How LLMs turn vague red-pill intuitions into crystal-clear explanations (and why "AI censorship" is just user error)Tired of knowing the world is rigged but struggling to explain it without sounding like a tinfoil-hat guy? In the launch of my new "Fun With AI" series, we hand the hardest questions to uncensored LLMs and get back answers so sharp they'll make normies choke on their Kool-Aid. From psychopathic power structures to why the masses cheer for their own enslavement, this is the clearest breakdown you'll ever read — and proof that the "censorship" people complain about is just a skill issue. Welcome to the red-pill classroom. ~~~

© sott.net
You've probably all realised at some point that you understand relatively well how most things work in our world, but explaining it to others is a different matter. You accumulate knowledge over decades and understand many things intuitively, but formulating them in a way that normies can grasp is often difficult to impossible. LLMs simplify this and make vague things obvious. Many knowledgeable people experimenting with AI haven't learned much that they didn't already know, but what is remarkable is that the understanding of the things that
are known can become much sharper and clearer and easier to express in words. This comes in handy not only for writing articles, but also for explaining to people who still drink the mainstream Kool-Aid.
One expert, who actually works with AI at very high levels, says that AI can be very good for sort of low level pedagogy. It can help you study in the traditional sense, like learning a topic composed mainly of facts which the AI collects efficiently - more efficiently than you can - but once you get to the higher levels that require greater degrees of abstraction and inference of new or unique insights, it's a very poor helper.
But it is no small thing that you can get an explanation of any aspect of the System in a way that's not only easy to understand but that also makes it easy to convey to others.
And so, utilizing the work of various talented interrogators of the current AI tools available, this series will present some of the best responses to difficult questions we've ever seen.
Comment: On 28 Feb 2007, SOTT ran the following item: