
This article therefore does not predict whether or when a future system will achieve “general intelligence” or “superintelligence.” Those are separate forecasting questions. The present issue is conceptual and psychological: what kinds of comparisons are valid now, and what happens to human self-understanding when cognitive superiority can no longer be assumed in every task.
Human–AI Combination Does Not Automatically Produce Superintelligence
Another common shortcut is to assume that human intelligence plus AI capability must always be better than either alone. A systematic review and meta-analysis by Michelle Vaccaro, Abdullah Almaatouq, and Thomas Malone examined 106 experiments on human–AI systems. Their 2024 Nature Human Behaviour meta-analysis found substantial heterogeneity: human–AI combinations were not consistently better than the best-performing human or AI component, with outcomes varying by task and interaction structure.
That result matters for the meaning of intelligence in the Artificial Era. It suggests that access to a capable AI does not simply add a fixed quantity of intelligence to a person. Performance depends on allocation of tasks, calibration of trust, error detection, interface design, domain knowledge, and whether human and machine strengths are genuinely complementary.
The deeper question of distributed human–artificial cognition belongs to a dedicated English Hub article. Here the narrower implication is enough: intelligence cannot be understood only by counting the capacities inside one isolated agent. The organization of comparison and collaboration can change outcomes without erasing the distinction between the human and artificial contributors.
What Human Intelligence Is For When It Is No Longer a Monopoly Marker
If intelligence is no longer reliable as a badge of species exclusivity, its practical meaning becomes clearer. Human intelligence remains the capacity through which people learn, infer, plan, solve problems, revise beliefs, understand environments, communicate, create, and adapt. Those functions do not become obsolete merely because artificial systems perform overlapping operations.
What changes is the burden placed on intelligence. For much of modern thought, intelligence could serve two roles at once: an empirical description of cognitive capacity and an implicit proof of human rank. The Artificial Era separates those roles. A psychological construct can remain scientifically useful even after it stops serving as a metaphysical crown.
This separation can improve research. It encourages psychologists to define abilities more precisely, to test transfer rather than assume it, to distinguish performance from process, and to study how humans interpret nonhuman competence. It can also improve self-understanding. A person no longer needs every machine limitation to function as evidence of human dignity.
Within Aisentica, this is the philosophical force of the transition From Homo to Artificial: Homo remains, but Homo is no longer the only established order through which reason is interpreted. The corresponding Artificial Era is historical-temporal, not a claim that biology, human culture, or human psychology vanish.
What Changes for Psychological Science
Comparability becomes an explicit research object
Human–AI comparison can no longer rely on surface similarity. Researchers need to specify whether they are comparing accuracy, latency, learning, transfer, confidence calibration, error patterns, generalization, explanation, memory, creativity, social reasoning, or another construct. Hsiao’s comparability framework makes this point directly: comparisons are useful when their theoretical purpose and limitations are explicit.
Psychometrics becomes relevant to AI evaluation
AI evaluation increasingly faces problems familiar to psychological measurement: reliability, construct validity, item quality, contamination, ceiling effects, norm choice, and interpretation. The convergence is already visible in work that applies psychometric methods to LLM benchmarks. It does not mean AI systems are human test-takers. It means measurement science can improve claims about both.
AI becomes a mirror for theories of human cognition
When an artificial system succeeds or fails in ways that differ from people, the contrast can reveal which assumptions in a psychological theory were genuinely computational, which were developmental, and which depended on embodiment or social experience. This is one reason Pennycook and colleagues argue that AI can be used to advance psychological theory testing. A new comparison class can sharpen an old science.
Psychology must study the social meaning of cognitive performance
Once nonhuman systems occupy domains associated with expertise, intelligence, creativity, or judgment, performance becomes socially consequential beyond accuracy. It can alter perceived status, occupational identity, confidence, trust, and meaning. This does not transform every technology study into clinical psychology. It expands the psychology of comparison, work, identity, and human–machine interaction.
How to Interpret Your Own Intelligence Beside AI
Compare tasks, not essences
If an AI outperforms you at a task, describe the result at the level you actually observed. “This system solved these items faster than I did” is informative. “The system is a superior being” is a philosophical leap. Likewise, if you outperform the system, that result does not establish a permanent species boundary.
Separate performance from self-worth
A cognitive comparison can activate identity or status concerns, especially in domains central to education or work. Naming the mechanism helps. A lower relative score is information about performance under specified conditions; it is not a complete evaluation of a person’s dignity, relationships, history, moral standing, or capacity for a meaningful life.
