In a podcast interview with Kara Swisher, Sam Altman, the CEO of OpenAI, talked about naming their large language model (LLM) GPT-4 using a combination of “letters plus a number” to avoid people from anthropomorphizing the AI. This has not stopped other AI companies from giving their creations human names. Naming aside, it is almost impossible to avoid using human terms to describe AI. The use of the word “agentic”, with quotation marks, points to how the development of AI is butting up against our current vocabulary. We use words that are conventionally reserved for human minds. When chatbots take time to respond to prompts, it is difficult not to label that processing of information as some form of “thinking”. When a chatbot is able to process our prompt in the way that we intended, it makes it feel like it “understands” what we are communicating. The leading issues around AI similarly use human terminology. “Hallucination” occurs when a chatbot confidently provides a response that is completely made up. A huge area of AI research is dedicated to the “alignment” problem, which according to Wikipedia, “aims to steer AI systems towards humans’ intended goals, preferences, or ethical principles.” To the uninformed, this sounds very much like civic and moral education for students.

Humans tend toward anthropomorphism. We explain things for human understanding and often anthropomorphism helps to communicate abstract ideas. Nature documentary hosts would give names to every individual in a pride of lions and lionesses, describe their fights as familial or tribal feuds, and dramatize the animals’ lives from a human perspective. The 18th-century Scottish philosopher Adam Smith uses the term “invisible hand” to describe how self-interest can lead to beneficial social outcomes. Researchers have found that anthropomorphic language can help us learn and remember what we have learned. As AIs exhibit increasingly human-like capabilities, it will be a challenge for people to not anthropomorphize them because we will use human-analogous words to describe them.

If we are not careful in delineating AI, which is ultimately a set of mathematical operations, from its human-like characteristics, we may become more accepting of using it for other purposes. One particularly tricky area is the use of AI as relational agents. The former U.S. Surgeon General, Vivek Murthy called loneliness a public health “epidemic”, this view is echoed by many. A 2019 survey by Cigna, a health insurer, found that 61 percent of Americans report feeling lonely. It is not unimaginable for people to think that conversational AI can help relieve loneliness, which the US CDC reports is linked to serious health conditions in older adults. If there is demand for such services and money to be made, businesses will meet that demand, especially since most cutting-edge AI research is conducted by commercial enterprises. In fact, there are already similar situations occurring. In Japan, owners of the Sony Aibo robot dog are known to conduct funerals for their robot companions. While the robot dogs are definitely not alive, they have touched the lives of their owners in a real way. An article in the San Francisco Chronicle reported on how a Canadian man created a chatbot modeled after his dead fiancé to help with his grief. If chatbots were to make it easier for people to feel less lonely, would it lower the effort that people put into forging real relationships with actual full human beings, which may not be as acquiescent as their artificial companions? How would human society evolve in those circumstances? As technology has been often used as a wedge to divide society, would AI drive us further apart?

Besides the more overt issues that come with anthropomorphizing AI, there may able be less perceptible changes that occur beneath our noses. Machines are tools that humans use to multiply and extend our own physical and mental efforts. Until now, the user interface between humans and machines was distinct from human communication. We turn dials and knobs, flick switches, and push buttons to operate physical machines. We drag a mouse, type into a screen, and use programming languages to get computers to do our bidding. Now, we use natural language to communicate with chatbots. For the first time in history, the medium in which we interact with a machine is the same as that of cultural communication. We may eventually come to a point where most natural language communication takes place not between humans, but with a machine. How might that change language over time? How would that change the way that humans interact with one another? In a TED talk by Greg Brockman, President of OpenAI, he joked about saying “please” to ChatGPT, adding that it is “always good to be polite.” However, the fact is that machines do not have feelings — do we dispense with courtesies in our communication with AI? If we continue to say “please” and “thank you”, are we unwittingly and subconsciously anthropomorphizing AI?

