AI & Globalization

Challenges Posed by Globalization

Scarcity of non-English datasets leads to Anglocentric cultural biases. Further, AI can accelerate digital colonization and labor exploitation. AI developed and controlled by dominant corporate and political entities in the Global North can significantly impact social well-being in the Global South. There is a long history of technological fantasies of all-seeing, all-knowing machines or artifacts, such as maps of the world to symbolize possession of lands. The idea that the most intelligent person should rule over the world goes all the way back to ancient Greece. In many cultures, intelligence denotes access to knowledge.

In the era of cartography, map-making was a pragmatic and symbolic means to gain knowledge about the world and to possess the world. If it is on one’s map, one symbolically owns it. The following seventeenth-century map of the world is one example. See a high-resolution scan of the map on the Library of Congress’s website.

How Different Cultures Approach AI

Generative artificial intelligence should be governed with care in order to enhance social justice. More importantly, we need to understand how different cultures perceive AI and so-called “intelligent” machines in unique ways. Science fiction and speculative fiction reflect (and construct) societies’ anxieties and aspiration regarding technology.

Attitudes toward AI vary around the world, because cultural assumptions about human-machine relationships are shaped by local histories and ideologies. In the United States, AI is sometimes imagined to be the ultimate God-like technology in films such as Transcendence (2014) and The Matrix (1999).

Narratives in countries such as South Korea, India and Russia sometimes thematize the idea of “catching up to” Western forms of technology led by the US and UK.

Given the cultural and linguistic diversity around the world, there are many contrasting ways to say “AI.” The English term artificial intelligence, the use of machines to emulate some aspects of human speech and thought, was coined in 1955 by John McCarthy. The catchy term is so captivating that it replaced all of the alternative terms such as —

engineering psychology
neural cybernetics
non-numerical computing
automatic programming
machine intelligence
hypothetical automata (Penn 2020: 206 fn).

The abbreviation AI was popularized by Steven Spielberg’s 2001 film A.I. Artificial Intelligence.

In Germanic languages, art becomes part of the picture, as in künstliche Intellienz in German, kunstmatige intelligentie in Dutch, and kunstig intelligens in Norwegian. The word Kunst could also mean knowledge or ability. Technology is linked, therefore, to art.

In Russian, the term for AI is Иску́сственный ра́зум [Iskússtvennyy rázum], or artificial reason. Rázum means mind or reason. The emphasis of a reasoning mind provides an interesting contrast to the general term intelligence in English that has often been debated. It also joins the tradition of anthropomorphizing technologies.

In Japan, since the 1990s, scientists have used the term jinkō chinō (人工知能, じんこちの) to denote Western machine learning models. The notion of jinkō, or artificial, is associated with Western technologies that are distinct from art, craftwork, and other types of technologies that have indigenous links within Japan. Jinkō suggests tools and machines that are produced as opposed to artworks such as sculpture that are crafted.

What makes us human? One common thread around the world is the fetishization of speech which is assumed to be what makes us human, especially in Judeo-Christian cultures. Individuals with speech impairment are sometimes thought to have low intelligence. Speech (textual or verbal) is regarded as social shorthand of intelligence and civilized status. This attitude has contributed to the proliferation of screen representations of sentient AI robots who write or talk. These representations are often accompanied with human anxieties and hope.

The question of what makes us human is asked somewhat differently in such cultures such as Japan. In the Shinto Buddhist tradition, all beings and objects have souls (kami). In the West, only humans have souls in the classic conception of mind and body. In traditional Japan, the notion of “yaoyorozu no kami” (八百万の神, eight million gods) implies there are too many deities (kami) to count; humans are not the only beings to have souls. This has important implication on a more harmonious human-nature co-existence, how one relates to their environment, and human-machine relations.

Lost in Translation

Currently, large language models remain Anglo-American centric in their linguistic foundation and cultural outlooks. These AI models miss out on “culturally specific details that can have real-world consequences.” Further, not everything is digitally represented. Only a very small portion of the collective, historical human expressions and experiences are digitally represented and accessible. Think oral culture, un-codified emotions, gestures, events not deemed worthy to be archived, and languages with fewer speakers.

Only a very small number of the 7000 languages globally are represented digitally. Mongolian, for example, is a script that digital software does not recognize and cannot process (you cannot send a text message in Mongolian). Mongolian is the only living language that is not digitally searchable.

AI in World Cinema

Without fragile flesh and blood, AI figures (in embodied or disembodied forms) are often perceived as powerful, god-like figures. Film cultures around the world approach the figure of AI differently, reflecting each culture’s obsession and aspiration. Here are a Japanese film and an American film espousing contrasting approaches to allegorizing the rise of AI.

AI Amok (dir. Yû Irie, 2020) is a Japanese science fiction film and a cautionary tale similar to its Western counterparts. However, it focuses more on the human inventor’s tribulations than on the story of the AI itself. Kosuke Kiryu develops a medical AI called Nozomi (Hope) to cure many ailments. One day it goes haywire and starts to evaluate citizens and labels those deemed worthless for termination. Here is a trailer of the film.

Terminator 3: Rise of the Machines (dir. Jonathan Mostow, 2003) focuses on–as mandated by its franchise–the rise of an artificially intelligent system that has become self-aware. The scene in this short clip depicts the fateful moment when Robert Brewster, a US Air Force general, presses y (for yes) on a keyboard in Cyber Research Systems’ headquarter to activate Skynet. He does so in order to fight a computer virus, only to inadvertently initiate the apocalypse known as Judgment Day. Skynet turns against humans, seeing humanity as a threat to its existence.

Your Turn

What comes to mind when you think of AI in English or another language? What might AI connote? What implications do words such as “intelligence” have? Begin with a few nouns and adjectives. Please elaborate on them. Illustrate your main points with concrete examples.

The term AI in English and other languages is erroneously used to refer to a single computational technology. However, it is more accurate to say AI refers to a type of social attitude and even aspiration (Cave et al, 32). AI has historically been associated with ranking algorithms that sort and categorize individuals. The word intelligence has also historically been associated with intelligence quotient (IQ) tests and used to legitimize colonial oppressions. Each culture aspires to rather different things when it comes to AI. Consequently, we have to take into consideration cultural differences and globalization when we discuss how governments and international bodies should regulate AI.

For further insights on AI and world languages, read Stephen Cave, Kanta Dihal, et al, “The Meanings of AI: A Cross-Cultural Comparison,” in Imagining AI: How the World Sees Intelligent Machines, ed. Cave et al (Oxford University Press, 2023), pp. 16-36.

AI & Globalization

Further Reading