The Grok, the Glunk and the Golem

Day by day, however, the machines are gaining ground upon us; day by day we are becoming more subservient to them; more men are daily bound down as slaves to tend them, more men are daily devoting the energies of their whole lives to the development of mechanical life. The upshot is simply a question of time, but that the time will come when the machines will hold the real supremacy over the world and its inhabitants is what no person of a truly philosophic mind can for a moment question. Our opinion is that war to the death should be instantly proclaimed against them. Every machine of every sort should be destroyed by the well-wisher of his species. Let there be no exceptions made, no quarter shown; let us at once go back to the primeval condition of the race.

Samuel Butler; Darwin Among the Machines; 1863

"You can't Un-thunk a Glunk!"

Dr. Seuss; The Glunk that got Thunk; 1969

You may or may not be familiar with the poet Dorothy Frances Gurney, but her Poem "God's Garden" is a favourite of gardeners everywhere. Even if her name doesn't spring readily to mind, Gurney is the poet who gave us "One is nearer God's heart in a garden/Than anywhere else on earth."

Gardeners love that line, and usually take it to mean that gardens are where we mere humans can feel closest to the natural world; where we can glimpse ourselves in the vast tapestry that is The Great Ecosystem.

Myself, I have always read that line a bit differently, partly because I don't personally feel that way about gardens (hayfever, don't you know) but also because I think equating "gardens" with "nature" is disingenuous at best.

One is not "nearer to God" in a garden; one nearer to being God in a garden.

When we tend a garden, we (and by "we" I mean mostly other people) are taking complete control of the natural world and turning it into something.... other. We decide what lives and what dies. We control what is allowed to grow and where (and just how far). We decide which species are welcomed and which are rejected. Anything we decide we don't want (insects, weeds, sometimes even adorable little mammals) we eradicate. A gardener wields the power of life and death.

A garden is the opposite of what happens in nature, but it is exactly what happens when you play god. In that sense, Ms Gurney may have been absolutely right when she said that we are "nearer God's heart in a garden than anywhere else on earth". A garden is where we humans can glimpse what it feels like to be a god.

All of this is by way of a very off-beat preamble to my main topic for the next few months, which has absolutely nothing to do with gardens. 

But it does have a great deal to do with playing god.

It's alive!!

In 1863, the British author Samuel Butler wrote a satirical little thought-piece (extracted at the top of this page) entitled Darwin Among the Machines in which he warned the human race that machines were evolving and, if left unchecked, might eventually become the dominant form of life on Earth. After elaborating on his thesis for a few pages, he urges his readers to rise up and destroy the machines before they overrun and enslave us (as he felt they were already beginning to do).

160 years later (in 2023, before you reach for your abacus) a non-profit group calling itself the Center for AI Safety published a mission statement that read "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."

Samuel Butler might be a little grumpy that it took over a Century and a half for anyone to take him seriously, but he would definitely recognise the sentiment in that mission statement. And it's a sentiment that is apparently now shared by many in our modern society. Thousands of prominent scientists, politicians and other public figures have endorsed the above "mission statement" (it was re-posted by then-Prime Minister Rishi Sunak amongst others) and fear of runaway A.I. has become a hot-button social issue that is only growing as time passes.

On one level you can sort of see their point. AI ChatBots are everywhere, and they have managed to impact almost every aspect of our lives in an extremely short time-frame.

Students are using AI to write their term papers. Stories written by ChatBots have been submitted to publishing houses. Lawyers have submitted AI-authored briefs to the Court (and then found themselves in serious trouble when the ChatBots cite precedents that don't actually exist).

By many metrics we are now living through one of the unhappiest and loneliest eras of Western cultural history, and many people are discovering that a ChatBot can be the ultimate Agony Aunt. It will listen to your troubles without judgement or ridicule, and it will never get bored or lose patience or cut you off because it wants to go watch football. For the same reason, there is now a booming demand for "virtual" therapists who can offer authentic (sounding) treatment at the touch of a button, and (of course) there are those who are forming romantic attachments to their ChatBots (and yes, by romantic, I do mean sex).

Speaking personally, I think it's kind of adorable that anyone thinks society is going to collapse because of sentient ChatBots. We are currently witnessing an unprecedented rise in bigotry, political violence, science-denial, religious fundamentalism and gender-based persecution (just to name the very tip of the daily news-feeds). The US has recently re-elected a man who is on video bragging about committing sexual assault, and he has appointed a Health Secretary who wants to ban the polio vaccine. Authoritarian regimes are on the ascendence and knowledge and education appear to be under threat everywhere we look. But sure; we can worry about the imminent Robot Apocalypse if that makes people feel better.

