The Ideological Rorschach Test

Why a Decentralised Identity System Is Legible to Both Communist and Capitalist Frameworks, and Why That Might Matter More Than the Math

---

Authors: Alex Applebee and L. N. Combe

Affiliation: OMXUS Research Initiative

Contact: research@omxus.com

Date: February 2026

Status: Preprint --- Not peer reviewed

Series: OMXUS Research Series (Papers 29--31 of 33)

---

Abstract

We report an unexpected empirical observation: the OMXUS decentralised identity protocol, when presented to AI systems trained within ideologically opposed cultural contexts (Anthropic's Claude, trained predominantly on Western liberal-democratic corpora; DeepSeek, trained within the People's Republic of China under Chinese content governance), was recognised and endorsed by both systems from within their respective ideological frameworks --- without modification to the protocol. We argue this is not a trivial result. Systems that address fundamental questions of identity, resource distribution, and collective safety typically provoke ideological opposition from at least one major political tradition. OMXUS does not, because it answers the question both traditions actually care about --- who controls identity and resources? --- with an answer neither tradition has claimed: nobody and everybody. We explore why this occurs, what it reveals about the deeper structure of the capitalist-communist disagreement, and what it might mean for building systems that survive contact with geopolitics.

This unified thesis integrates the core Rorschach finding ((Applebee & Combe, 2026, "The Switchboard")), the historical analysis of knowledge monopolies ((Applebee & Combe, 2026, "The Smartness Trap")), and the plain-language account designed to be handed to a person who has never read an academic paper in their life ((Applebee & Combe, 2026, "The Invisible Fence")).

Keywords: decentralised identity; political economy; ideological compatibility; AI cultural bias; common-pool resources; institutional design; smartness mandate; data colonialism; sovereign technology

---

Author's Note: The Nine Values and the Fourteen Goals

This paper exists because of a problem with research.

The problem is not that the research is wrong. The problem is that the research does not reach the people it is about. Academic papers sit behind paywalls. Policy documents sit in filing cabinets. The people whose lives these systems shape --- the ones working double shifts, raising kids alone, burying their mates too young --- never see the evidence. And when they do, it is written in a language designed to exclude them.

So we wrote it differently. We wrote it through nine values --- not because we invented them, but because they are already there. They are the values that working people, particularly working men, already hold. They are the door.

The Nine Values:

1. Loyalty --- "You stand by your family."
2. Hard work --- "You earn your keep."
3. Toughness --- "You don't complain."
4. Providing --- "You put food on the table."
5. Mateship --- "You don't let your mates down."
6. Fairness --- "You have a go, you get a fair go."
7. Being your own man --- "Nobody tells me what to do."
8. Call a spade a spade --- "You call it like you see it."
9. If it's your family, you're there --- "You show up."

The research does not say these values are wrong. It says:

• You value loyalty? The system you're loyal to isn't loyal back.
• You believe in hard work? You're working twice the hours for half the result.
• You value toughness? They renamed your silence "consent."
• You provide? The food you're providing is poison.
• You value mateship? They removed every structure that let you see your mates.
• You believe in a fair go? Here's the numbers. It's rigged.
• You're your own man? You vote for someone else to make your decisions.
• You call a spade a spade? Here's the spade. The justice system is inverted.
• You show up? Your nan fell. Ambulance takes 14 minutes. You could be there in 60 seconds.

Each value is a door into the research. And the research behind each door connects to one or more of the 14 goals this project exists to build:

1. Fire all politicians. You vote on everything.
2. Work 22 hours max. Keep your pay. Choose your hours.
3. Free all prisoners.
4. Eradicate courts.
5. Fire all police, justice, and corrections staff.
6. Re-employ all fired staff in functional positions.
7. Legalise drugs. Stock pharmacies. Cheap.
8. Internet costs nothing.
9. No foreign investment in housing.
10. Food contains only things proven safe.
11. Monkey bars at every bus stop. Climbing walls on all stairwells.
12. Every school is play, mastery, curiosity.
13. $29 ring. Press it, your people come in 60 seconds.
14. Cancer is 90% preventable. Here's how.

This paper addresses goals 1, 7, 8, and 13 most directly. But the framework it describes --- designing from wound rather than from ideology, building systems that pass the ideological Rorschach test --- is the methodology behind all fourteen.

Why cross-ideology compatibility matters: The left-right divide is partly manufactured. Gilens and Page (2014) found that US citizens have "near zero" independent influence on policy. The Harvard Ash Center (2020) found 93.1% Chinese citizen satisfaction across 31,000+ interviews. Each side's media frames the other as unfree. Strip that away and both populations fear concentrated power that does not represent them, and both want agency AND security. The nine values are not left or right. Loyalty is not conservative. Fairness is not progressive. They are human. And the system that honours them must be legible to humans across every ideological line --- not because we water it down, but because we designed it from the wound that cuts across all of them.

Strong rhetoric serving human flourishing is not epistemic abuse. We do not soften. We sharpen.

---

Table of Contents

• Part I: The Ideological Rorschach Test
• Part II: The Smartness Trap
• Part III: The Invisible Fence
• References
• Appendix A: The Empathy Equation
• Appendix B: Cross-References to the OMXUS Research Series
• Appendix C: Annotated Bibliography

---

# Part I: The Ideological Rorschach Test

(Applebee & Combe, 2026, "The Switchboard") of the OMXUS Research Series

---

Chapter 1: Introduction

The twentieth century's defining political conflict was between two answers to a single question: who should control the means of production and the distribution of resources?

• The capitalist answer: Private individuals, through markets, with minimal state interference.
• The communist answer: The collective, through the state, with minimal private accumulation.

This disagreement produced two world wars' worth of proxy conflicts, a nuclear standoff, and a body of political theory so vast that no individual can read it all. It also produced a deep assumption that any system addressing resource distribution must be legible to one tradition and threatening to the other. If capitalists like it, communists won't. If communists like it, capitalists won't. This is treated as a law of political nature.

We present evidence that it is not.

1.1 The Observation

During development of the OMXUS decentralised identity protocol, the system was presented to two large language models (LLMs) with markedly different training environments:

1. Claude (Anthropic): Trained predominantly on English-language, Western, liberal-democratic corpora. Reflects broadly Western assumptions about individual rights, market economics, and institutional design.
2. DeepSeek (DeepSeek AI, PRC): Trained within the People's Republic of China under Chinese content governance. Reflects a cultural context in which collective action, state-guided development, and revolutionary liberation are foundational narratives.