Perhaps we need to expand our vocabulary to distinguish between human and AI behavior. Instead of using quotation marks, perhaps we could add a prefix that suggests the simulated nature of the observed behavior: sim-thinking, sim-understanding, sim-intentions. It does not quite roll off the tongue, but it may help us be more intentional in our descriptions. In response to an interviewer’s questions about how LLMs are “just predicting the next word”, Geoffrey Hinton, a pioneer in AI research, responded, “What do you need to understand about what’s being said so far in order to predict the next word accurately? And basically, you have to understand what’s being said to predict that next word, so you’re just autocomplete too.” Hinton got into AI research through cognitive science and wanted to understand the human mind. His response just goes to show how little we comprehend whatever happens in our heads. Hopefully, AI can someday help us with this. The tables might flip and we may see AI as our reflection — maybe we find out sim-thinking and thinking are not that different after all — if we survive the AI upheaval that is.

In the previous two parts of the essay, I’ve discussed how our senses and mind could limit our ability to understand the world. I will be concluding this three-part essay by turning my focus to culture. First, a working definition of culture: “The arts and other manifestations of human intellectual achievement regarded collectively.” This is one of the broader versions of the word, which encompasses all collective human creation (including technology) and across different geographical areas. 

No man is an island. I think it is important to state the significance of this, even though it seems plainly obvious. All of our thoughts are shaped by prior thinking conceived by someone else. For instance, when we try to communicate and manifest abstract thoughts and feelings verbally, we use words that we did not invent. When collectively aggregated, the whole of this precedent thinking is equivalent to culture. 

One approach to wrap our heads around this is structuralism, which began in the early 20th century (unsurprisingly) within the field of linguistics. Structural linguists realized that the meaning of a word is dependent on how they relate to other words in the language. Earlier, we defined the word “culture” using a string of other words. Every word is defined by other words. We can imagine language as a network of relationships between words. The implication of this is that a word has no meaning on its own, except where it fits structurally in the system. Over time, this idea became applied in other fields like anthropology and sociology, notably by figures like Claude Lévi-Strauss. Structuralism then became a “general theory of culture and methodology that implies that elements of human culture must be understood by way of their relationship to a broader system.” Structuralism, simply put, is an approach to understanding cultural “phenomena using the metaphor of language.”

The structuralist approach can be similarly applied to what we think, feel, know and understand. Coming back to the main thesis of this essay — what and how we understand is shaped and limited by culture. Several thinkers have explored this in their own ways. Zeitgeist, a German word that literally translates as “time spirit” (or less clunkily, “spirit of the time”) is a term that is commonly associated with Hegel. The term is defined as “the defining spirit or mood of a particular period of history as shown by the ideas and beliefs of the time.” This shows that there is an acknowledgment of how certain ideas and beliefs are bound to a specific time at least since the 1800s. Marx later built upon the idea with the bedrock concepts base and superstructure. He defined base as the economic production of society and superstructure as the non-economic aspects of society, like culture, politics, religion and media. (Do note that my definition of culture includes both base and superstructure, but we can continue for the time being.) Marx’s thesis is that products of culture (superstructure) are shaped by means of production (base). This, to some extent, was built on Hegel’s zeitgeist and explains how and why ideas and beliefs change over time. 

The two (similar) concepts that are most relevant to this essay came later. The first is episteme, coined by Michel Foucault. The second and perhaps more popularly known idea is paradigm (shift) by Thomas Kuhn. In Foucault’s book, The Order of Things, he describes episteme: “In any given culture and at any given moment, there is always only one episteme that defines the conditions of possibility of all knowledge, whether expressed in a theory or silently invested in a practice.” In other words, Foucault claims that the episteme sets the boundaries of what can be even thought of by individuals of a culture – a sort of ‘epistemological unconscious’ of an era. Kuhn, a historian of science, described paradigm shift in his book The Structure of Scientific Revolutions, as “the successive transition from one paradigm to another via revolution” and claimed that it “is the usual developmental pattern of mature science.” While Kuhn used the term purely within the scientific context, it has become more generally used over time. Examples of scientific paradigm shifts include the Copernican Revolution, Darwin’s theory of evolution and, more recently, Einstein’s theory of special relativity. Each of these shook the scientific establishment of the time and, in the case of the former, resulted in banned books and Galileo’s house imprisonment. We can see from the first two examples that society can be resistant to change, despite overwhelming evidence. This further cements the notion that ideas can sometimes be too far beyond what can be accepted by predominant culture. 

