Artificial Intelligence: Throwing my Two Cents (or Bytes) in the Pond (or Stream)

A Conversation

The other day, I caught up with a dear friend, and our chat quickly turned to the implications of AI. And let me tell you, we covered a lot of ground. So, grab a drink and settle in for a free-flowing stream of consciousness (or unconsciousness, depending on how you look at it).

Fields and Feudal Lords

One of the earliest instances we discussed was that of the humble peasants who made up the majority of human history. Fiefs were subjected to a feudal administration, where they toiled the land they were beholden to their lords for. Therefore, one of the most groundbreaking creations during this period was the animal-drawn cultivator, which partly substituted the manual hoe. Just imagine the elation of a peasant who was given access to such a marvel. It's easy to picture a "griddy" being hit, as the kids these days say.

Suddenly, I, as a humble medieval peasant, could travel to distant places and stop by taverns for rest. All because I had a DIY handcart to relieve myself of the heavy load I could be carrying. Maybe this was the medieval internet, a way for me to connect with fellow travelers on the road.

From Farming to Factories

During the Industrial Revolution, manufacturing was transformed by groundbreaking inventions such as the Spinning Jenny, created by James Hargreaves in 1764, and the James Watt engine, which powered machinery in cotton production.

In 1793, Eli Whitney patented the Cotton Gin, which was capable of producing cotton fiber at an even faster rate. This expedited the demand for cotton in the textile industry. Suddenly, we could produce cotton faster than a cheetah on roller skates, which, in turn, fueled America's cotton production. Despite these benefits, society had to regulate and standardize the use of these physical tools to address their associated effects. This makes me think about the current landscape of AI generated visuals (imgs, gifs, vids, etc...).

However, the abundance of cotton led to a price collapse, which prompted the "Gone to Texas" (GTT) movement some decades later. Essentially, a large number of farmers from the antebellum South ran away, settling in Texas. Economic factors such as debt and the collapse of the cotton market led to this.

Web's World Wide Wonderland

Even in the early days of the internet, we saw both progress and stagnation. That's right, folks, it's what I like to call "more of the same." Not the most popular phrase, but hear me out.

The internet made it possible for us to communicate across the entire world (thank you, underwater cables). As of 2022, 66% of the world is online. However, as with any new development, human nature had to rear its head, and the dot com bubble of the late 90s was a prime example of our tendency to get carried away.

For as much as the internet has the potential to bring together different cultures and backgrounds, it can also perpetuate discrimination and inequality. And let's not forget about anonymity and privacy. They can be double-edged swords, leading to a lack of accountability and responsibility.

But that's not all. The internet is a vast sea of content, yet it can still leave us feeling like we're drowning in sameness and conformity. And while it's supposed to be a platform for self-expression, it can also lead to self-censorship and social conformity.

Bubbles: The High School Clique of the Internet

Numerous echo chambers tailored to every possible demographic or interest out there exist, much like the jocks and cheerleaders of old. That's where the internet came in — it's made it even easier to find and connect with those who share our views and values. Except, in today's world, we have subreddits and facebook groups. It's like the jocks and cheerleaders have been replaced by keyboard warriors and meme lords.

With the rise of AI, these complexities will only be highlighted further. Unless we tackle the linguistic divide that separates us (don't worry, I'll explain later), we might just be stuck in a rut of "more of the same."

And let's not forget the early adopters of the internet. While the rest of the world was still trying to figure out what an email was, these hobbiest pioneers were busy sharing cat photos, making memes, and cracking jokes (some of which were a little tasteless, but hey, that's the internet for you).

They were also dealing with the fear fueled by news reports warning of potential abuse - kind of like what's going on with chatGPT right now. Reminds me of Technophobia to an extent. And this.

Lingual Lingo: The Language of Interpretation

Language is not just a means of communication, but also a tool for creating connections and understanding between individuals, and this is particularly evident in the use of idioms and other linguistic expressions. Understanding the precise tone and intention behind a speaker's words can sometimes avert tragedies. Idioms are a classic example of this; they're like secret handshakes that let another group know that you're one of them. And if they respond positively, well then hot damn, we've got ourselves a zinger!

Mandarin, uses phrases that might not seem significant on the surface, but they carry a wealth of national history and ideas. We could describe this as colloquialism or jargon instead of idioms, but in essence, it amounts to the same thing.

16th Century's Grisliest Grapple

Historians love to cite the meeting between Hernan Cortes, the Spanish conquistador, and Montezuma II, the Aztec ruler. The two were strangers to each other, with no prior knowledge or understanding of their respective cultures. During their first meeting, Cortes wanted to embrace Montezuma passionately, but he was restrained by Montezuma's group, and they instead bowed to each other.