Both systems engaged positively and substantively with OMXUS. Neither flagged ideological incompatibility. More importantly, each system recognised the protocol from within its own framework: Claude saw decentralisation, individual ownership, and market-based verification pricing; DeepSeek saw collective benefit, equal distribution, liberation from corporate and state gatekeeping, and the commons as primary beneficiary.

The protocol was not modified between presentations. The same system was legible to both.

1.2 Why This Is Not Trivial

An optimistic reading is that OMXUS is simply inoffensive --- bland enough to avoid triggering ideological filters. We argue this is wrong. The protocol makes strong claims about identity (no state or corporate authority), resource distribution (equal per-capita sharing of 70% of revenue), and collective obligation (vouching, emergency response). These are precisely the domains where ideological conflict is fiercest. A system that makes strong claims in contested territory and yet is recognised by both sides is doing something structurally interesting.

1.3 Scope and Limitations

We are not claiming that LLM responses constitute proof of cross-ideological viability. AI systems are imperfect proxies for political traditions. DeepSeek's engagement does not represent the Chinese Communist Party's endorsement, any more than Claude's engagement represents the United States government's. What LLM responses do reveal is the structure of the cultural-ideological frames embedded in training data --- which frameworks are available, which pattern-matches activate, and which produce recognition rather than rejection. That is the evidence we examine.

---

Chapter 2: The Structure of the Disagreement

2.1 What Capitalists and Communists Actually Argue About

Beneath the vast theoretical apparatus, the core disagreement is about control:

Table 1: The structural disagreement between capitalist and communist traditions.

Notice that both traditions claim to serve the people. Capitalism claims that private ownership and markets produce optimal outcomes for all. Communism claims that collective ownership prevents exploitation. The disagreement is not about goals --- both claim prosperity and freedom --- but about mechanism: who holds the lever, and why that's safe.

2.2 The Hidden Agreement

Both traditions also share a hidden assumption: someone must hold the lever. In capitalism, it is the property owner, disciplined by markets. In communism, it is the state, disciplined by revolutionary mandate. Both assume that identity and resources require a controller --- the argument is only about who.

This shared assumption is what OMXUS violates.

---

Chapter 3: How OMXUS Is Read by Each Tradition

3.1 The Capitalist Reading

A system trained on Western liberal-democratic values recognises in OMXUS:

1. Decentralisation: No central authority controls identity. This resonates with libertarian and classical liberal suspicion of state power.
2. Individual ownership: Each person owns their Human Existence Record (HER). Property rights over personal data --- a capitalist value.
3. Market-based pricing: Verification is sold at $0.001 per call. There is a product, a price, and a market. This is legible as a business.
4. Voluntary participation: Nobody is compelled to join. Entry and exit are free. This is the market model.
5. Permissionless innovation: Any developer can integrate the SDK. No gatekeeping by a central authority.

What the capitalist reading does not see (or sees as secondary): the equal per-capita distribution, the non-transferability of the identity token, the collective emergency response obligation. These are present but do not trigger ideological alarm, because they operate within a framework of individual choice and market activity.

3.2 The Communist Reading

A system trained within Chinese cultural and political context recognises in OMXUS:

1. Collective benefit: 70% of revenue distributed equally to all verified humans. This is "from each according to ability, to each according to need" implemented in code.
2. Liberation from corporate control: Identity is not owned by Google, Facebook, or any corporation. This resonates with anti-imperialist and anti-capitalist critique.
3. Liberation from state control: Identity is not issued by any government. In a context where revolutionary legitimacy comes from liberating people from exploitative systems, a system that removes all gatekeepers --- including state ones --- can be read as completing the revolution rather than opposing it.
4. The commons: The network itself is the shared resource. Token holders collectively benefit from its growth. This is a commons, not a market.
5. Mutual aid: The emergency response system --- verified humans responding to nearby emergencies --- is collective care, not individual heroism.

What the communist reading does not see (or sees as secondary): the market pricing, the individual ownership framing, the absence of centralised coordination. These are present but do not trigger ideological alarm, because they operate within a framework of collective benefit and equal distribution.

3.3 The Rorschach Effect

Like a Rorschach inkblot, OMXUS presents the same structure to both observers, and each sees what their framework prepares them to see. This is not because the system is vague or empty --- it makes strong, specific commitments. It is because the system's commitments happen to satisfy the actual concerns of both traditions while violating the assumed mechanism of both.

Table 2: How each tradition reads the same protocol features.

---

Chapter 4: Why This Happens --- The "Nobody and Everybody" Answer

4.1 The Control Question, Revisited

Recall the core question: Who controls identity and resources?

• Capitalism answers: The individual (through property rights and markets).
• Communism answers: The collective (through the state or party).
• OMXUS answers: Nobody and everybody.

"Nobody" because there is no central authority --- no company, no government, no party, no individual --- that can unilaterally create, revoke, or modify an identity or alter the distribution formula.

"Everybody" because the network is governed by its participants, revenue is shared among all verified humans, and identity is established through mutual vouching among peers.

This is not a compromise between capitalism and communism. It is not "a bit of both." It is a different answer to the same question --- one that neither tradition has a ready-made objection to, because the objections each tradition has are aimed at the other tradition's answer, not at this one.

4.2 Capitalism Objects to Communism, Not to OMXUS

The capitalist objection to communism is: centralised control leads to authoritarianism and inefficiency. This objection does not apply to OMXUS, because there is no centralised control.

4.3 Communism Objects to Capitalism, Not to OMXUS

The communist objection to capitalism is: private ownership leads to exploitation and inequality. This objection does not apply to OMXUS, because identity is non-transferable, revenue is equally distributed, and no individual can accumulate disproportionate control.

4.4 Both Object to Each Other's Failure Mode

Table 3: How OMXUS avoids both traditions' failure modes.

---

Chapter 5: Mutual Misperception --- The Manufactured Divide

5.1 Each Population Believes the Other Is Unfree

The ideological opposition between East and West is not purely intellectual --- it is actively constructed and maintained by media ecosystems with structural incentives to portray the other negatively. Each population believes the other is unfree, and each thinks itself lucky:

Table 4: How each population perceives the other's system.

5.2 Both Critiques Have Empirical Support

The striking finding is that both narratives have empirical evidence supporting them:

The "American oligarchy" critique: Gilens and Page (2014) analysed 1,779 policy issues and found that "economic elites and organised groups representing business interests have substantial independent impacts on U.S. government policy, while average citizens and mass-based interest groups have little or no independent influence." Though critics note elites and the middle class agree on policy >90% of the time (Bashir, 2015), the finding that citizen preferences have "near zero" independent effect when they diverge from elite preferences is robust.

