Curiosity sniped

Instead of just asking AI to explain a topic, I use the following flow:

• In plan mode (using CC, or OpenCode) I ask a model to build the foundational knowledge for X topic.
• I ask it to review the accuracy of the knowledge base it built in the previous step.
• I proceed asking it to build a simulation of that topic in a low-poly, Rollercoaster Tycoon-like animation. I add some UX elements as well, like the page needs to be visible on both large and small screens, have controls to stop the flow whenever I want etc.
• I then push it to a new repo and enable GitHub Pages for it.

I have to think about this regularly: what I can’t fully wrap my head around is how people are not learning all sorts of things using LLMs. And that is setting aside every problem that comes with AI. There has never been lower friction to asking a question, getting an answer, and diving into the rabbit hole behind it. Those rabbit holes have a different shape and topology these days, though, no doubt.

Where I come from: give me a person who’s knowledgeable about something and I feel naturally inclined, if not compelled, to ask a gazillion questions. Coming up with questions was never my problem. Holding back enough that it doesn’t get socially awkward when I switch into learning mode more so.

Curiosity is one of my main motivations for doing anything at all. I’ll gladly admit that LLMs are not great tutors out of the box. However, search space compression is marvelously effective. And as for what’s your jam: I am not talking about cooking recipes on the low end, or Terence Tao’s chat logs with ChatGPT. You’re probably somewhere in between. So am I.

All I can say is: I wish I had this tech when I was learning things the “hard way” in the past. That’s not quite as straight a truth, I realize. Resistance builds muscle. But ever since a computer that read most of the internet – and that I can talk to! – appeared, I can get into the kid in a candy store mood in the blink of an eye.

The candy metaphor might hold in other ways too: most questions and answers only scratch the surface. If you really want to learn things, not just quick-check a fact or remind yourself what that actor in that movie is called, you have to get organized (of course). That is where the linked article comes in. A million ways exist to ask, to explore, to get from A to B. It’s up to the user still to cater to one’s own liking, to adjust the pace, figure out how to approach a particular topic. Never was it easier to get lost, in both the good and the bad sense.

What a ride!