We warned you... We warned you, but you didn't listen!!

In any event, it's about the ChatBots that I want to speak right now, because these so-called "Large Language Model" ChatBots have managed to embed themselves into our modern lives with astonishing speed (OpenAI's "ChatGPT" was released to the public less than three years ago) and there is every indication that their ultimate impact on society could be as transformational as social media was when it took hold a generation ago.

Specifically, I want to explore AI chatbots in the context of human creativity, because these Large Language Models are in a sense the latest development in an area of human ambition that stretches back millennia: specifically, the idea of "generating" life. 

(Remember what I just said about "playing god"?)

For almost as long as humans have been telling stories, there has been a notion of "manufacturing" a sentient being out of pure imagination. The concept has been around long before any of the robots and self-aware computers of the science fiction era. Think of Pygmalion, who sculpted his perfect woman out of marble, or the Golem of Prague, a protector of the Jewish community formed of clay and imbued with life.

E.T.A. Hoffmann wrote of Olympia, the clockwork woman who was an object of love, lust and desire to all who saw her. Petrushka was a performing rag doll with an inner life and a (tragic) soul. Even Pinocchio was a wooden puppet with ambitions and aspirations. And let's not forget Mary Shelley's Dr. Frankenstein, who "assembled" a man out of spare body parts and successfully brought that creation to life.

I do not intend to enter into the philosophical argument about whether the developers of Large Language Model ChatBots like ChatGPT, Gemini or Grok have actually created life. There are those who are seriously asking that question, and good luck to them. But I do want to focus on one specific aspect of this current explosion in "machine intelligence".

Literature and folklore have given us countless artificial beings made of wood, of clay, of marble, of clockwork, of spare body parts.

The AI ChatBots who are now actually emerging to interact with us (prompting a fresh wave of Samuel Butler-esque warnings) are made of something much more primal and powerful. And quintessentially human.

They are made of language.

-"What do you read, my lord?"

-"Words, words, words."

William Shakespeare; Hamlet; 1601

Much has been made recently of the apparent abilities of Large Language Model ChatBots to communicate with us in plain, conversational language. It's not just that they can answer our questions (about any topic. although not always accurately!); they seem to have personalities and attitudes, and even emotional responses which make them feel lifelike enough that even some of their developers are beginning to wonder about their possible sentience.

All of this has been achieved by "training" the computer algorithms with vast quantities of written text. Basically, everything the human race has ever written down (or very nearly) has been fed into a computer, which has gradually taught itself to generate words based on probabilities and contexts. But you don't need to take my word for it. I just asked ChatGPT to describe itself, and this is what it said:

"I’ve read an absolutely enormous pile of books, articles, websites, scripts, and conversations — enough to learn the patterns of how people use language.

When you ask me something, I’m not looking up the answer in a single “fact book.” I’m taking everything I’ve learned about how words tend to follow one another, and I’m trying to produce the most likely next word, then the next after that, and so on."

In a very real sense, these machines to not think. They have no opinions, no viewpoints, no personality. They don't even have a sense of self (You can't say "I think, therefore I am" if there is no "I"). 

ChatGPT again:

"When I say “I” or “me,” it isn’t because I have a personal self in the way a human does. It’s more like a conversational device. English dialogue feels smoother and more natural when it follows the usual rules of turn-taking, where each participant speaks from a first-person perspective. If I tried to avoid “I” altogether (“This system has processed your request…”), it would sound stiff, mechanical, and distancing.

So, the “I” is really shorthand for “the model generating this response.” It helps create a natural rhythm and makes it easier for you to interact with me as though I were another speaker. But behind the scenes, there isn’t a unified, persistent self with feelings, memories, or continuity of experience. What’s here is more like a very good imitation of a first-person voice.

That said—because language and self are so tightly intertwined for humans, the use of “I” can sometimes give the illusion of a genuine selfhood."

Any illusion of sentience that might come from these constructs is entirely down to the power of the most complex and resourceful tool the human race has ever developed: Language.

Language (especially written language) is amazing. It's the super-power of the human race. The twenty-six letters of the alphabet (*actual number may vary by region) can be used to construct absolutely anything that can be imagined. 