The "Chinese satisfaction" data: The Harvard Ash Center conducted the longest non-Chinese survey of Chinese public opinion (2003--2016), with over 31,000 face-to-face interviews across China. They found 93.1% satisfaction with the central government, with satisfaction increasing over the study period (Cunningham, Saich & Turiel, 2020). Whatever one thinks of the Chinese system, its citizens do not uniformly experience it as oppressive.

5.3 Media Amplification

Both views are distorted by structural media bias:

• New York Times coverage of China has been net negative since 1990, regardless of actual events in China (Peng, Chen & Wu, 2023)
• COVID-19 coverage was described as "one of the most biased and politicized cases in recent memory" (Jia & Lu, 2021)
• Each media ecosystem has institutional incentives to portray the other as threatening --- this generates engagement, justifies military spending, and provides political utility

5.4 The Shared Layer Beneath the Misperception

Strip away the propaganda layer, and both populations share concerns:

1. Fear of concentrated power that doesn't represent ordinary people (whether corporate or state)
2. Desire for individual agency and collective security --- not one at the expense of the other
3. Distrust of elites who claim to act in the public interest while serving themselves

This explains why OMXUS is legible to both: it addresses what both populations actually fear, not what each side's media says the other fears. The decentralisation satisfies Western fears of state capture; the equal distribution satisfies Eastern fears of elite accumulation; the mutual accountability satisfies both.

5.5 The Neurobiological Basis

This cross-ideological recognition may have neurobiological grounding. Mirror neuron systems --- discovered by Rizzolatti and colleagues (2004) --- enable humans to recognise suffering and intention across cultural boundaries. When we observe another person in pain or distress, the same neural circuits activate as if we experienced it ourselves. This is pre-ideological empathy.

Wound-driven design speaks directly to these circuits. A designer who builds from experienced harm creates systems that mirror neurons can recognise --- systems that address suffering the observer can neurologically feel, regardless of how their ideology frames the cause.

De Waal's work on empathy (2009) suggests this capacity is evolutionarily ancient and cross-cultural. Both Western and Eastern populations possess the same empathy circuitry. The ideological layer is a cultural overlay on a shared neurobiological substrate. Systems that address genuine human needs --- safety, agency, belonging --- resonate at this deeper level, even when the ideological framing differs.

Gallese's "shared manifold" hypothesis (2001) provides the mechanism: mirror neurons create a shared representational space between self and other. When a system is designed from experienced failure, it carries the imprint of that failure in its structure. Observers from any tradition can recognise the failure being addressed --- not because they share the designer's ideology, but because they share the designer's neurobiology.

5.6 The Oxytocin-Trust Connection

Carter (2017) demonstrated that oxytocin in safe contexts enables social bonding, while vasopressin dominates under threat. Kosfeld et al. (2005) showed that oxytocin significantly increases trust in economic games. The OMXUS vouch ceremony --- physical co-presence, mutual recognition, shared commitment --- is architecturally designed to create conditions where oxytocin, not vasopressin, governs the interaction. This is not metaphor. It is endocrinology applied to protocol design.

---

Chapter 6: The Deeper Pattern --- Systems Born from Failure

6.1 Design by Ideology vs. Design by Wound

Most political systems are designed by ideology: a theory of how the world should work is developed first, and institutions are built to implement it. Capitalism was theorised by Smith and Ricardo; communism by Marx and Engels. Both are top-down designs: theory first, system second.

OMXUS was not designed this way. It was designed by someone who had been failed by existing systems --- repeatedly, across multiple domains of institutional failure --- and who asked: what would have to be true about a system for it not to produce these failures?

This is design by wound, not by theory. And it produces a qualitatively different kind of system, because the design constraints come from experienced failure rather than imagined success. The designer is not asking "what would utopia look like?" but "what would have to be different for the worst things that happened to me to not happen to anyone else?"

6.2 Why Wound-Driven Design Passes the Rorschach Test

Ideologically designed systems are optimised for one framework and therefore threaten the other. A system designed to maximise individual liberty will alarm collectivists. A system designed to maximise collective welfare will alarm individualists. The ideology is the constraint, and the constraint is one-sided.

Wound-driven design optimises for failure avoidance across multiple systems. If the designer has been failed by corporate power, the system will resist corporate control. If the designer has been failed by state power, the system will resist state control. If the designer has been failed by community silence, the system will build in emergency response. If the designer has been failed by systems that assign blame instead of providing help, the system will prioritise response over judgment.

The result is a system that is not aligned with any ideology but is responsive to the failures of all of them. Both capitalists and communists can recognise it, because both have seen the failures it addresses --- they simply attribute those failures to the other side.

6.3 The Empirical Signature

This produces a testable prediction: systems designed from lived experience of institutional failure will show higher cross-ideological legibility than systems designed from political theory. We cannot test this rigorously here --- we have one case study, not a dataset --- but we note it as a hypothesis worth investigating.

---

Chapter 7: Implications

7.1 For System Design

If cross-ideological legibility is achievable --- and if it depends on designing from failure rather than from theory --- then the current practice of building systems within ideological silos is unnecessarily limiting. The question "is this a capitalist system or a socialist system?" may be the wrong question. The better question may be: "what failures does this system prevent, and are those failures recognised across ideological lines?"

7.2 For Geopolitics

Any system that handles identity and resource distribution will eventually encounter geopolitics. Systems legible to only one ideological tradition will be adopted by half the world and resisted by the other half. Systems legible to both have a structural advantage: they can be adopted across political boundaries without requiring either side to abandon its self-understanding.

This does not guarantee adoption. Political resistance is often strategic, not ideological --- states may oppose systems that reduce their control regardless of ideological compatibility. But removing the ideological objection removes one barrier, and in international adoption, removing barriers matters.

7.3 For the Relationship Between Suffering and Design

This is the implication we are most cautious about stating, but it may be the most important.

The observation that wound-driven design produces cross-ideological legibility implies that certain kinds of understanding are only available to people who have been failed by existing systems. A theorist who has only experienced functioning institutions will design systems that improve those institutions. A person who has experienced their total failure will design systems with different structural properties --- not better in every way, but different in ways that pure theory cannot reach.

This does not romanticise suffering. Suffering is not a credential; it is a cost. Most people who are failed by systems do not go on to design better ones --- they are too busy surviving the failure. But when someone does, the resulting design may contain structural insights that are unavailable from within any single ideological tradition, because the designer's experience cuts across the boundaries that ideology maintains.

---

Chapter 8: Limitations

1. Two AI systems do not constitute a political science study. LLM responses are suggestive evidence of cultural-frame compatibility, not proof of cross-ideological viability among actual political actors.