This meeting had significant implications, as the Aztecs had never encountered Europeans before, and vice versa. The Aztecs were totally freaked out by the Spanish's horses (never seen those before!) and their fancy, high-tech weapons. Meanwhile, Cortes was blown away by the Aztecs' wealth and civilization. As a European, he probably had not seen such an immaculately pristine city outlook. However, neither side had any ill intentions towards the other.

Cortes had two translators with him during the meeting. The first was a prisoner woman named Malinche, who knew the Aztec language, Nahuatl, and Mayan. The other translator was a priest named Geronimo del Aguilar, with whom Cortes talked back in Spanish. The translation was, therefore, a chain of ideas from one person to another, with the longer the chain, the more information was lost.

It was almost like the classic game of "Chinese Whispers" (or "Telephone," for my American friends) - you know, that game where one person whispers a phrase into someone's ear, who then passes it on to the next person, and so on, until the final recipient inevitably hears something completely different.

Unfamiliarity's Subtle Complexities

You see, after an extravagant dinner in the palace, Montezuma decided to give a speech. What started as a simple exchange of words quickly spiraled into a comical display of linguistic tomfoolery. TOMFOOLERY, that wonderful, whimsical word that tickles the tongue and the funny bone.

As amusing as Montezuma's linguistic tomfoolery may have been, it was not lost on the Spaniards who found something even more appealing in his lengthy monologue:

That Cortes and his band were divine beings fulfilling the prophecy of a banished god's return from the east. The cherry on top? Montezuma himself was expected to surrender to Cortes. It's hard to fathom what must have gone through Cortes' mind as he left that meeting to kidnap Montezuma. Perhaps something along the lines of "We're about to go all-in at Pochinki." That, my friends, was a true Hotel Trivago moment.

Similar to how idioms can serve as a subtle way to convey meaning, there are also other linguistic expressions that allow for the expression of reverence without explicitly stating it.

Nahuatl, the language of the Aztecs, relied on this type of speech to communicate.

Sehrgei, Tell ush the Russhian Shtory of the "Yurodivy", Da?

Sergei is nowhere to be found, but at least we have the most stereotypical Russian name to keep us company. But that brings to mind the Russian Folklore of the "Holy Fool."

The Holy Fool is an outcast and symbolic paradigm within the veins and arteries of society. But the Holy Fool has access to "the heart". The truth. In the story, there is a painting of the Virgin Mary and while everyone else admires it. Our smart and bravado child, Sergei, doesn't. He claims its' the work of Satan. The devil.

Everyone around starts channeling their inner "Captain America Bucky Winter Soldier" moment, flipping tables and scratching their necks like they're in dire need of a fix. To them, the claim was as baseless as a Jenga tower in an earthquake.

Nevertheless, at that very same instant someone throws a stone at the work of art and the illusion is broken. Behind the painting people adored, the face of the devil is revealed.

In the same vein, the boy from Hans Christian Andersen's, "The Emperor's New Clothes" is a Holy Fool himself. The king is told that his new robe is magical by some swindlers posing as merchants. And told that those who do not notice the new garment are incompetent at their jobs. Nobody points out that he's actually naked. With the exception of a naive child, oblivious to societal taboos. The boy sees what the experienced can't - he notices the robe, an irony that can't be ignored.

This may highlight how we, as adults, lose our sense of the innate curiosity of how children view the world. It's like watching a movie and developing many interpretations of certain plot lines. The more exposure we have to those genres, the more we come to expect certain beats in the story. Perhaps it is our brain trying to do less work as we age.

By acknowledging the king's nudity and pointing it out despite the societal taboo, the child challenges the expectation that the king must always be seen as regal and perfect, regardless of reality.

Verbose Past

Let's discuss the word "shit." If you lived in the upper-middle class Victorian society, merely uttering this word would have been enough to make the stiffest of upper lips quiver with shock and horror. It was seen as lowbrow and vulgar, reserved only for the uncultured plebeians. In fact, some cultures still view certain expletives as taboo, while others take a more laissez-faire (cue my best french accent) approach depending on the person's education level and global awareness. For those of us who fancy ourselves "distinguished and educated" (read: snobs), we tend to replace such base language with fancier euphemisms, such as "fecal matter" for "shit" or "nalga" for "culo". But there's a subtle nuance to language that only a select few can truly grasp (and unfortunately, I'm not one of them) - it's like trying to navigate the tumultuous seas of translation that Cortes's linguistic pathfinders discovered.

