Communication Skills are critical with AI use

[AION]

 I noticed that most people don't use AI the right way. Even people with very high credentials and a high educational level, they seem not to know how to communicate with AI and how to use AI for you. They try to be the oracle while AI is the oracle. Just like the Greek mythology, you only get an answer to the question you ask. So if you ask stupid questions, you get stupid answers. And if you don't ask questions, you don't even get answers. And the beauty is with the mythological oracle you could only ask questions and get an answer, but with AI, you can ask it to ask you intelligent questions about your problem/opportunity and totally flipping the script. Having the right paradigm with AI use is critical and I see nobody talk about it.  

Edited June 5 by AION

[LordFall]

You're 100% correct here. People complain that AI ruins critical thinking and just gives you the answer. Why not just AI to challenge your thinking?

I just had a heated argument with someone on Reddit that accused me of projecting and I thought they were saying illogical stuff that didn't make sense and were themselves projecting. I put the entire convo in ChatGPT which said that I was indeed projecting and invalidating the person's point. Very interesting new technology that opens up new ways self-develop like never before.

I'm quite good at rationalization so it's hard for me to see when I'm wrong but I'm open to feedback so this is super exciting for me. 

[Carl-Richard]

I'll try to get ChatGPT to ask me some questions at the end of every answer it gives me 👍

Edited June 6 by Carl-Richard

[AION]

Flipping the script is key in paradigm shifts. 

[Basman]

I found AI to be helpful with expanding my thinking and generating ideas. I found that it can soften your view on specific topics and is particularly useful for contemplating very niche topics that people aren't talking about, for example Danish politics. I don't use AI all that often however because I feel like it doesn't provide me all that much value ultimately though that could just be a skill issue.

You can never trust it to tell the truth though. It will always just be an "idea". It would be a mistake to ever give it epistemic control. That is the key distinction of responsible AI use in my opinion. I never go in looking for an answer from it. 

[AION]

One can even deconstruct AI to such an extent that it won’t like it. This only shows how much power the owners of AI will have. Owners decide the rules of AI. They will be the new oligarchs and they will have massive amounts of power. In my opinion the only branch worth studying at this point is AI because all else will be taken over. Education will also face massive paradigm changes because even today education sucks. And even great minds like Leo Gura teach with an outdated paradigm of teaching. 

Edited June 7 by AION

[Nilsi]

Everyone talks about how to use AI. They repeat the same line: it’s a powerful technology, and you just need to ask better questions. But that’s not interesting. What matters isn’t how to use it, but what it reveals.

Because once you understand how these large language models work - what they actually do - you realize something much deeper. You realize it’s not a commentary on intelligence or technology at all. It’s a commentary on language.

Here’s what I mean.

LLMs don’t think. They don’t reason. They don’t “know” anything. What they do is predict. They predict the next most likely word, one token at a time, based on a vast statistical model of human language. That’s all. That’s it. The most probable next word. And then the one after that. And then the one after that.

Now think about what that implies.

Let’s say - just for the sake of argument - that 99% of all human communication was chaotic, entropic, incoherent. Total noise. And only 1% was stable, rational, structured, “meaningful.” Even in that scenario, a model trained to predict the most likely next word would still gravitate toward that 1%. Why? Because coherence is narrow. There are only so many ways to be coherent. Only so many words that can come after a given sentence that preserve flow, syntax, stability. But there are infinite ways to be incoherent. Infinite ways to rupture, interrupt, fragment.

And this is the point. This is why LLMs seem so agreeable, so reasonable, so syntactically smooth. Not because they’re designed to be polite. Not because they’re trained to follow instructions. But because that’s how language itself works. The coherence isn’t artificial - it’s structural.

And this isn’t just about AI anymore. It’s about you.

Because if you’ve ever been in a high-stakes situation - say, on a sales call, heart pounding, mind scattered, adrenaline surging - you’ll know the feeling. You don’t know what to say. You start the sentence blindly. You just open your mouth. And then, somehow, the rest of the words come. They line up. They follow. You improvise. But really, you don’t. They arrive like dominos falling. One word makes the next inevitable. You’re not thinking. You’re predicting. You’re doing exactly what the language model does. You’re surfing probability.

And this is what Lacan meant when he said the unconscious is structured like a language.

It means that what we call “reality” - the everyday, stable flow of sense - is mediated through a statistical process. Language stabilizes experience by smoothing it, by filtering it through chains of signifiers that make sense after the fact. It feels real, but it’s retroactive. It’s a simulation. And what you take to be spontaneous thought might just be a probabilistic reflex.

AI shows us this. Not because it simulates intelligence well - but because it simulates language perfectly. And when it does that, it reveals something terrifying: that language doesn’t describe reality - it hides it.

Because the Real - the Lacanian Real - isn’t what language expresses. It’s what it misses. It’s rupture. It’s the hot girl walking by when you’re trying to close a million-euro deal. It’s the stutter, the silence, the loss of words. It’s what doesn’t fit.

Which brings us to the Singularity.

People say AI will wake up. It will become conscious. It will break through, self-improve, transcend, whatever. But that’s a fantasy. It’s the Imaginary at full blast. They picture it like heaven. Upload your mind. Merge with the machine. Escape finitude. Escape death.

But nothing will happen. Not to you.

Because even if something “emerges,” even if the machine evolves - your experience of the world won’t rupture. It will stabilize. It will cohere. It will simulate. You’ll still be inside language. Inside a feedback loop of symbols, patterns, predictions. You’ll still be speaking in sentences. You’ll still be bound by signifiers.

The Singularity won’t save you. It won’t break anything. It will only deepen the trap.

So no, this isn’t a critique of AI. It’s a critique of you. Of language. Of the Symbolic. Of the false coherence that structures your reality. Because if LLMs teach us anything, it’s this:

You’re not speaking. You’re being spoken.

And the Real? The Real is elsewhere.