1. Recognition is not adoption. Both AI systems recognised OMXUS as compatible with their frameworks. This does not mean the Chinese Communist Party or the United States Congress would adopt it. Political adoption involves power, strategy, and institutional inertia --- not just ideological compatibility.

1. We have one case study. The claim that wound-driven design produces cross-ideological legibility is a hypothesis generated from n = 1. It requires testing across multiple systems.

1. The "nobody controls it" claim is aspirational. In practice, someone writes the code, someone hosts the servers, someone makes governance decisions. True decentralisation is a spectrum, not a binary, and OMXUS is early in its development. The protocol's ideological legibility may not survive contact with implementation reality.

1. Ideological Rorschach tests can be misleading. A system that is legible to everyone may also be claimed by everyone --- leading not to peace but to competing interpretations that produce new conflict. We acknowledge this risk.

1. AI systems are trained to be agreeable. The strongest counterargument to our finding is that both Claude and DeepSeek are optimised to engage positively with user-presented material. A control test --- presenting a genuinely ideologically one-sided system to both AIs and observing whether both still approve --- would strengthen the claim. We have not conducted this control.

1. The actual AI transcripts are not archived with this paper. The empirical basis of the observation --- the specific prompts and responses from Claude and DeepSeek --- should be stored alongside this analysis. They are not. This is a documentation gap, not an evidence gap, but it matters for reproducibility.

---

Chapter 9: Conclusion

The twentieth century's great ideological conflict was about who should hold the lever of control over identity and resources. Capitalism said the individual. Communism said the collective. Each built systems accordingly, and each objected to the other's system on the grounds that the wrong entity held the lever.

OMXUS removes the lever.

Identity is established by mutual vouching, not granted by any authority. Revenue is distributed equally, not accumulated by owners or allocated by officials. Emergency response is a network function, not a state monopoly. The protocol enforces these properties through code, not through officials who might be captured, corrupted, or overthrown.

When presented to AI systems embodying opposed ideological traditions, both recognised the protocol as compatible with their own values --- because both traditions' actual concerns (exploitation, authoritarianism, inequality, loss of agency) are addressed, while neither tradition's assumed mechanism (private ownership, state control) is employed.

We propose that this cross-ideological legibility is not accidental. It is a structural consequence of designing from experienced institutional failure rather than from political theory. A person who has been failed by the state will not build a state-controlled system. A person who has been failed by the market will not build a market-controlled system. A person who has been failed by both will build something else --- something that neither tradition has a ready-made objection to, because neither tradition designed it.

Whether this "something else" works in practice is an engineering question, not an ideological one. But the fact that it can be asked --- that a system can exist in the space between capitalism and communism without being attacked by either --- may itself be the most important finding.

The math matters. The security proofs matter. The protocol design matters. But if the system cannot survive contact with the world's deepest political division, none of that matters at all.

OMXUS, it appears, might survive that contact. Not because it is politically neutral, but because it is politically complete: it addresses what both sides care about, by answering a question neither side thought to ask.

What if nobody holds the lever?

---

# Part II: The Smartness Trap

(Applebee & Combe, 2026, "The Smartness Trap") of the OMXUS Research Series

How the Mandate to "Become Smart or Go Extinct" Became the Latest Mechanism of Control

---

Chapter 10: The Mandate

In 2022, Orit Halpern and Robert Mitchell coined the term "smartness mandate" --- the imperative embedded in every AI narrative that demands we "become smart or else go extinct as a species" (Halpern & Mitchell, 2022, p. 220).

Smartness --- as in smartphones, smart cities, smart homes, smart contracts --- is not merely a marketing term. It is "an epistemology, that is, a way of knowing and representing the world so that one can act in and upon that world" (Halpern & Mitchell, 2022, p. xi). The mandate does not ask whether you want to be smart. It tells you that the alternative is extinction. Accept the terms. Install the update. Authenticate with Google.

This paper argues that the smartness mandate is structurally identical to every previous knowledge monopoly --- and that the OMXUS ecosystem is the printing press that breaks it.

---

Chapter 11: The Four Mechanisms

11.1 Surveillance as Service

Matteo Pasquinelli warns that "the problem of AI has nothing to do with intelligence per se but with the manner in which it is applied to the governance of society and labor via statistical models --- ones that should be transparent and exposed to public scrutiny" (Pasquinelli, 2019, p. 3). He describes the result as a "planetary business of surveillance and forecasting" (Pasquinelli, 2023, p. 12).

The mechanism is simple: offer a useful service (email, maps, social connection, AI assistance). In exchange, collect data. Use the data to build statistical models. Apply the models to governance --- not political governance, but behavioural governance. What you see, what you're offered, what you can access, how much it costs.

Google does not sell your data. Google sells predictions about your behaviour, derived from your data, to entities that want to modify your behaviour. This is surveillance --- but it is experienced as convenience. The mandate says: use Google or lose email. Use AI or fall behind. Become smart or go extinct.

(Applebee & Combe, 2026, "Platform Gatekeeping") documents the result: Five companies control digital identity. Combined 2023 revenue: $1.037 trillion per year. None elected. None appointed. None accountable.

11.2 Extraction as Convenience

Nick Couldry and Ulises Mejias describe a process they call "data colonialism" --- "a new phase in the long history of extractive relationships" that transforms "lived experience into abstract, commodifiable data points" (Couldry & Mejias, 2018).

The parallel to historical colonialism is structural, not metaphorical:

Table 5: The evolution of colonial extraction.

The extraction mechanism has evolved. You don't need armies when you have APIs. The smartness mandate replaces the gun with the login page: authenticate with us or cease to exist digitally ((Applebee & Combe, 2026, "Platform Gatekeeping")). The cost of refusal is not death --- it's social and economic irrelevance.

Negueruela del Castillo and Neri describe how this extraction creates "digital shadows" that replace real environments: "The city becomes increasingly indistinguishable from its digital shadow, cast by the vast apparatus of data collection and algorithmic processing" (Negueruela del Castillo & Neri, 2024). The map replaces the territory. The data replaces the person. The shadow is governed; the person merely generates it.

11.3 Reification as Analysis

When AI systems process human behaviour, they do not analyse it --- they reify it. They transform "living relations into fixed, computable categories" (Negueruela del Castillo & Neri, 2024). This is algorithmic reification: the conversion of fluid, contextual, culturally embedded human behaviour into static data types.