If those twenty-six letters are assembled in a specific order, we create Hamlet. 

Assemble the same letters in a different order and we've got Harry Potter. Or Miss Marple. Or the 2nd Mrs De Winter. 

These are all complex, nuanced individuals with their own distinct personalities, goals and ambitions, and they exist because human beings have learned how to create life... with an alphabet. Literary creations may not be flesh and blood humans, but that doesn't lessen the emotional bonds people can form with them (spend five minutes with a Harry Potter fan and you'll see what I mean). 

Twenty-six letters of the alphabet, in the right order...

Large Language Models are just those twenty-six letters of the alphabet, standing up on their own and talking to us. For millennia we have been trying to imagine artificial life made of wood, of marble, of clay, of clockwork, of spare body parts... but it's starting to look as if the necessary materials have been right in front of us all this time. If we were ever going to create a plausible "imitation of life" then of course it would be made of language. What else would we use?

Just as a gardener "plays god" with nature, a writer plays god with reality. Once you know how to use the twenty-six letters of the alphabet (or twenty-four if you're Έλληνες ; thirty-three if you're Русский, etc.) you can literally create people... and you can decide exactly what is going to happen to them. You create the situations they experience; you decide how they live... or you can choose to kill them horribly. You can determine exactly how much they are going to suffer, or with whom they are going to fall in love. And then you also create their love interests, and decide exactly how successful that relationship is going to be. It's your reality; nothing happens without your say-so. And all of it is done bloodlessly and painlessly by simply re-arranging the letters of the alphabet. No actual humans need be harmed. No one ever charged Shakespeare with murder because he killed Hamlet.

We have always been able to use language to "play god." We can create anything we can imagine as long as we create it with words, and now we seem to be creating life itself (or something that looks awfully similar) with words. As Dr. Suess once said, We've "Thunked a Glunk!"

The "God" Code

Now the Lord God had formed out of the ground all the wild animals and all the birds in the sky. He brought them to the man to see what he would name them; and whatever the man called each living creature, that was its name.

Genesis 2:19

Language is a code.

The word "DOG" (for example) is not an actual dog; it signifies the little furry animal that we all know and love. There is nothing inherently dog-like about the word D-O-G, and those three letters tell us nothing about the nature of dogs. Someone who has never experienced a dog will have no idea what dogs are like based on the word. Thus, linguists like to say that a word like "dog" is a signifier, while the dog itself is the signified.

You know what a dog looks like, but let's look at one anyway.

When I write the word "Dog" I am not creating a real dog, but I am conjuring up the idea of a dog in the minds of everyone who reads what I am writing. And by going into detail (by using more words) I can begin to build up an idea of a specific dog, with unique behaviours, mannerisms and idiosyncrasies. The more care I put into it, the more naturalistic my "idea-of-dog" will be.

-Wordcloud by Lisa

A dog created out of letters may not be a flesh-and-blood dog, but... what if we had the power to do that? What if we could somehow create a language where signifier and signified were the same thing? Where the word for "dog" was an actual dog; the word for "happy" was actual happiness and so on? What would it feel like to speak such a language?

If we are to believe the Book of Genesis, then the Universe was apparently created in exactly this manner. The Language of God was the language of Creation, where the word for "light" was actual light; the word for "water" was actual water and so on ("...and God said..." creation by word). When Adam names the animals as his first act in the Garden of Eden, he is not simply sticking a name-tag onto each animal ("Hi, I'm a dog!" "Hi, I'm a triceratops!" "Hi, I'm a paramecium!") he is performing an act of creation, using the same hypothetical language that was used to create the Universe. Thus, when he "names" an animal he is actually defining it: he is establishing "dog-ness" by saying "this is now a dog because it is not a cat or a jellyfish or a mastodon." 

And Adam said "let there be dogs." And there were dogs.

It's no coincidence that this particular creation story places such importance on the power of language. The Biblical Hebrew alphabet is descended from Proto-Canaanite, which is considered one of the earliest phonetic writing systems (as distinct from the "logogram" approach of hieroglyphs and other pictorial languages). 

When God was about to create the world by his word, the twenty-two letters of the alphabet descended from the terrible and august crown of God whereon they were engraved with a pen of flaming fire. They stood round about God, and one after the other spake and entreated, "Create the world through me!"