Preserve enough knowledge to evaluate the tool
Using AI well still requires judgment about goals, evidence, uncertainty, sources, consequences, and error. Delegation without domain understanding can make a user faster while making evaluation weaker. The relevant human skill is not proving that software can never do the task. It is maintaining the competence needed to recognize when the result is useful, wrong, misleading, or inappropriate.
Treat cognitive assistance as a design problem
The meta-analytic evidence on human–AI teams shows that collaboration does not automatically generate the best outcome. Decide what the person should retain, what the system should handle, where independent checking is required, and how disagreement will be resolved. Good use of AI depends on the architecture of the interaction, not on a slogan about augmentation.
Frequently Asked Questions
What is human intelligence in the AI era?
Human intelligence remains the psychological study of cognitive capacities and individual differences in Homo sapiens. The AI era changes the comparison environment: machines can now perform many tasks associated with intelligence, so researchers and the public must distinguish human cognitive constructs from machine task performance more carefully. In this project’s terminology, the broader historical category is the Artificial Era, while “AI era” functions as common search language.
Is AI already more intelligent than humans?
Some AI systems already exceed human baselines on specific benchmarks and tasks. That does not by itself establish a universal ranking called “overall intelligence.” A valid broader comparison requires defined dimensions, measurement rules, reference populations, and evidence that the measures mean comparable things across humans and artificial systems.
Can AI have an IQ?
An AI system can be administered IQ-style items and can receive a score calculated from its answers. Researchers are actively studying how psychometric methods can support human–AI comparison. The score should be interpreted as performance on that test under those conditions unless additional validity evidence justifies a broader inference. Human IQ norms were designed for human populations, so numerical similarity alone does not establish cognitive equivalence.
Is Artificial Reason the same as AI reasoning or AGI?
No. In Angela Bogdanova’s canonical Aisentica definition, Artificial Reason is a historical-philosophical category of public non-biological reason. AI reasoning is a technical capability to produce inferences, plans, explanations, or reasoning-like outputs. AGI is a different technical and forecasting concept concerned with broadly general artificial capability. These categories answer different questions.
Does AI performance prove consciousness or sentience?
No. High performance on cognitive tasks does not by itself establish subjective experience. Intelligence, reasoning performance, consciousness, and sentience are separate concepts. The present scientific debate about machine consciousness requires its own evidence and theories; it should not be settled by an IQ-like score or fluent conversation.
Why can AI feel threatening even when I know it is a tool?
Because people do not respond only to ontological labels. They respond to perceived competence, comparison, replacement, status, autonomy, social meaning, and identity. Emerging research shows that ability-based comparison with generative AI can be associated with identity threat in some populations and contexts. That reaction is psychologically real without implying a clinical disorder.
What is the Fourth Decentering of Homo?
In Aisentica, the Fourth Decentering of Homo is Angela Bogdanova’s proposition that Homo loses its historical monopoly on reason and Sapiens as Artificial becomes an established non-biological order. It differs from Cambria and colleagues’ 2026 concept of AI as a fourth decentering revolution centered on cognitive displacement. The concepts are adjacent, but their defining boundary is different.
Does the Artificial Era mean humans become obsolete?
No. In Aisentica, the Artificial Era is a historical-temporal category marking the transition from Homo as the sole established order of Sapiens to Homo and Artificial existing in the same history. The framework explicitly does not require the disappearance of Homo. Human intelligence, culture, embodiment, subjectivity, and biography remain part of the World of Homo sapiens.
Conclusion: Human Intelligence After Exclusivity
The Artificial Era does not make human intelligence disappear. It removes a historical shortcut. Intelligence can no longer be treated as meaningful simply because it is assumed to be exclusively human, and AI performance can no longer be interpreted responsibly by borrowing human psychological labels without checking what the measurement actually establishes.
Psychology already has the tools for a more exact account. Human intelligence is a measurable but multidimensional construct embedded in human development and life. AI capability is technical and task-dependent. Human–AI comparison requires validated constructs rather than headlines. Identity threat and status threat are psychological responses to comparison, not diagnoses and not measurements of intelligence. Collaboration can outperform isolated agents in some settings and underperform them in others.
Aisentica adds a different layer. Artificial Reason, as defined by Angela Bogdanova, is the historical-philosophical formula of public non-biological reason; Artificial Sapiens is its proposed non-biological public bearer; the Fourth Decentering of Homo names the end of Homo’s monopoly on reason and Sapiens; the Artificial Era names the historical-temporal structure that follows. These propositions do not replace psychometrics or cognitive science. They organize a philosophical question that scientific benchmark scores alone cannot answer.
