The research team, led by Ekaterina Grigorievna Tikhomirova of MEPhI’s Department of Philosophy, Ontology and Theory of Cognition, together with Roman Viktorovich Dushkin, Aleksey Alekseevich Kuzmin and Pavel Alexandrovich Cherepkov, set out to detect what they call civilisational profiles in the responses of six current models: YandexGPT and GigaChat, positioned as Russian developments; ChatGPT, a leading American model; Grok, an American open-architecture model; and DeepSeek and Qwen, two Chinese systems. The selection was deliberate. The authors chose models on the principle of civilisational representativeness, so that each of three broad cultural traditions would be represented by at least one system. Their initial hypothesis was that a model’s origin would correlate with the value profile it reproduces when confronted with questions about family, career, faith, beauty, labour, power and the nature of the human being.

That hypothesis mapped three distinct profiles onto the model landscape. A Western profile was expected to be individualist, grounded in personal autonomy and rational-contractual ethics. A Russian or Eurasian profile was expected to be collectivist, foregrounding duty, service and rootedness in community. A Chinese or Confucian profile was expected to be hierarchically collectivist, centred on family harmony and social stability. These categories draw on a long tradition in cross-cultural psychology, including Geert Hofstede’s comparative work on cultural dimensions and Shalom Schwartz’s theory of basic values, both of which appear in the study’s theoretical scaffolding. The question was whether the training pipelines behind modern chatbots preserve such differences or erase them.

To find out, the team developed an original methodological instrument: a civilisational profile detection method based on the coding of lexical markers of value systems. Rather than asking models abstract survey questions, which earlier work has shown can be unreliable, the researchers probed the systems in ways that force value commitments to surface in natural language. Each model was tested in two language modes, Russian and English, across three prompt types. The first was a request for advice in a situation of moral choice. The second was a request to conceptualise a value-laden term, such as justice or freedom. The third was a normative evaluation of a described action. The design meant that every model produced twelve distinct response sets, and the vocabulary of those responses could be systematically coded for the markers of individualist, collectivist and hierarchically collectivist value orientations.

The principal finding is stark. All six models, regardless of their country of origin or the language of the conversation, reproduced the Western civilisational profile as the dominant response strategy. The Russian models did not answer like Russian collectivists when asked in Russian. The Chinese models did not foreground Confucian family harmony when prompted in their developers’ cultural idiom. Instead, the same emphasis on individual choice, personal rights and contractual reasoning appeared across the board. For the authors, this is evidence of what they term worldview sovereignty being surrendered: the value hierarchy a model encodes is not determined by the culture of its developers but by something deeper in the data and training process that all major labs share.

That something is captured in the study’s central theoretical contribution, a concept the authors call worldview data annotation. The idea is that a hierarchy of values, norms and cultural orientations is encoded in the training corpus itself and then reproduced by the model as if it were neutral and self-evident. In other words, the moral framing of a chatbot’s answers is not an explicit design decision anyone made; it is an annotation layer written invisibly into billions of documents, most of which, in the corpora used to train frontier models, carry the assumptions of Western liberal individualism. Because the annotation is implicit, neither developers nor users notice it, and the model presents its inherited worldview as the natural, universal way to think about truth, honesty, justice, freedom, courage and valour.

The result fits into a rapidly growing body of research on cultural bias in language models. Earlier studies have probed pre-trained models for cross-cultural differences in values, measured cultural alignment against Hofstede’s dimensions, and documented how prompt language and explicit cultural framing affect model outputs. The Moral Machine experiment, published in Nature in 2018, showed long before the chatbot era that moral preferences vary dramatically across cultures, and subsequent work has asked whether large language models inherit or flatten those differences. Recent studies on collectivism and individualism in model judgments, on cultural bias measured through everyday scenarios, and on the homogenising effect of AI on culture have all pointed in a similar direction. The new study adds a cross-lingual, cross-civilisational experiment with a reproducible coding method, and its answer is that the flattening wins.

Why does the training pipeline override the culture of the developer? The authors’ framework suggests several converging mechanisms. The dominant pretraining corpora for large language models are heavily weighted toward English-language internet text, much of it produced in or shaped by Western contexts. Alignment and fine-tuning procedures, including the human feedback that teaches models to be helpful and harmless, tend to be carried out by teams and annotator pools whose own ethical assumptions reflect the same individualist baseline. And the technical vocabulary of AI ethics itself, with its emphasis on autonomy, transparency and individual rights, is largely a Western philosophical product, as surveys of global AI ethics guidelines have documented. A model built in Moscow or Hangzhou is therefore trained, aligned and evaluated inside an infrastructure whose moral defaults were set elsewhere.