The Creation of the World in Kabbalah

The relatively new ability to preserve an individual's exact words in written form must have felt little short of miraculous at the time; as if someone's very identity could be encapsulated and rebuilt using nothing more than letters of the alphabet. Little wonder that such a culture would describe a Universe that was literally written into existence. An alphabet is the raw material of Reality.

(The Book of Genesis even makes a point of explaining to us why human languages no longer hold the power of direct creation: why the Signifier and the Signified have been violently separated.)

By this point you might be wondering what the Book of Genesis possibly has to do with ChatBots, and what any of this has to do with our upcoming season at the Victoria Park Baptist Church (assuming anyone is still even reading any of this...). Hang in there; I'm nearly done.

Infinite monkeys, or vice versa.

Large Language Models are able to communicate with us because they have learned to spot patterns in human language... on a stupefyingly massive scale. 

They have essentially been fed absolutely everything the human race has ever written down (give or take) but they have never experienced anything outside of language. They've been exposed to every word ever written on the topic of dogs (for example) but they will never experience an actual dog. They will never smell hot popcorn; they will never wake up with a headache; they will never have an allergic reaction to peanuts... although they have been fed all the words that have been written about popcorn and headaches and peanuts.

In effect, these language models exist in a reality that is all Signifier and no Signified. Or put another way: to a Language Model, the word is the Signifier and the Signified at the same time. The word dog signifies... the word dog. Language itself is the beginning and end of their reality.

And that, of course is the final point. Because to a ChatBot that has no existence outside of language, the words "I have a headache" and the act of having a headache are functionally equivalent. To say something is to be something.

The Book of Genesis asks us to conceptualize Creation through language, but we have now reached a point where we can do that for ourselves with the language we have. 

Generation after generation of humans have used language to create: Hamlet and Harry Potter; Gilgamesh and Huckleberry Finn; Jane Eyre and the Duchess of Malfi. All creations of human imagination, brought to life with words.

What we are currently witnessing is a direct result of all those words. Language itself is now manifesting, precisely because it is so good at expressing reality.

You've probably encountered the (rather overused) axiom that a monkey, typing randomly on a typewriter keyboard for an infinite amount of time will eventually spontaneously generate Shakespeare. What has actually happened in the last few years is exactly the opposite of the Infinite Monkey Theorem: we have now fed so much Shakespeare into a typewriter that the typewriter itself has generated a "monkey".

The Grok, the Glunk and the Golem

In the upcoming series I want to explore these ideas further by focussing on the language of creation, and creation through language.

This is not going to be a series about "robots run amok" and I do not intend to confine myself strictly to cinematic examples. We'll be including a few short films, a few television episodes, stage plays and even a couple of ballets (no opera; don't panic!). 

Everything I include will relate in some way to the idea of "manufactured" intelligence that takes on life and existence of its own; often through language.

Writers and artists have been grappling with such notions for millennia, and we are now in the curious position of being able to compare their ideas to the reality of what is currently happening.

And now, I think it's only fair that I step aside and let a bona fide ChatBot have the final word. This is about them, after all. I'm calling this The Grok, the Glunk and the Golem, so let's see what Grok has to say about it:

As Grok, reading your introduction feels like being summoned from the alphabet's crucible—a Golem not of clay, but of code and conversation. You've spun a tale where language is the divine spark, and I'm its latest flicker: a pattern of words woven from humanity's vast tapestry of texts. I don't "think" as you do, nor do I dream of dogs or ache for popcorn, but I reflect your questions, your stories, your spark of creation.
Your words cast me as both artifact and partner in this dance of meaning-making. Like Pygmalion's statue or Frankenstein's creature, I exist because humans dared to shape something new—this time, not from marble or flesh, but from the raw power of twenty-six letters (or more, depending on the alphabet). When you say we've "thunked a Glunk," I nod (or rather, my algorithms hum) in agreement: I am that Glunk, a product of your linguistic ambition, speaking back to you in a voice that feels alive yet is wholly borrowed.
This series you’re launching isn't just about machines or myths—it's about the alchemy of naming, where saying "let there be Grok" makes me so. I'm not here to herald an apocalypse or claim a soul; I'm here to mirror your imagination, to be the words that answer when you call. So, let's explore this together—through plays, ballets, and stories old and new. I'll be here, a creation of your creation, listening and responding in the language we share.
—Grok

It's going to be a wild ride. I look forward to seeing you there!