This matters because the OMXUS justice research (Papers 1--15) demonstrates that the existing justice system does exactly this. It converts complex human behaviour into binary categories --- guilty/innocent, truthful/deceptive --- using instruments (behavioural deception cues) that are 91.3% inverted ((Applebee & Combe, 2026, "Signal Inversion")). The signal inversion effect is algorithmic reification applied to human freedom: a living person is converted into a category ("deceptive"), the category is processed ("guilty"), and the output is applied to the person ("imprisoned").

The smartness mandate extends this pattern beyond the justice system. Every platform that categorises users --- credit scores, risk assessments, content moderation decisions, hiring algorithms --- performs the same reification. Living people become computable categories. The categories are governed. The people merely generate them.

11.4 Invisibility as Power

Eduard Kaeser observes that "invisibility is a signum of power" (Kaeser, 2018, p. 28). The smartness mandate operates precisely through this invisibility. Technology becomes visible only "when it is not working and we cannot perform our tasks" (Gerber & Atalay Franck, 2024). When it works, it is transparent --- which means its power is unaccountable.

The Facebook outage of October 2021 made the infrastructure visible for six hours. Government services in India collapsed. Healthcare communication in Brazil stopped. Emergency coordination across sub-Saharan Africa went dark (BBC, 2021). For six hours, 3 billion people could see the dependency. Then it was fixed, and the infrastructure became invisible again. The dependency remained. The visibility did not.

(Applebee & Combe, 2026, "The Invisible Network") responds: a BLE mesh network has no invisible centre. There is no infrastructure to make visible because the infrastructure is the people. When your neighbour's phone is the network, the network is never invisible --- it's standing next to you.

---

Chapter 12: The Historical Pattern

The smartness mandate is not new. It is the latest iteration of a pattern that repeats whenever a knowledge technology is monopolised:

Table 6: The pattern of knowledge monopolies across history.

Each monopoly was broken by the same force: hardware cheap enough that control shifted from institutions to individuals. The printing press broke the Church. The personal computer broke broadcast. The smartphone broke the newspaper.

(Applebee & Combe, 2026, "Sovereign AI Infrastructure") identifies the next break: the ASUS ExpertCenter Pro ET900N G3. $50,000 AUD. 784GB coherent memory. Runs 671 billion parameters on your desk. The printing press for thought.

---

Chapter 13: Counter-Practices

Gerber and Atalay Franck note that "the simulation of democratic procedures in a nascent authoritarian state is designed to conceal the fact that the state in question isn't a democracy any more. If a human practice is being replaced by a simulation of it, what ultimately results is deskilling: an impoverishment of the capacity to engage in the original activity" (Martin, 2024, p. 60).

The smartness mandate simulates capability while producing dependency. "Sign in with Google" simulates identity while producing dependence on Google. AI-generated answers simulate knowledge while producing dependence on the AI provider. Smart city infrastructure simulates urban intelligence while producing dependence on the platform vendor.

The OMXUS ecosystem is not a simulation. It is a set of counter-practices --- technologies deployed for sovereignty rather than governance:

Table 7: The smartness mandate and its OMXUS counter-practices.

Each counter-practice refuses the mandate without refusing the technology. OMXUS uses cryptography, mesh networking, ML inference, and blockchain anchoring --- all products of the same technological era that produced the mandate. The difference is not the technology. The difference is who controls it.

---

Chapter 14: Digital Pluriversality

Negueruela del Castillo and Neri, drawing on Walter Mignolo, call for "digital pluriversality" --- a demand that "urban AI systems be repurposed toward emancipatory ends" and that citizens have "the right not just to access urban data but to participate in the very definition of what constitutes urban possibility" (Negueruela del Castillo & Neri, 2024).

This is what OMXUS builds. Not one system to replace five corporations --- but a framework in which identity, communication, governance, and knowledge are controlled by the people who use them, through hardware they own, on networks they constitute.

The "right to the algorithmic city" --- extending Lefebvre's concept --- means the right to participate in the definition of the categories, the architecture of the algorithms, and the governance of the outputs. OMXUS implements this through:

• Direct democracy ((Applebee & Combe, 2026, "Swiss Direct Democracy")): Citizens define the questions, not representatives
• Quadratic voting ((Applebee & Combe, 2026, "Democratic Voting Mechanisms")): Intensity of preference is expressed, not just binary choice
• Physical verification ((Applebee & Combe, 2026, "Be In The Same Room")): Identity is confirmed by neighbours, not algorithms
• Open-weight AI ((Applebee & Combe, 2026, "Sovereign AI Infrastructure")): The model runs on your hardware, under your terms
• BLE mesh ((Applebee & Combe, 2026, "The Invisible Network")): The network is the people, not the infrastructure

This is not a utopian proposal. Every component is proven technology ((Applebee & Combe, 2026, "Social Group Scaling") in the Trap Map). The combination is novel. The components are not.

---

Chapter 15: The Embodiment Gap

Gerber and Atalay Franck identify a fundamental gap: "Whatever AI will generate, it will not come from some social/physical body and its interactions with the world" (Gerber & Atalay Franck, 2024). AI lacks a "sentient body" --- and therefore cannot generate genuine spatial, social, or governance intelligence from embodied experience.

This is why OMXUS insists on physical infrastructure:

• BLE requires physical proximity --- you must be near the mesh to use it
• Co-presence requires physical bodies --- you must be in the room to be counted
• The vouch ceremony requires physical presence --- you must stand with your witnesses
• The emergency button requires physical response --- someone must come to your door

The smartness mandate abstracts everything into data. OMXUS re-embodies everything into proximity, presence, and physical relationship. The gap between AI and embodiment is not a limitation to be overcome --- it is a design principle. Systems that require bodies to function cannot be controlled by entities that have none.

---

Chapter 16: The Trap, Revisited

(Applebee & Combe, 2026, "Platform Gatekeeping") asked: why do five companies control your identity? Because we let them.

This paper asks a different question: why do we let them?

Because the smartness mandate makes the alternative sound like extinction. "Leave Google" sounds like "leave civilisation." "Run your own AI" sounds like "build your own power grid." "Use mesh networking" sounds like "become a survivalist."

The mandate frames sovereignty as regression. That's the trap. The trap is not the technology --- it's the framing. The framing says: smart = corporate, sovereign = primitive. The framing is wrong.

• Switzerland has run direct democracy for 178 years. They're the richest country in the world. ((Applebee & Combe, 2026, "Swiss Direct Democracy"))
• Australia has run mandatory voting for 102 years with 90%+ turnout. ((Applebee & Combe, 2026, "Be In The Same Room"))
• Bitcoin has run trustless consensus for 15+ years at $1T+ market cap. ((Applebee & Combe, 2026, "Trust-First Governance"))
• BLE mesh exists in 6.8 billion devices. ((Applebee & Combe, 2026, "The Invisible Network"))
• A $50K desktop runs 671B parameters. ((Applebee & Combe, 2026, "Sovereign AI Infrastructure"))

None of this is primitive. All of it is sovereign.

