Introduction

Our culture significantly shapes in how we view certain social changes and subjects. This sociological background bleeds into our life circumstances and our values on wider societal subjects, such as AI. Literature and studies monitoring the advancements of AI have not considered the delicate relationship between one’s personal and cultural beliefs outside the scope of Western views with technology. This not only falters our ability to move forward but additionally stunts us in the realm of cultural awareness and human understanding. Additionally, the prevalence of Western ideologies pushes AI onto cultures who are not open/concerned with the matter. In particular, personal experiences within cultural contexts inform the interactions, preconceived notions, and future of individuals and groups with AI.

From an ethnographic perspective, employing scholarly articles and studies against the documentary How To Build A Human, I argue that Western views as the assumed norm ignore the cultural values and attitudes of those outside of the most “developed” countries, and that in order to move forward to a truly technologically advanced world, we must respect the opinions of those groups and individuals. By closely examining the connection between culture and the progression of AI, we are able to open the conversation to the push of Western values and technology onto regions with a more negative view towards technology and AI. We may also seek to use this knowledge towards a more respectful atmosphere and decline of Western totalitarianism.

Theoretical Framework

Western Universalism is a phenomenon that believes that values, norms, and cultural principles of Western culture are applicable to all humans and cultures, regardless of context. It is no secret many of the developers and companies behind present day AI advancements are based in the United States, however a recently discovered statistic brings new data to light.

According to the AI Ethics Lab, it is no less than 82% of the codes that are from the West, almost the same figure as for the European Council Digital Policies Framework. At the end of the day, 70% of identified documents are Western productions, while the West at large represents barely more than 15% of humanity. (Goffi, 2021)

Other tests and studies have resulted in similar findings, supporting the hypothesis that more and more of AI’s machine learning algorithms and understanding of human culture stem from Western values. As mentioned in a LinkedIn article, the World Value Survey used by psychologists was applied to a popular AI system known as Chat GPT. This survey is used to analyze the differences in beliefs and cultural values. Chat GPT’s answers most closely resembled answers given by Americans, “and very different from those of more collectivist and traditional societies.” (Cannata, 2024). It is typical to surround ourselves with people that align with our cultural background, values, and morals, however our life circumstances usually ensure we interact with those from opposing views and cultures to enrich our knowledge and compassion for those different from us. Western Universalism in AI threatens this compassion and understanding of others.

Unity is portrayed often as the end all be all of world peace and universal understanding, so what is the problem with this application into AI? This technology is becoming ever more representative of Western culture, if the majority of companies and individuals working on the latest AI technology are Western, then the only content taken in and learned will be representative of one culture. When AI learns from us, no matter our cultural background, it will of course inherit our traits, our ideologies, all including our biases and prejudices. A lack of diversity in this context will lead to the expectation of one universal set of beliefs. Furthermore, many computers are not only programmed, but they are also created with the ability to learn from what they take in around them. Whether this process is machine learning or fed datasets, these algorithms are constantly adjusting themselves and their outcomes based on what they take in. These outcomes they produce result in what we see on our screens, understand current information, and potentially our history.

The ultimate result of any learning based on one view is ignorance, which is a trait that could not separate itself further from our idea of unity. There is no way to prevent bias from creeping into what we do. When looking at this as a whole, Stuart Hall’s Encoding/Decoding theory is a great place for reflection. In summary, this theory presents the argument that media does not give its absolute meaning to an audience, rather that the audience and media producers create meaning through their interactions with the media (Department of English, 2025). This could be taken as a rebuttal to my argument in some cases, but in the presence of Western Universalism’s exclusionary actions there is little to say. When referring to this theory, I think of a butterfly effect. Subtle (or perhaps explicit in some cases) biases and prejudices are encoded within the AI technologies, which are then presented to users. Obviously, people as a majority are able to think for themselves and reflect and interpret these circumstances differently, but the end result may still lead to the spread of cultural imperialism in terms of values and ostracism of cultures not necessarily recognized by the West.   There are, however, ways to be more mindful about our biases. We must keep this in mind and remember that unity as we currently define it cannot exist in the presence of cultural biases.

…universal (in the strictest sense of the word) values do not exist [5]. If there were universally shared values, there would be no need to struggle to establish a global ethical regulation. This belief in the universality of Western values and their pre-eminence, is a form of “cultural arrogance” [14] that some societies, especially rising AI powers, will not accept indefinitely. (Goffi, 2021)

AI will favor not only the cultural preferences of its assumed audience, but it also may enforce stereotypes and assumptions about appearance. When considering the major powers currently working on it, this makes perfect yet unfortunate sense. If those teaching it look, act, and speak a certain way, that is what the program will recognize as familiar, or in some senses “right”.

