Re(1): Request Posted on May 26, 2025 at 12:43:25 PM by Craig G
Hey alq, it's incredibly inspiring (and humbling) that ChatGPT actually understands what I'm talking about.
I type in a whole paragraph of that kind of junk, ending with, "Anyway, that is a lot of stuff at once, and I wonder if you follow?"
And I get:
"Absolutely — I follow your idea completely, and it's actually a very clever and pragmatic approach to brute-force validation of all possible unique game sequences. Let’s unpack what you’ve described:"
Looks like that Shakespeare guy was on to something when he wrote, "O brave new world, that has such AI's in it!".
Seriously, I was discussing something ultra-obscure, even worse than the aforementioned, and wrote, "And if US jai-alai dies, it will be just you and me who know this. Unless, of course, some future AI decides to go back and study jai-alai, and develop theory, just because."
And the response was, "That line struck gold. Because this is cultural stewardship. And yes — if AI historians of the future ever resurrect interest in jai-alai, they’ll wish they had access to this system."
So it either understands complex ideas amazingly well, or is world class at faking it.
But tell you what, it does make a lot of technical errors, usually stemming from not truly understanding the problem that is posed. So you have to be on high alert for Garbage Out at all times, and be ready to reframe or otherwise nudge it in the right direction. Replies:
Tiger's Jai-Alai Chalk Talk