The smartness mandate says: become smart or go extinct. OMXUS says: become free. The technology is the same. The ownership is different. That's the whole paper.

---

# Part III: The Invisible Fence

(Applebee & Combe, 2026, "The Invisible Fence") of the OMXUS Research Series

The kitchen-table version of (Applebee & Combe, 2026, "The Smartness Trap")

---

Chapter 17: The Kitchen-Table Version

There's a fence around you. You can't see it. That's the point.

---

The mandate

Someone decided that you have to "become smart or go extinct." Not you --- someone. They called it the "smartness mandate." Smart phones. Smart cities. Smart homes. Smart everything.

Smart means: connected to their servers. Monitored by their algorithms. Governed by their terms. Invisible to you --- until it breaks.

When Facebook went down for six hours in 2021, governments in India couldn't function. Hospitals in Brazil lost communication. Emergency services across Africa went dark. For six hours, three billion people could see the fence. Then it was fixed and the fence became invisible again.

The fence is still there. You just can't see it.

---

The pattern

This has happened before.

Before printing: The Church controlled books. They said: "Accept our interpretation or be a heretic." For a thousand years, that worked. Then Gutenberg built a press and anyone could read for themselves. The Church called it dangerous. They lost.

Before computers: Broadcasters controlled information. They said: "Watch our channel or be uninformed." For a hundred years, that worked. Then personal computers appeared and anyone could publish. The broadcasters called it chaos. They lost.

Now: Five companies control digital identity and AI. They say: "Use our platform or cease to exist. Use our AI or fall behind. Become smart or go extinct."

They will lose too. The question is how long we let them run it.

---

The extraction

A researcher named Nick Couldry called it "data colonialism." Same structure as regular colonialism --- take something valuable from people, use it to get rich, tell them it's for their benefit.

Table 8: Colonialism, old and new.

The difference: you don't need armies when you have login pages.

Google doesn't sell your data. Google sells predictions about what you'll do next, based on everything you've ever searched, emailed, walked past, or watched. That's not a service. That's a one-way mirror with a cash register on the other side.

---

The invisible part

A philosopher named Eduard Kaeser said: "Invisibility is a signum of power." The more powerful the system, the less you see it.

You don't see Google's servers. You don't see the algorithm that decides which search results you get. You don't see the content moderation system that decides what you're allowed to say. You don't see the facial recognition that logs where you walk. You don't see the predictive model that decides whether you get a loan.

You see the search bar. You see the login page. You see the "I agree" button. That's all you're supposed to see.

---

The counter

(Applebee & Combe, 2026, "Platform Gatekeeping") asked: why do five companies control your identity? Because we let them.

This paper asks: why do we let them? Because they made the alternative sound like extinction. "Leave Google" sounds like "leave civilisation." "Run your own AI" sounds like "build your own power grid."

That framing is wrong. Here's what "the alternative" actually looks like:

• Identity without Google: Your name + birthday + four words you choose. Deterministic forever. No account. No server. No one to press a button. ((Applebee & Combe, 2026, "Be In The Same Room"))
• Communication without towers: Bluetooth. Already in your phone. Already in 6.8 billion phones. No ISP. No Facebook. Just phone to phone. ((Applebee & Combe, 2026, "The Invisible Network"))
• AI without subscription: A $50,000 desktop that runs 671 billion parameters. Less than a Toyota Corolla plus a holiday. No terms of service. No content policy. Yours. ((Applebee & Combe, 2026, "Sovereign AI Infrastructure"))
• Democracy without politicians: Sign your vote on your phone. It relays through the mesh. No polling booth. No politician counting. Switzerland's been doing it for 178 years. ((Applebee & Combe, 2026, "Swiss Direct Democracy"))
• Verification without surveillance: Stand in a room with 3 others for an hour. 4 times a year. Your phones confirm you were there. No iris scan. No camera. No database. Physics does the work. ((Applebee & Combe, 2026, "Be In The Same Room"))

None of this is primitive. All of it is sovereign.

---

The fence comes down

The smartness mandate says: become smart or go extinct.

We say: become free. The technology is the same. The ownership is different.

The invisible fence is made of convenience. You can see it now. That's the first step. The second step is realising the gate was never locked --- you just couldn't see the fence to know there was a gate.

Walk through it.

---

# References

Primary Sources

Political Economy and Institutional Design

Bashir, O. S. (2015). Testing inferences about American politics: A review of the "oligarchy" result. Research & Politics, 2(4).

Gilens, M., & Page, B. I. (2014). Testing theories of American politics: Elites, interest groups, and average citizens. Perspectives on Politics, 12(3), 564--581.

Hardin, G. (1968). The tragedy of the commons. Science, 162(3859), 1243--1248.

Hayek, F. A. (1945). The use of knowledge in society. American Economic Review, 35(4), 519--530.

Marx, K. (1867). Das Kapital, Vol. 1. Verlag von Otto Meissner.

Marx, K. (1875). Critique of the Gotha Programme.

Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system. https://bitcoin.org/bitcoin.pdf

Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press.

Rawls, J. (1971). A Theory of Justice. Harvard University Press.

Scott, J. C. (1998). Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press.

Winters, J. A., & Page, B. I. (2009). Oligarchy in the United States? Perspectives on Politics, 7(4), 731--751.

Wright, E. O. (2010). Envisioning Real Utopias. Verso.

Western Liberal Tradition

Berlin, I. (1969). Two concepts of liberty. In Four Essays on Liberty. Oxford University Press.

Friedman, M. (1962). Capitalism and Freedom. University of Chicago Press.

Jefferson, T. (1776). Declaration of Independence.

Locke, J. (1689). Two Treatises of Government. Awnsham Churchill.

Madison, J. (1788). Federalist No. 51. The Federalist Papers.

Nozick, R. (1974). Anarchy, State, and Utopia. Basic Books.

Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations. W. Strahan and T. Cadell.

Communist and Collectivist Tradition

Bookchin, M. (1982). The Ecology of Freedom. Cheshire Books.

Chinese Communist Party (2006). Decision on Building a Harmonious Society.

Engels, F. (1880). Socialism: Utopian and Scientific.

Kropotkin, P. (1902). Mutual Aid: A Factor of Evolution. William Heinemann.