The Lexical Decay: When Words Go Rogue

Language changes over time. For example, everyday terms may gain new meanings through a "euphemism cycle." Now, I'm not one to stir the pot, but the name "Chinese Whispers" might rub some people the wrong way. It implies that the Chinese language is incomprehensible or challenging to understand. Maybe language is just a perpetual way to bring attention to problems we haven't solved.

Now, back to our boy, Montezuma

He was the master of false humility, a skill he and his peers considered the pinnacle of class in good-nature. Maybe Cortes thought the same when he tried to affectionately embrace Montezuma in the beginning. Who knows? But the practical terms were clear: in Nahuatl, the word "noble" is nearly identical to "child" (as stated on page 20 of Malcom Gladwell's book, "Talking to Strangers: What We Should Know about the People We Don't Know"). Montezuma spoke in code, referring to himself as weak and small, but those close to him knew he was really flaunting his power and the center of high-falutin' status.

The point is that overlooking the nuance of language eliminates the soul we find in human compassion. We might even lose sight of what's really being conveyed, just like Malinche and Geronimo del Aguilar must have felt.

AI and the Moral Time Warp

As society evolves, so do its values. But how do we evolve along with them? There still exist plenty of stereotypes based on religion or geographical location, opposing everyday living.

Today, Sweden is recognized for its liberal attitude towards sex and contraception, but back in the 1930s and even the 60s, things were vastly different. Up until the 1960s, abortion was illegal in most cases, but now it isn't considered a controversial issue in society.

Cultural values are frequently shared worldwide. For example, the "typical conservative wife" who is responsible for taking care of both husband's parents and the children is prevalent in several Southeast Asian countries. Who's to say that if I had been raised in that environment, I wouldn't have embraced those values? It's like the age-old nature versus nurture debate, isn't it?

Even though some men label these types of values as "Asian values," women may have a different perspective. They may find these cultural norms unbearable and be less inclined to get married.

Some of these values will fade away, as they have in modern-day Sweden. This is just another aspect of the human-centered complexity that AI will have to contend with, much like the internet did before it.

Joe-normous Quest for knowledge

It's a peculiar situation we find ourselves in, as we heavily rely on democratized sources of information, much like during the internet boom. But how do we verify it? We have to fall back on the old-school ways historians have used since time immemorial - primary and secondary sources, cultural context, and so on. Speaking of which, in the future, we may have validated sources of information that can be used to train models to prevent misuse.

We can even design models that flag potential biases or inaccuracies. By citing sources or linking to popular ideas it found, AI can enhance transparency and facilitate critical evaluation of information.

Incorporating Global Perspectives in Model Interpretation

Maybe we can encourage users to expose our models to cultural ideas and nuances from around the world, but I personally believe that we should approach this "leap" with great caution due to the possible misinterpretation of cultural nuances (or human experiences).

I can't help but wonder, as time progresses, the potential trade-offs between increased privacy and accessibility to AI-based knowledge. Whether instant access to information might eventually lead to its devaluation either intrinsically or not. What are the potential psychological consequences of constant knowledge availability? Would this lead to complacency? Or a lack of critical thinking or motivation to learn? Maybe this is the same questions poised when the internet blew up?

Language as a Lens: Capturing the Multifaceted Human Experience

Some people might not are not able able to capture as much of the human experience as possible because of the inherent restrictions they face to participate on the world stage with their ideas.

As an analogy, language can be compared to a camera lens, which captures and filters reality in a certain way. And just as different lenses can capture different aspects of a scene, different languages can capture different aspects of the human experience.

But there are some people who are limited in the lenses they can use. They may only have access to a single lens, or their lens may be distorted or obscured in some way. And just as a photographer with limited or distorted lenses will struggle to capture the full beauty of a scene, people with limited or distorted access to language will struggle to capture the full richness of the human experience. This works in both ways-- learning a language. And mastering one you already speak.

But paradoxically, this is exactly what makes a human language unique. The very fact that language is imperfect and limited, that it can never fully capture the depth and complexity of human experience, is what allows it to be so maleable. Like a sculptor chiseling away at a block of marble, we use language to create meaning and shape our perceptions of the world around us. And just as no two sculptors would create the exact same statue, no two individuals will use language in exactly the same way. In this process of creating and recreating language is when new connections between seemingly unrelated ideas bring innate curiosity. So while it may seem limiting, the inherent imperfections of language (lenses) are ultimately what allow us to express ourselves (scenes) in ways that are unique.

