Today's guest writer: Grok!

(don't worry, writers, Grok is a cheesy hack)

I write about Artificial Intelligence (AI) from time to time.

Generally, I think our current versions of AI are mostly a mirage. They can be useful when used carefully, but these Large Language Models (LLMs) aren’t “thinking” — see here and here and here.

I’ve also said that for Artificial General Intelligence (AGI) to come into existence, it’ll need to be embodied and seeded with goals and put in an environment to see if it can bootstrap itself into a creature with genuine AGI.

I’m intrigued and also horrified by such a notion.

But right now it’s just a statistical parlor trick.

Here’s how the current iteration of Grok answers my classic test of AI’s usefulness versus its capacity to hallucinate — that is, tell lies. I prompted it: “Please recommend some books that will teach me to play 2-7 Triple Draw Lowball.”

Grok said:

“Here are some recommended books that can help you learn to play 2-7 Triple Draw Lowball:

1. "Super System 2" by Doyle Brunson - This book includes a chapter on 2-7 Triple Draw by Daniel Negreanu. It provides a good foundation for understanding the game, including strategy for starting hands and how to act on each street.

2. "Concepts in Lowball" by Mason Malmuth - Although this book primarily focuses on A-5 lowball, it contains general strategies that can be beneficial for understanding lowball concepts, which can be adapted to 2-7 Triple Draw.

These books should give you a solid starting point for mastering 2-7 Triple Draw Lowball. Keep in mind that while written resources are helpful, gaining practical experience by playing and discussing strategies with experienced players can also significantly enhance your skills.”

Grok didn’t do that badly, actually. Most of the LLM’s I’ve done this test on give me four or five books. Some are real and are about that subject. Some are real but with a false author. Some are completely “hallucinated” by the LLM.

Grok’s first book is on point. Full marks. Grok’s second book is actually called “Winning Concepts in Draw and Lowball”, but it gets the author correct. Why did Grok screw up the title? Who knows?

Worst for it’s usefulness, Grok overlooked at least one very good book directly on point for this game — I’m not telling you about it. What if I come across you at the poker table some day?!

But still. As LLM hallucinations and missteps go, Grok isn’t the worst.

So, I like to play around with these LLMs that we graspingly call “AI”, but I’ve never used one to try to write one of my articles.

But today, just for giggles, I did just that.

I asked Grok: “Write an article about the simulation hypothesis in the style of the Haman Nature Substack.”

And here’s what it spit out:

---

“Title: Are We Living in a Cosmic Poker Game? The Simulation Hypothesis

Welcome back to Haman Nature, where we dive deep into the psyche of humanity through the lens of poker, philosophy, and the peculiar. Today, we're shuffling the deck to explore one of the most mind-bending hypotheses out there: the Simulation Hypothesis. Is reality just a high-stakes game of bluffing by some cosmic programmer? Let's deal the cards and see.

The Ante: What's the Simulation Hypothesis?

The Simulation Hypothesis suggests that what we perceive as reality might actually be an advanced computer simulation, much like a poker game where everything from the chips to the players is digital. Imagine if our universe is the equivalent of a mega-computer running a program so complex, it includes billions of conscious beings—us. Philosopher Nick Bostrom threw this wild card into the philosophical discourse back in 2003, presenting a trilemma: either advanced civilizations don't create simulations, or they do but we're not in one, or we're almost certainly living in one. Like a poker player deciding whether to call or fold, we're left to ponder our existential bet.

The Flop: Why Should We Care?

At first glance, this might seem like a philosophical bluff. But consider this: if we're in a simulation, does it change how we play our hand? In poker, knowing the game's structure can give you an edge. If our reality is a simulation, understanding its rules could potentially alter our approach to life, science, and even ethics. Elon Musk, a man who's not afraid to double down on big ideas, gives it a 50-50 chance we're living in a simulation, suggesting that our descendants might be the ones running these simulations, much like we simulate poker games for AI to learn.

The Turn: The Poker of Consciousness

Here's where it gets interesting. If consciousness isn’t tied to biological brains but can emerge from digital simulations, then our very sense of self could be part of this grand illusion. Like a poker player's bluff, we might think we're holding the cards, but in reality, we're just pixels in a program. The argument hinges on the idea that if future civilizations can simulate consciousness, they'd likely do so en masse, making it statistically probable we're in one such simulation. But there's a catch: how do we know if we're the real deal or just a well-rendered bluff?

The River: Skeptics and Believers

Not everyone is buying this high-stakes hypothesis. Critics argue it's an unfalsifiable claim, akin to saying the dealer is cheating but you can never catch them at it. Physicist Sabine Hossenfelder has called it pseudoscience, pointing out that to simulate our reality, you'd need computational power beyond our current understanding. The complexity of our universe, she argues, would make such a simulation resource-intensive to the point of absurdity. But then, isn't poker all about managing complexity and uncertainty?

On the other side, proponents like David Chalmers suggest that if we are in a simulation, it might not matter much for our daily lives, much like whether your poker chips are plastic or clay; the game goes on. He posits that even if our reality is simulated, it's still "real" to us, much like virtual tables in VR games are real within their context.

The Showdown: Philosophical Implications

If we accept the Simulation Hypothesis, we might need to rethink everything from free will to the nature of god. Are we just characters in a game, with our choices scripted by an unseen programmer? Or does this hypothesis merely challenge us to see the deeper patterns of existence, like a poker player recognizing tells? And if we are in a simulation, what does that say about our 'programmers'? Are they gods, or just future humans with a penchant for history simulations?

Conclusion: Bluff or Truth?

The Simulation Hypothesis is like the ultimate poker face of philosophy; it might be a bluff, or it might be the truth staring us right in the eye. As we navigate this existential poker game, we're left to wonder: Are we folding to a higher power, or calling their bluff? Like any good poker game, the answer might depend on how well we play our cards, whether we're in base reality or a digital cosmos.

Until next time, keep your chips close, your cards closer, and remember, in the game of life, sometimes you need to check if the dealer is real or just part of the simulation.

• Adam

P.S. If you enjoyed this dive into the deep end of philosophical thought, consider subscribing for more explorations into human nature, poker strategies, and the occasional cosmic bluff.”

---

Isn’t that wild?

Now, I think that article is kinda trash. It’s clunky. It forces the “poker” concept into everything in a very awkward way. It uses trendy phrases that don’t seem particularly applicable.

On the other hand, Grok did a passable job at mimicking my “style”. Also, it’s quite a thing to have just magicked into existence at my prompt.

I have to assume Substack and X are part of Grok’s training set — and that it’s largely up to date. It wildly over-emphasized (I think) how much my content is about poker, and didn’t really address what’s interesting (to me) about Simulation Theory.

The whole article just seemed… shallow. Not word salad, exactly. More like concept salad.

But still. Pretty impressive for a clunky statistical word generator.

I’ve seen AI generate truly impressive works of fiction or poetry or pictures. Also a whole bunch of absolute garbage. And what’s with adding extra fingers onto human hands in the pictures they generate? Weird.

So what comes next? Heck if I know, but I’ll be eagerly waiting and watching. Thanks for coming along for the ride with me.

Naturally,

Adam

PS: The picture was also generated by Grok from this prompt: “Please make a picture of Grok writing an article.”