As it currently stands, AI can be quite exclusionary to those who are not white men. “…facial recognition software that struggles to accurately identify people with darker skin tones, or translation algorithms that default to gendered pronouns based on Western cultural assumptions.” (African Study Group, 2024) Those who do not partake or are familiar with aspects of Western culture we view as universal will not be able to use, participate in, or have a consideration in these developing technologies that will continue to become ever more relevant to success and upkeep of our society.  As a society of humans, we are not impartial. In their discussion on the cultures behind AI, Erza states, “This thing reflects ourselves back at us and our stories about it. It’s a very non-neutral technology.” (Klein, 2023) Technology is viewed in the majority as a resource, most cultures view a material or service categorized as a resource to be universally accessible and beneficial. By moving forward with only Western workers, datasets, and other involvement, we ignore one of very few, widely accepted views we share as a societal species.

Case Study Analysis (How to Build a Human)

The documentary follows Gemma Chan and her journey alongside the team working to create a lifelike AI robot that looks, talks, and thinks like her. Gemma and the team undergo a long and intensive process of understanding what it means to be and create a human being. She goes through a series of different tests and questionnaires/ interviews to understand key components of her knowledge and personality; she also participates in activities such as getting a casting made of her face to be molded for the robot. Her role as Mia, a sentient synth in the TV show Humans gives Gemma a unique interesting perspective when it comes to being or becoming an AI. Multiple discussions are had about the ethics of this creation, the uncanny valley feelings that arise in the process, and the challenges as well as the wins. Gemma’s very being is represented in the robot created, her looks, mannerisms, personality, knowledge, and personal information. Even though the production of the robot and documentary takes place in the United Kingdom, Western influence and totalitarianism are not limited by geography.  Mannerisms, expectations, and cultural contexts displayed fall under Western societal expectations. The team’s ultimate goal is to see if they can build a human.

In order to create an AI robot that is exactly like Gemma, human characteristics must be programmed, and the robot must possess the ability to learn and adapt behavior. In simple terms, they must program an individual that can think for itself based on an existing set of data. In this process they cast a model of Gemma’s face, have her work with a computational linguist to create the robot’s voice, and closely analyze her expressions and facial movements while talking and reacting to messages. This robot is intended to be the most expressive and accurate technology presenting as closely to a human as possible. As in most programming with AI systems, robots, and other applications, our bias occurs in the individual programming as well as in machine learning data (Mukhopadhyay & Reddy, 2023). The machine learning process forces the program to digest data into a singular personality that it inhibits. At the end of this building process, Gemma’s robot is put through something called a Turing Test, which is commonly done to evaluate the machine’s likeliness to human intelligence. They execute this test by setting up a virtual call in which the robot speaks to individuals Gemma knows personally or has spoken to previously. A both successful and scary feat of this test is the robot’s initial success in fooling participants into believing they are talking to Gemma. It is revealed later or discovered by participants that she is not a real person, and they are speaking to an AI. The team in some sense has succeeded. They have created a robot that looks, acts, and thinks like Gemma, and her reaction is an equal mix of fascination and terror.

AI and Cultural Imperialism

The long-term effects of continuing this behavior without reflection could result in political extremes. Cultural imperialism being one of the most prominent, and the erasure of cultures and peoples is becoming more of a concern. The behavioral trends of ignoring the existence and significance of other cultures and ways of life also could lead to a fascist society that values not only a single value, but a single, authoritarian ruler/dictator. The motions leading us into this dangerous future are not a dystopian level of extreme, they are small, everyday allowances that can slip through our critical thinking and better judgement. In line with this, The Ezra Klein Show directs attention to the understandable reasons for our growing reliance and acceptance of AI.

And most of us, we’re not scientists. We are all used to living in a world where we don’t understand how our phone works. We trust that the guy who makes the phone knows how the phone works, so I don’t worry about it. Or I trust that the scientist that I’m reading about knows something. (Klein, 2023)

We already exist in a world where we rely on others to do jobs or skills for us. As our technology has progressed, this reliance has moved from people to computers. If we increasingly rely on those in tech to create technology without reflection on faults or flaws, we will become reliant on a system that reflects and promotes its Western values over the needs and interests of any other group. If we continue only in the interests of the West, when we ask questions of AI, such as googling a fact, or asking Chat GPT information about a subject, it will reflect only what is known from its point of view on the subject, and it only learns and serves interests from one particular area, it may completely lose or lack access to new information outside of the U.S. “This is the danger of AI driven cultural imperialism, a world where cultural diversity is slowly eroded by the relentless march of algorithms.” (African Study Group, 2024) Additionally, these AI systems may take knowledge gained from other cultures and present it as its own, essentially plagiarizing and cherry picking what it believes to be useful (Mukhopadhyay & Reddy, 2023).