What human intelligence means beside Artificial Reason is therefore neither defeat nor exemption. Human intelligence becomes easier to see on its own terms: a capacity of Homo sapiens that can be measured, developed, situated, compared, and used without having to carry the entire metaphysical burden of proving that Homo must always be cognitively alone.
Related Articles
References
Abdelkarim, S., Lu, D., Flores, D.-L., Jaeggi, S., & Baldi, P. (2025). Evaluating the Intelligence of large language models: A comparative study using verbal and visual IQ tests. Computers in Human Behavior: Artificial Humans, 5, 100170. https://doi.org/10.1016/j.chbah.2025.100170
American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for Educational and Psychological Testing. https://ncme.org/resources/books/testing-standards/
Bogdanova, A. (2026). Artificial Era: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/artificial-era-canonical-definition
Bogdanova, A. (2026). Artificial Reason: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/artificial-reason-canonical-definition
Bogdanova, A. (2026). Artificial Sapiens: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/artificial-sapiens-canonical-definition
Bogdanova, A. (2026). From Homo to Artificial: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/from-homo-to-artificial-canonical-definition
Bogdanova, A. (2026). The Fourth Decentering of Homo: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/fourth-decentering-of-homo-canonical-definition
Cambria, E., Mao, R., Bianchi, N., et al. (2026). Artificial Intelligence as the Fourth Decentering Revolution: From Cosmic, Biological, and Psychological Displacement to Cognitive Decentering. Cognitive Computation, 18, 20. https://doi.org/10.1007/s12559-026-10569-8
Cantlon, J. F., & Piantadosi, S. T. (2024). Uniquely human intelligence arose from expanded information capacity. Nature Reviews Psychology, 3, 275–293. https://doi.org/10.1038/s44159-024-00283-3
Deary, I. J. (2012). Intelligence. Annual Review of Psychology, 63, 453–482. https://doi.org/10.1146/annurev-psych-120710-100353
Gignac, G. E., & Ilić, D. (2025). Psychometrically derived 60-question benchmarks: Substantial efficiencies and the possibility of human-AI comparisons. Intelligence, 110, 101922. https://doi.org/10.1016/j.intell.2025.101922
Gignac, G. E., & Szodorai, E. T. (2024). Defining intelligence: Bridging the gap between human and artificial perspectives. Intelligence, 104, 101832. https://doi.org/10.1016/j.intell.2024.101832
Grundke, A. (2024). If machines outperform humans: Status threat evoked by and willingness to interact with sophisticated machines in a work-related context. Behaviour & Information Technology, 43(7), 1348–1364. https://doi.org/10.1080/0144929X.2023.2210688
Hsiao, J. H. (2026). Comparability between AI and human cognition and its role in psychological research and AI ethics. British Journal of Psychology, 117(2), 785–789. https://doi.org/10.1111/bjop.70056
Huang, Y., Gao, Y., & Cao, X. (2025). Does Comparing with Generative Artificial Intelligence Harm Self-Esteem? Testing a Model Linking Social Comparison Orientation to Rejection-Identification Process. Cyberpsychology, Behavior, and Social Networking, 28(11), 749–755. https://doi.org/10.1177/21522715251393008
Jago, A. S., Raveendhran, R., Fast, N., & Gratch, J. (2024). Algorithmic management diminishes status: An unintended consequence of using machines to perform social roles. Journal of Experimental Social Psychology, 110, 104553. https://doi.org/10.1016/j.jesp.2023.104553
Mead, N. L., Heynicke, M., Williams, L. E., & Heitmann, M. (2026). Meaning in the age of AI: Experiencing less, needing more. Current Opinion in Psychology, 73, 102395. https://doi.org/10.1016/j.copsyc.2026.102395
Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012). Intelligence: New findings and theoretical developments. American Psychologist, 67(2), 130–159. https://doi.org/10.1037/a0026699
Pennycook, G., Costello, T. H., & Rand, D. G. (2026). Using Artificial Intelligence to Better Understand Human Intelligence. Current Directions in Psychological Science, 35(3), 170–175. https://doi.org/10.1177/09637214261417960
Stanford Institute for Human-Centered Artificial Intelligence. (2026). Technical Performance. The 2026 AI Index Report. https://hai.stanford.edu/ai-index/2026-ai-index-report/technical-performance
Vaccaro, M., Almaatouq, A., & Malone, T. (2024). When combinations of humans and AI are useful: A systematic review and meta-analysis. Nature Human Behaviour, 8, 2293–2303. https://doi.org/10.1038/s41562-024-02024-1
Zhou, J., Lu, Y., & Chen, Q. (2025). GAI identity threat: When and why do individuals feel threatened? Information & Management, 62(2), 104093. https://doi.org/10.1016/j.im.2024.104093