Culture shapes and, therefore, limits our understanding in a variety of ways. Culture defines who gains access to knowledge and understanding. According to UNICEF, only 49% of countries have equal access to primary education for both boys and girls. The numbers only get worse higher along the educational pathway. The gender disparity in education can be traced back to gender stereotypes and biases. Such implicit biases extend to inaccurate and unfair views of people based on their race, socioeconomic status and even their profession. They are insidiously absorbed through experience based on the social norms of our time and go undetected unless they are specifically made conscious. A form of philosophy and social sciences known as critical theory, started by the Frankfurt School in the early 1900s, aims to free human beings from prevailing forms of domination and oppression by calling attention to existing beliefs and practices. A development known as critical race theory, which seeks to examine the intersection of race and law in the USA, has recently been facing pushback in states such as Texas and Pennsylvania through book bans or restrictions within K-12 education. In this, we see a formal restriction of understanding by culture (in the form of a public institution). Further upstream in knowledge production, research deemed to be socially taboo can be severely limited. An example is the legal contradiction faced by scholars looking into the medicinal benefits of marijuana. The issue is nicely summed up by the following sentence from this article by Arit John: “marijuana is illegal because the DEA says it has no proven medical value, but researchers have to get approval from the DEA to research marijuana’s medical value.” 

Beyond such visible examples, I think it is important to emphasize that a majority of how our individual understanding is shaped by the culture we are embedded in is hidden in plain sight. It is only in retrospect that misguided views and practices may seem obvious today. Up until the 1980s in the UK, homosexuality was a mental disorder treated by electroconvulsive therapy. Homosexuality was removed from the World Health Organisation’s International Classification of Diseases (ICD) only in 1992. Besides comparing cultural attitudes with those from the past, they can also be identified through intersubjectivity by comparing different cultures. In Singapore, homosexual acts are considered illegal based on Section 377A of the Penal Code, an inheritance from its past as a British colony. Other former colonies like Hong Kong and Australia have since repealed the law. Culture implicitly and explicitly defines what is normal within a group or society. As stated by Marshall McLuhan in his book The Medium is the Massage, “Environments are not passive wrappings, but are, rather, active processes which are invisible. The ground rules, pervasive structure, and overall patterns of environments elude easy perception.” This echoes a story from a speech by David Foster Wallace in which an older fish asks younger fishes about the water, to which they later respond, “What the hell is water?” Normality is invisible in our daily lives, we do not notice it because it is the ground on which we (and all of our perceptions and thoughts) stand.

Like words, culture is self-referential. Culture shapes culture. This not only applies to how current culture gives rise to future culture but also operates in the reverse direction, where today’s culture can be used to look at yesterday’s culture. This reminds me of how the art critic Jerry Saltz says in this lecture that “all art is contemporary art because I’m seeing it now.” Strangely, our visions of the future and our recollection of the past are and can only be done through the filter of the present moment. To repurpose a famous quote on McLuhan by his friend John Culkin — culture shapes the understanding of individuals, and individuals go on to shape culture. It is our collective human enterprise. Talks about culture often lead to the distinction between nature and culture, which distinguishes what is of/by human beings. Funny thing is, the nature-culture discourse is itself facilitated through culture. It seems, therefore, that all understanding is filtered through culture.

As I wrap up, I would like to address some issues that have increasingly become noticeable while writing this essay. First, I have rather simplistically equated knowing and understanding when they are differentiable mental processes. Second, there seem to be different flavors of understanding, which can be mostly grouped into two categories: objective and subjective. The physical sciences fall into the former, while the humanities and social sciences seem to fall into the latter. The issue here is that interpretation seems to play very different roles in either. For objective questioning (e.g. why does an apple fall toward the earth?), there is usually a convergence towards a single theory, whereas, for subjective questioning (e.g. why do people generally think that babies are cute?), there is a divergence in different approaches to understanding a single issue (sometimes even opposing viewpoints within an approach), none of which is definitive in explaining a phenomenon. Third and finally, how much of our understanding is motivated by our perspective and how much of our perspective is derived from understanding? Perhaps I will attempt these questions in future essays.

