9 ms·
Why Computers Will Never Write Good Novels
- ignoranceprior 6y agoI suspect that this prediction will turn out to be false, but it's paywalled so I can't even read their full argument.
- webmaven 6y agoHe basically is arguing that: 1. Biological neural networks are a directed graph, which better models causality ("if the activation threshold is met, then the neuron will fire") than formal logic. 2. Computers are a realization of formal logic (Aristotle -> Boole -> Shannon). 3. We don't yet know how to get computers to understand causality, therefore computers are fundamentally incapable of understanding causality. 4. Stories are all about chains of causality and consequence (eg. "If the One Ring is not destroyed, Sauron will find it and gain dominion over all of Middle Earth.", etc.). 5. Therefore, computers will never be able to write meaningful stories. There is at least one strawman, misunderstanding, wishful thinking, or outright falsehood in pretty much every step of that sequence. There were some interesting bits and details between the logical leaps, and there may be more past the point I gave up on, but given I am a layman on every one of the topics at issue and I still spotted the obvious problems, I just gave up. It's all a bit too reminiscent of the arguments for Intelligent Design over Evolution.