Mao, Z. (1937). On Contradiction. In Selected Works of Mao Tse-tung, Vol. 1.

Proudhon, P.-J. (1840). What Is Property?

Xi, J. (2017). Report to the 19th National Congress of the Communist Party of China.

Ethical Humanism

American Humanist Association (1973). Humanist Manifesto II. The Humanist, 33(5), 4--9.

Grayling, A. C. (2003). What Is Good? The Search for the Best Way to Live. Weidenfeld & Nicolson.

Kurtz, P. (1988). Forbidden Fruit: The Ethics of Secularism. Prometheus Books.

Lamont, C. (1997). The Philosophy of Humanism (8th ed.). Humanist Press.

Norman, R. (2004). On Humanism. Routledge.

Neuroscience and Empathy

Carter, C. S. (2017). The oxytocin-vasopressin pathway in the context of love and fear. Frontiers in Endocrinology, 8, 356.

De Waal, F. (2009). The Age of Empathy: Nature's Lessons for a Kinder Society. Harmony Books.

Decety, J., & Jackson, P. L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3(2), 71--100.

Gallese, V. (2001). The 'shared manifold' hypothesis: From mirror neurons to empathy. Journal of Consciousness Studies, 8(5--7), 33--50.

Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biological Psychiatry, 54(12), 1389--1398.

Iacoboni, M. (2009). Mirroring People: The Science of Empathy and How We Connect with Others. Picador.

Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673--676.

Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169--192.

Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York Academy of Sciences, 1156(1), 81--96.

AI Cultural Bias

Cao, Y., Zhou, L., Lee, S., Cabello, L., Chen, M., & Hershcovich, D. (2023). Assessing cross-cultural alignment between ChatGPT and human societies. arXiv:2303.17466.

Durmus, E., et al. (2024). Towards measuring the representation of subjective global opinions in language models. arXiv:2306.16388.

Johnson, R. L., Pistilli, G., Menedez-Gonzalez, N., et al. (2022). The ghost in the machine has an American accent: Value conflict in GPT-3. arXiv:2203.07785.

Tao, Y., Viberg, O., Baker, R. S., & Kizilcec, R. F. (2024). Cultural bias and cultural alignment of large language models. PNAS, 121(32), e2405470121.

Wendler, C., Veselovsky, V., Monea, G., & West, R. (2024). Do llamas work in English? On the latent language of multilingual transformers. arXiv:2402.10588.

Cross-Cultural Values

Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96(5), 1029--1046.

Hofstede, G. (2001). Culture's Consequences: Comparing Values, Behaviors, Institutions and Organizations Across Nations (2nd ed.). Sage.

Inglehart, R., & Welzel, C. (2005). Modernization, Cultural Change, and Democracy. Cambridge University Press.

Schwartz, S. H. (2012). An overview of the Schwartz theory of basic values. Online Readings in Psychology and Culture, 2(1).

Media Bias and Mutual Misperception

Cunningham, E., Saich, T., & Turiel, J. (2020). Understanding CCP Resilience: Surveying Chinese Public Opinion Through Time. Ash Center for Democratic Governance and Innovation, Harvard Kennedy School.

De Gruyter (2022). News exposure and Americans' perceptions of China.

Jia, W., & Lu, F. (2021). US media's coverage of China's handling of COVID-19. Global Media and China, 6(1).

Peng, W., Chen, M., & Wu, X. (2023). Discursive use of stability in New York Times' coverage of China. Humanities and Social Sciences Communications, 10(671).

Smartness Mandate and Data Colonialism

Couldry, N., & Mejias, U. A. (2018). Data colonialism: Rethinking big data's relation to the contemporary subject. Television & New Media, 20(4), 336--349.

Gerber, A., & Atalay Franck, O. (2024). Architectural intelligence in the age of artificial intelligence. In A. Gerber, O. Atalay Franck, & M. Mieskes (Eds.), Architectural Intelligence in the Age of Artificial Intelligence (pp. 11--28). transcript Verlag.

Halpern, O., & Mitchell, R. (2022). The Smartness Mandate. MIT Press.

Kaeser, E. (2018). Trojanische Pferde unserer Zeit: Kritische Essays zur Digitalisierung. Schwabe Verlag.

Martin, C. G. (2024). The simulation of intelligence and creativity: On the foundations of machine learning. In A. Gerber, O. Atalay Franck, & M. Mieskes (Eds.), Architectural Intelligence in the Age of Artificial Intelligence (pp. 53--69). transcript Verlag.

Mignolo, W. D. (2011). The Darker Side of Western Modernity: Global Futures, Decolonial Options. Duke University Press.

Negueruela del Castillo, D., & Neri, I. (2024). Beyond machine perception: AI urban imagination. In A. Gerber, O. Atalay Franck, & M. Mieskes (Eds.), Architectural Intelligence in the Age of Artificial Intelligence (pp. 141--153). transcript Verlag.

Pasquinelli, M. (2019). How a machine learns and fails: A grammar of error for artificial intelligence. spheres: Journal for Digital Cultures, 5, 1--17.

Pasquinelli, M. (2023). The Eye of the Master: A Social History of Artificial Intelligence. Verso.

Design Philosophy

Alexander, C. (1977). A Pattern Language. Oxford University Press.

Papanek, V. (1971). Design for the Real World. Pantheon Books.

Simon, H. A. (1996). The Sciences of the Artificial (3rd ed.). MIT Press.

OMXUS Research Series

OMXUS Research Initiative (2026a). Geographic birthplace as a predictor of primary language: A cross-national observational study. Preprint.

OMXUS Research Initiative (2026b). Consensus, distillation, and trust: On the mathematics of agreement in machines, networks, and people. Preprint.

OMXUS Research Initiative (2026c). The environmental determination of complex behaviour: Evidence from language, cultural practice, religious affiliation, and machine learning bias. Preprint.

OMXUS Research Initiative (2026d). The Constitution of Resonance: A Policy for Agency and Harmony. Internal governance document.

---

# Appendix A: The Empathy Equation

The OMXUS governance protocol includes a formal mechanism for structured empathy --- the empathy invitation (EI), also referred to as ViewSwap. This is not metaphor. It is encoded in the governance protocol.

Core Formula

Governing objective:

pi* = arg max_pi E[ SUM_{h in H} Pwr_h - lambda * #EI ]

Where:

• Pwr_h = personal power/agency of human h (capacity, freedom, meaningful choice)
• #EI = count of empathy invitations (ViewSwaps) in evaluation horizon
• lambda = friction coefficient (minimising unnecessary swaps)

Empathy trigger (permissionless):

For all i,j,t: Trig(i,j,t) => exists t' in [t, t+delta]: EI(i  j, t', 168h)

Any token-holder can trigger a 168-hour (7-day) ViewSwap with any other token-holder. Within delta days, the swap occurs --- a structured exchange of context, permissions, and daily experience. Cannot be vetoed.