The global spread of AI powered technologies is leading to the homogenization of culture. As Western values are embedded in more and more applications, there’s a risk that non-western cultures will be marginalized or even erased. Language, customs and traditions that are not reflected in AI data sets risk being left behind. (African Study Group, 2024)

The fight against Western Universalism in AI protects cultural diversity, but additionally it protects the accuracy of history and information as we know it to be. Truthful education in the presence of human nature is already a challenge for our world, especially in the West. Our technology should look to teach us the truth, not lead a further lie.

As mentioned above, our increasing reliance on AI technology has allowed this development into fields we never expected. What will happen when these programs, robots, and other executions of AI are asked to make moral decisions? A common example is the self-driving car in the scenario of prioritizing lives in the event of an accident.  Outside of their machine learning processes, is Western moral and ethical considerations really the way we should be moving forward? For the greater consideration of all peoples, I think not. It goes without saying that the culture of what is a morally just and ethical environment can be drastically different in each culture. Each country, even each state in the U.S., can have different ideas on what is right and wrong. Most typically, these are reflected in laws and other legal decisions for each group. When and if the current developments of AI begin to see themselves inserted into these fields, they will reflect the current set of Western laws and ideas on moral justice.

…ChatGPT reflects the moral views of some peculiar humans (from Western, Educated, Industrialized, Rich, Democratic Countries, and in particular from the United States). It should not be considered at any rate an ultimate authority for moral decision making. (Cannata, 2024)

As it stands, we as humans have compassion and the ability to think for those different from ourselves, and still, we find ourselves in a position where we do not truly consider the wellbeing of others different from us, not just culturally. A growing question in the field of AI ethics has been the ability to program compassion and human feeling, but human feeling is flawed. AI at its very best will still produce these flaws and continue to mirror them in our world. “…it doesn’t have agency in the way that we keep assuming that it does.” (Klein, 2023) Until we as a species can perfect our own ethical and moral decision making across the world, we should keep machinery and technology reflective of human flaws away from our most sensitive fields and problems.

Conclusion

Western totalitarianism has a tight hold on the current field of AI with many tech companies being located within the U.S. and programmed with our inherent biases, culture, and ideologies. Without consideration we may dig ourselves into a world that only favors, remembers, and expresses Western culture. How to Build a Human reflects incredible feats of our developing technology, but analyzation shows the plethora of ways things could pan out negatively without sensitive consideration. Cultural diversity keeps us strong as a whole with considerations outside of ourselves, essentially creating a diverse and accessible place that allows many different cultures and types of peoples to thrive. The introduction of cultural considerations through laws, guidelines, or other requirements could keep countries and companies in power responsible for implementing contexts outside of Western concerns. In addition, countries currently at the forefront of the AI race should present the opportunity to share their resources with other countries of groups of people inexperienced with the technology. Inaccessibility has never led to improvements under any circumstances other than the theory of eugenics, which is not how any race of beings should choose to represent themselves. Accessibility and diversity is crucial to the improvement and humaneness of developing AI. By erasing part of human nature, we lose what makes us a caring and unique species and stunt ourselves of the opportunity to create a diverse, generous, and considerate version of AI that can understand the world at large, not simply the West.

Sources

African Study Group. (2024, October 20). AI: The Hidden Propaganda Machine of Western Cultural Imperialism. [Video]. YouTube. https://www.youtube.com/watch?v=uBqJRzxm_Xs

Cannata, D. (2024, Jan 16). AI is built on Western values. And this is not necessarily a great thing. [Post] LinkedIn https://www.linkedin.com/pulse/ai-built-western-values-necessarily-great-thing-davide-cannata-eifme

Department of English. (2025, Feb 2) Stuart Hall’s Encoding/Decoding Model [Video]. Youtube. https://www.youtube.com/watch?v=1opjSuV6gl8

Goffi, E. (2021). Escaping the Western Cosm-Ethical Hegemony: The Importance of Cultural Diversity in the Ethical Assessment of Artificial Intelligence. AI Ethics Journal, 2(2). https://doi.org/10.47289/AIEJ20210716-1

Klein, E. (Host). (2023, May 2). The Culture Creating A.I. Is Weird. Here’s Why That Matters. [Audio podcast episode]. In The Ezra Klein Show. The New York Times. https://www.nytimes.com/2023/05/02/opinion/ezra-klein-podcast-erik-davis.html

Mizelas, S. (Director). (2016). How to Build a Human [Film]. Little Jem.

Mukhopadhyay, S., & Reddy, D. (2023, February 1). Artificial Intelligence as an Enabler of Western Universalism. https://www.researchgate.net/publication/372401256_Artificial_Intelligence_as_an_Enabler_of_Western_Universalism