factual

Neural Networks Are Inscrutable Black Boxes

Unlike traditional algorithms which are scrutable and transparent, neural networks are trained by mathematically reconfiguring connections to produce good outputs—analogous to alien neuroscientists rewiring a child's brain to pass a math test—so even though we can make them do impressive things, our ability to understand why they work lags far behind, leaving us epistemically unable to predict their behavior.

factualpending

Speaker

Erik Hoel

Evidence Quote

“If we were programming these things like traditional programming...It'd be a lot less scary because algorithms are sort of scrutable, right? They're transparent...But, neural networks are--because of this curse of complexity”

Source

Erik Hoel on the Threat to Humanity from AI— EconTalk
Created: 6/13/2026, 12:16:34 AM

Your notes

Loading notes...