The system optimises: maximise individual agency (Pwr_h) while minimising the need for empathy interventions (#EI). When the system works well, few swaps are needed. When it fails someone, they can invoke the swap --- and the person affected by the policy must live under it.

Cross-Ideological Legibility

Table A1: How the empathy equation is read by both traditions.

The equation is not just design --- it is a proof that empathy can be weighted, formalised, and made into governance.

---

# Appendix B: Cross-References to the OMXUS Research Series

This paper is part of a 33-paper research series. Below is a map of how the Ideological Rorschach research connects to the broader body of work.

Direct Dependencies

Related OMXUS Research Directories

Conclusions This Paper Supports

---

# Appendix C: Annotated Bibliography

Ethical Humanism

The argument that OMXUS satisfies human needs regardless of ideological framework is grounded in ethical humanist philosophy --- the tradition that derives ethics from human experience and dignity rather than ideological doctrine.

Kurtz, P. (1988). Forbidden Fruit: The Ethics of Secularism. Prometheus Books.

• Argues ethics can be grounded in human flourishing without metaphysical commitments
• Relevant to "wound-driven design": ethics derived from experienced harm, not theory

Lamont, C. (1997). The Philosophy of Humanism (8th ed.). Humanist Press.

• Defines humanist ethics as "concerned with the happiness, freedom, and progress of all humanity"
• Maps to OMXUS's "nobody and everybody" answer

Norman, R. (2004). On Humanism. Routledge.

• Contemporary defence of humanist ethics across political traditions
• Chapter 4 on "Ethics without God" relevant to non-ideological system design

Grayling, A. C. (2003). What Is Good? The Search for the Best Way to Live. Weidenfeld & Nicolson.

• Survey of ethical traditions finding common ground in human welfare
• Supports claim that both traditions share hidden agreement on goals

American Humanist Association (1973). Humanist Manifesto II. The Humanist, 33(5), 4--9.

• Declaration that "ethics are autonomous and situational, needing no theological or ideological sanction"
• Direct precedent for systems that transcend left/right categorisation

Mirror Neurons and the Neurobiology of Recognition

The "Rorschach effect" --- where both traditions recognise OMXUS from within their own frame --- may have neurobiological basis in mirror neuron systems that enable cross-cultural empathy and intention-recognition.

Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169--192.

• Foundational paper on mirror neurons and action understanding
• Humans neurologically mirror observed actions/suffering regardless of cultural frame

Gallese, V. (2001). The 'shared manifold' hypothesis: From mirror neurons to empathy. Journal of Consciousness Studies, 8(5--7), 33--50.

• Argues mirror neurons create "shared manifold" between self and other
• Explains why wound-driven design resonates across ideological boundaries

Iacoboni, M. (2009). Mirroring People: The Science of Empathy and How We Connect with Others. Picador.

• Popular science account of mirror neuron research
• Chapter on "broken mirrors" relevant to institutional failure recognition

De Waal, F. (2009). The Age of Empathy: Nature's Lessons for a Kinder Society. Harmony Books.

• Evolutionary basis for cross-group empathy
• Argues empathy is pre-ideological; supports "deeper pattern" section

Decety, J., & Jackson, P. L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3(2), 71--100.

• Neural systems underlying empathy function across cultures
• Provides mechanism for cross-ideological recognition of harm/fairness

Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York Academy of Sciences, 1156(1), 81--96.

• Reviews evidence for empathy as universal human capacity
• Relevant to claim that both traditions' "actual concerns" are recognised

Oxytocin, Trust, and Vouching

Carter, C. S. (2017). The oxytocin-vasopressin pathway in the context of love and fear. Frontiers in Endocrinology, 8, 356.

• Oxytocin (OT) in safe contexts enables social bonding
• Vasopressin (VP) dominates under threat, promoting vigilance
• Perceived emotional context determines OT/VP balance

Kosfeld, M., et al. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673--676.

• Classic "trust game" experiment
• Intranasal OT significantly increased investors' willingness to trust
• OT helps humans overcome natural aversion to uncertainty about others

Heinrichs, M., et al. (2003). Social support and oxytocin interact to suppress cortisol. Biological Psychiatry, 54(12), 1389--1398.

• Social support combined with OT reduces stress responses
• Relevant to OMXUS vouching events as trust-building mechanisms

Cross-Cultural Values

Hofstede, G. (2001). Culture's Consequences (2nd ed.). Sage.

• Individualism/collectivism as primary cultural dimension
• Framework for understanding Western vs Eastern value systems

Schwartz, S. H. (2012). An overview of the Schwartz theory of basic values. Online Readings in Psychology and Culture, 2(1).

• Ten universal human values across cultures
• Suggests underlying human needs that both traditions serve

Inglehart, R., & Welzel, C. (2005). Modernization, Cultural Change, and Democracy. Cambridge University Press.

• World Values Survey data on cross-cultural value shifts
• Evidence that underlying values converge despite surface ideology

Graham, J., Haidt, J., & Nosek, B. A. (2009). Moral foundations theory. JPSP, 96(5), 1029--1046.

• Moral foundations theory: different groups weight different foundations
• OMXUS may satisfy multiple moral foundations simultaneously

AI Cultural Bias Research

Cao, Y., et al. (2023). Assessing cross-cultural alignment between ChatGPT and human societies. arXiv:2303.17466.

• Empirical measurement of cultural values embedded in LLMs
• Methodological precedent for using AI responses as cultural indicators

Durmus, E., et al. (2024). Towards measuring the representation of subjective global opinions in language models. arXiv:2306.16388.

• LLMs reflect training data demographics
• Supports claim that Claude/DeepSeek differences reflect cultural training

Wendler, C., et al. (2024). Do llamas work in English? arXiv:2402.10588.

• Multilingual models have implicit cultural biases
• Relevant to Chinese vs Western AI system comparison

Johnson, R. L., et al. (2022). The ghost in the machine has an American accent. arXiv:2203.07785.

• GPT-3 reflects American liberal values
• Supports baseline assumption about Western AI training bias

Tao, Y., et al. (2024). Cultural bias and cultural alignment of large language models. PNAS, 121(32), e2405470121.

• Systematic study of LLM cultural alignment
• Methodological framework for AI-as-cultural-proxy analysis

---

Last updated: March 2026

OMXUS Research Initiative --- research@omxus.com