Neural nets are like chess: we grasp the rules but can't fully predict the game
JFPuget · x · 2026-09-24
Chomba Bupe argues that "understanding" is a top-level concept: we understand the linear algebra and statistics governing neural nets the same way we understand chess rules — well enough to know they can't spontaneously wake up and kill everyone, even if we can't fully predict every game. A reply counters that engineers fully understand microprocessors, which is a crucial difference.
More from AGI Musings
- Abe Davis warns AI decouples knowledge from understanding—and settling for knowledge is dangerous — weichiuma · 2026-09-24
- Safety researcher to US AI chief: companies themselves warn unregulated AI races create unsafe practices — davidmanheim · 2026-09-24
- Google lacks the product DNA for a personal agent while Meta readies Muse, argues signulll — signulll · 2026-09-24
- EA blasted as 'unhinged': the philosopher who wants to save insects by exterminating them — anselm · 2026-09-24
- My fiance believes AI extinction is a decade away — how justified is that fear? — sailingintothedark · 2026-09-24
- UCSD design professor: AI makes us create more but think less, and worse — HaijunXia · 2026-09-24