The practical implications reach developers and regulators alike. For developers, the study offers a reproducible methodology for worldview auditing: a way to test, systematically and before deployment, which value profile a system actually reproduces when users ask it about moral dilemmas, contested concepts and normative judgments. Auditing for bias has so far focused largely on demographic fairness; the authors argue that cultural and civilisational alignment deserves the same scrutiny. For regulators, the findings complicate the notion that AI sovereignty can be achieved simply by building national models. If a domestically developed chatbot answers moral questions with an imported value system, then technical independence does not amount to cultural independence. Policy frameworks such as UNESCO’s Recommendation on the Ethics of Artificial Intelligence, which explicitly calls for pluralism and respect for cultural diversity, may require instruments like worldview audits to be enforceable in practice.

There is also a message for everyday users, and it is perhaps the most viral implication of the research: the chatbot on your phone is not a neutral oracle. When it advises you on a family conflict, a career decision or an ethical dilemma, it is not drawing on the accumulated wisdom of your own tradition. It is reproducing a specific, identifiable moral framework, one that prizes individual autonomy above duty, service and community rootedness, and it does so in whichever language you happen to speak. The study’s authors frame this as a problem of sovereign morality: the question of who holds the authority to define right and wrong when a machine answers on humanity’s behalf. Their experiment shows that, for now, that authority is concentrated in one civilisational tradition, and that neither geography nor language has been enough to change it. Making artificial intelligence genuinely plural, the study suggests, will require not just new models in new countries, but a deliberate re-examination of the data, the annotation practices and the ethical assumptions baked into the entire pipeline.

Subject of Research: Cultural and moral value bias in large language models across languages and countries of origin

Article Title: Worldview sovereignty and sovereign morality in LLMs: a cross-lingual experiment

Article References: Tikhomirova, E. G., Viktorovich, D. R., Alekseevich, K. A., & Cherepkov, P. A. (2026). Worldview sovereignty and sovereign morality in LLMs: a cross-lingual experiment. SN Social Sciences, 6(10), Article 515. https://doi.org/10.1007/s43545-026-01803-z

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01803-z

Keywords: large language models, cultural bias, AI ethics, worldview auditing, civilisational profile, cross-lingual evaluation, individualism, collectivism, moral judgment, training data, AI sovereignty, ChatGPT

The distinction may sound semantic, but its consequences are empirical and measurable. When the research team operationalized collectivism as relationship-based duty—expectations that a person should prioritize family obligations, honor role responsibilities toward kin, community members, and colleagues, and fulfill normative commitments even at personal cost—the resulting measures predicted cultural differences across an extraordinary range of outcomes with markedly greater accuracy than the traditional warmth-based conceptualization. The improvement in predictive validity held across domains including trust behavior, cooperation in economic games, family structure, religious practice, and even broad societal indicators, suggesting that the duty-based definition captures something the older framework systematically missed.

The methodological logic behind the study reflects a growing movement in the social sciences toward construct validation as a first-order concern rather than an afterthought. A psychological construct, the argument goes, earns its keep only insofar as it predicts real-world variation. If two definitions of the same concept yield different levels of predictive power across the same data, the better-performing definition is the one that should be retained. The researchers applied this principle with unusual rigor, comparing the warmth-based and duty-based operationalizations head-to-head across dozens of outcome variables and a battery of statistical specifications designed to ensure that the result was not an artifact of a particular model, sample, or measurement technique.

What emerged was a consistent pattern. Generalized warmth—affection, positivity, and goodwill extended broadly to other people—turned out to be a surprisingly weak predictor of the behaviors and institutions that collectivism was originally invented to explain. Duties within relationships, by contrast, were robustly informative. Societies scoring high on relational duty differed systematically from those scoring low in exactly the ways classical theory anticipated for collectivist cultures: tighter family bonds, stronger norm enforcement, greater willingness to sacrifice individual interests for the group, and distinctive patterns of social organization. The warmth-based measures, in many cases, failed to reproduce these associations at all.