Labels, therefore, are simplifications of usually more complex experiences. Additionally, they are ideal versions of whatever they are meant to point to. For instance, most people would say that they know what the word “black” means and will be able to identify black things in their environment. Let us say that we get someone (you can try this too) to look for one black object. Once they have found this object, they are asked to look for another black object, preferably one that is darker than the first. Now, we have two objects in front of us that are black. One of them will likely be darker than the other. By definition, the lighter black is not black, but a grey. Suppose this person repeats this process — they will likely be able to find an even darker black, rendering all previous examples grey. As of now, at least on our planet, this process will lead to the blackest material ever created, which is developed by researchers at MIT. The title was previously held by Vantablack, which caused some controversy when the British artist Anish Kapoor managed to acquire exclusive rights to it. Black is defined as “the very darkest color owing to the absence of or complete absorption of light; the opposite of white.” The only thing that is truly black in our universe, is a black hole. However, it is unlikely that anyone will ever perceive one up close unless they are interested in a one-way ticket into the darkness. However, the fact that we don’t need to perceive this true black, means that an ideal version of black already exists in our minds. Therefore, while we do experience imperfect instances of black through perception, the concept of pure black is one that is understood by the mind.

This perspective seems to echo Plato’s theory of forms, which posits that true reality exists separate from the physical reality in which we reside. In this higher reality are the ideal and perfect essences of all things, which people can access only through our thought and reason. While I do not think that such a realm exists, I do believe that in human language, the use of oppositional labels ultimately leads to the imaginary extrapolation of extremes. To put it another way, whenever we use opposite terms, they become such exaggerations of themselves that they can no longer exist in the real world. To illustrate this, we shall refer to the second half of the definition of black, which mentions “black” as the opposite of “white”. The eye perceives white when the three types of cone cells in our retina are equally stimulated by strong light. Similar to black, we will always be able to find an even brighter white, rendering every other white we have perceived up to that point as a grey. Unlike our search for the purest black, however, our quest for the brightest white will be cut short by permanent eye damage. The film director Ridley Scott once asked, “Life isn’t black and white. It’s a million gray areas, don’t you find?” To which, I would agree. Hence, strictly speaking, black and white mostly exist as ideal absolutes in our minds, while versions that we perceive in everyday life are shades of grey. 

This act of labeling applies not only to color but to every other aspect of our lives. Are people (innately) good or evil? Should societies organize themselves around capitalist or socialist economies? Should we prioritize individual freedom or the common good? Should governments be conservative or progressive? While such questions often expect one choice or the other, the actual answer is often a combination of both choices or lie somewhere in between them. We should be cautious whenever any pair of labels are presented to us as binary opposites. Oftentimes, these pairings are arbitrary and not mutually exclusive, creating false dichotomies. Moreover, I think that it is quite unrealistic to assume that the complex richness of our world can be reduced to one simplistic idea. By identifying the gradient that exists between supposed opposites and focusing our attention on appreciating subtlety rather than polarity, we can have much more productive discussions that will expand our knowledge and push us forward.

Additionally, we often look for opposites when they do not exist. Sometimes thinking in a purely binary way yields little use. Instead, we can think about how labels relate to one another and what type of space exists between or among them. Labels can be thought of as points in an indeterminate thought space (similar to the one described by Peter Gärdenfors). By connecting two of these points, we get a one-dimensional line. Sometimes, we can connect three or more of such points, creating two-dimensional planes (funny example by xkcd) or three-dimensional spaces.

To further complicate the matter, some labels that we use are social constructs. This means that the labels themselves are not fixed but are continually renegotiated within our society. One of the efforts of feminism, for instance, is to question the conventional roles of men and women. This process changes our understanding of these labels and their relation to our identity.

I shall conclude by stressing that just because our labels are simplified abstractions does not mean that they are unimportant or meaningless. Labels are useful as they help us navigate the world and distinguish different experiences and phenomena. Labels may even have a direct effect on our perception. Researchers have found that the language we speak affects the colors that we can perceive. We should just be aware that the world is a lot more complex and dynamic than the labels that we use to represent it. Contrary to my previous statements about black and white, I do not think that we should start calling things dark and light grey. We should not be paralyzed by the ideal quality of labels such that we become afraid to use them. For instance, I believe that gender is non-binary but, to echo a recent opinion piece by Nick Cohen, if I look like a man and act like a man, then maybe I should identify as a man. One of my favorite slang words, which seems to be used with increasing frequency, is “ish”. “Ish” reinjects complexity and approximation back into otherwise oversimplified categorical labels and frees us into using ideal terms in more flexible ways. Hence, I am a man(ish).