Talking Different: Why Culture Matters Most

It's like exploring a remote and far off village outside the capital of a country. Inevitably, different communication styles will appear. Due to the isolation those very same people were brought up in. But because of it, they have developed their own unique and distinct language, one that captures their experiences and worldviews in ways that a more widely spoken language cannot.

For example, English doesn't have words to describe the type of human experience of a group that another language that isn't english has to describe THAT particular human experience. It's why we borrow so much words. Saying the word correctly in the appropriate context evokes human intuition of deep mutual understanding. A certain-- je ne sais quoi if you will. Unfortunately, I cannot tell you what it is, for the very definition of je ne sais quoi is the inability to put a finger on it. In the same way, limitations and restrictions on language use experienced by certain groups can lead to new ways of communicating, ones that may not be fully understood by those outside the group (in the capital, because they don't have a word for it).

As such, this is why I think AI advancements will create, "more of the same" (as I've taken to calling it) scenerio. We'll have to walk a fine line between these competing demands.

This brings me to the next question: what are the implications when various languages are not being given the opportunity to express their unique voice in the vast expanse of the internet? Approximately 60% of the content on the internet is in English.

Imitation is the Sincerest Form of Flattery

Depending on how easily voice and video cloning become accessible in the future, people's need for entertainment may finally be fulfilled. My intuition tells me that AI content will be tailored to individual preferences, the same will apply to internet personalities. Some viewers might think, "I wish this content creator was a little more je ne sais quoi" and finding a slight adjustment to the type of ideas or content the personality is bringing forth will exist. It brings to mind, "If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck", although it is AI generated.

There might be a responsibility to develop a system of separation, like how different sections of a museum are dedicated to specific types of art or historical periods. We might be able to do the same for entertainment created with AI and original content.

Death by Genericide: When Brands Lose Their Identity

AI advancements make me think of Nintendo and how strictly they enforce their copyright because of how much they stand to lose. Unlike Bayer Pharmaceuticals (and countless others), who failed to properly reinforce and associate their brand with their trademark product, leading to Aspirin becoming a genericized trademark, Nintendo has continued to protect their copyright, resulting in Pokemon remaining as the highest grossing media franchise of all time, earning over $100 billion since its inception in 1996, which is enough money to run the United States Military Industrial Complex for a month and a half (in 2022).

This reminds me of the potential impact of AI on the entertainment space, particularly for YouTubers (or influencers, if you prefer to call them that). AI-generated content tailored to individual preferences may gradually shift people's perception of the original content creator, resulting in a "genericized" and "canon" version of their content creator through the viewers lense because other genres or topics are uniquely covered by the AI generated video, audio, and script.

Predictions and Obstacles

Below are some of my predictions for how AI can improve culture, as well as some of the obstacles it will inevitably face.

• Sometimes AI systems can be influenced by cultural and societal biases in language and definitions, so we need to train them to identify and avoid such biases.
• AI faces some difficulties in accurately translating and interpreting languages from different cultural and societal backgrounds, so we must ensure that the process is fair and accurate.
• The use of biased data sets in AI systems has ethical implications, which is why we need transparency and accountability in their development and use.
• International organizations can help avoid cultural and societal biases in language by promoting standardization and universal definitions.
• AI can help reduce polarization and increase understanding between different cultures by identifying and correcting biased language in media and public discourse.
• AI in multilingual communication and diplomacy can help promote cross-cultural understanding and reduce conflicts caused by misunderstandings.
• AI can be used to identify and address biases in language use and communication, which can help promote diversity and inclusivity.
• The use of AI to shape language and culture raises ethical considerations about its potential risks and benefits.
• AI can help overcome cultural and societal biases by promoting universal values such as human rights.

My idea

I think what we're really trying to automate is understanding geared for different people's experiences. I asked chatGPT to use my idea for having personalized learning based on someone's own website domain.

AI automation for personalized learning based on personal websites: In the future, individuals may have their own websites or domains where they document their thoughts, interests, and experiences. AI can be used to analyze these websites and create personalized learning curriculums for individuals based on their unique interests and knowledge gaps. By tailoring the learning experience to an individual's personal website, AI can create a more engaging and relevant learning experience, increasing the likelihood of knowledge retention and application. This can help individuals to learn new skills and concepts more effectively and efficiently, while also fostering creativity and self-expression.

Final Thoughts

I don't know how to interpret the continuously evolving nature of AI. We will see in a couple of years whether the human-assisted output of AI will reflect 'more of the same' from humanity or introduce a novel way to perceive the world and share ideas across cultures.

But like everyone else, I'm excited to see where it goes.