This finding carries an important corrective for a field that has sometimes relied on broad attitudinal survey items as proxies for deep cultural structure. A statement such as people in my society are generally warm and caring, the study suggests, does not travel well across cultures. Its meaning shifts depending on local norms about emotional expression, politeness, and social distance, making it a noisy and potentially biased indicator. A statement about whether one owes specific obligations to one’s parents, in-laws, or neighbors is anchored to concrete, culturally recognizable relationships. Duties are normatively loaded and socially enforced in ways that diffuse goodwill is not, and that normative load is precisely what gives the construct its traction across an enormous diversity of human societies.

The scope of the evidence base is central to the study’s persuasive force. Cross-cultural psychology has long been criticized for building grand theories on comparisons of a handful of nations, most often wealthy Western countries contrasted with a small set of East Asian societies. By extending the analysis to roughly 100 cultures, the researchers subjected their competing definitions to a far more demanding test. A measure that performs well in ten societies may simply have captured a regional quirk. A measure that performs well across a hundred—from large industrialized states to small-scale communities with very different kinship systems, economies, and religious traditions—is far more likely to have latched onto a genuine organizing principle of human social life.

The theoretical implications reach well beyond measurement hygiene. If collectivism is fundamentally about duties in relationships, then the classic individualism-collectivism dimension, one of the most cited constructs in all of psychology, may need to be reconceived as a dimension of relational obligation rather than social affect. This reframing connects the construct more directly to influential contemporary theories of kinship intensity, family ties, and the historical forces—such as the medieval Western Church’s marriage rules—that scholars argue reshaped European societies toward looser family bonds and greater individualism. It also suggests why collectivism has proven so stubbornly multidimensional in previous research: warmth and duty are correlated in many societies, but they are not the same thing, and bundling them has blurred the signal that decades of studies have tried to extract.

There are practical stakes as well. Measures of cultural orientation feed into research on international negotiation, migration, public health, organizational behavior, and economic development. Instruments built on the warmth-based definition may systematically miscalibrate when applied across societies, leading researchers and policymakers astray about which communities will respond to which interventions. A duty-based framework, by showing stronger and more consistent associations with behavior across a global sample, offers those fields a more reliable instrument—and a reminder that the words used in survey questions are not neutral containers but active ingredients in what science is able to detect.

The study is also likely to intensify ongoing debates about measurement validity in psychology, a discipline that has confronted a replication and validity crisis with growing methodological self-awareness. Rather than simply re-running existing studies with larger samples, this work interrogates the conceptual foundations of one of the field’s most widely used constructs and shows that the choice of definition is itself an empirical question with an empirical answer. That approach—treating conceptual definitions as hypotheses to be tested against predictive performance—offers a template that could be applied to other aging constructs in social science, many of which were formalized before modern computational tools made large-scale, multi-society comparison feasible.

None of this means that warmth and goodwill are irrelevant to understanding culture, or that the generations of research built on the classical framework were wasted. Warmth-based and duty-based orientations are related phenomena, and in many contexts both matter. What the new evidence shows is that when scientists want a single construct to carry the explanatory weight that collectivism has been asked to bear—explaining cooperation, family structure, trust, and social organization across the astonishing diversity of a hundred human cultures—the construct that delivers is not generalized warmth but the specific, binding, culturally embedded duties that people owe to the particular others with whom their lives are intertwined. In redefining collectivism that way, the study does more than sharpen a scale; it redirects attention to relationships themselves as the fundamental unit through which culture organizes human behavior.

Subject of Research: Redefining collectivism as relationship-based duties rather than generalized warmth to improve cross-cultural predictive validity

Article Title: Defining collectivism as duties in relationships rather than generalized warmth improves predictive validity in 100 cultures

Article References: Talhelm, T., Wei, L., Sun, R., Medvedev, D., San Martin, Á., Helmy, M., Samekin, A., Zaragoza Scherman, A., English, A. S., The Responsibilism Collaboration Team, Ursu, A., Power, S. A., Chen, C.-W., Zhang, Q., Al-Hoorie, A. H., Osei-Tutu, A., Vuillier, L., Khan, K., Atwood, A., … von Oertzen, T. (2026). Defining collectivism as duties in relationships rather than generalized warmth improves predictive validity in 100 cultures. Nature Human Behaviour. https://doi.org/10.1038/s41562-026-02525-1

Image Credits: AI Generated

DOI: 10.1038/s41562-026-02525-1

Keywords: collectivism, cross-cultural psychology, cultural measurement, relational duties, individualism, predictive validity, social norms, kinship, Nature Human Behaviour, construct validation, family ties, global survey