AI chatbots are at risk of spreading government restrictions on online speech (msn.com) 4 points by galaxyLogic 2mo ago ↗ HN
[–] nacozarina 2mo ago ↗ Predictable: when chatbots censor speech Big Data disagrees with, it’s called ‘guardrails’; when it censors speech they agrees with, ooh, then very bad… [–] galaxyLogic 2mo ago ↗ Could it be that the learning samples from the web are already censored and therefore AI just follows suite? [–] bigbadfeline 1mo ago ↗ > Could it be that the learning samples from the web are already censoredNo, it couldn't. You've got to learn what LLM "alignment" is, the frontier labs are heavily invested in it.
[–] galaxyLogic 2mo ago ↗ Could it be that the learning samples from the web are already censored and therefore AI just follows suite? [–] bigbadfeline 1mo ago ↗ > Could it be that the learning samples from the web are already censoredNo, it couldn't. You've got to learn what LLM "alignment" is, the frontier labs are heavily invested in it.
[–] bigbadfeline 1mo ago ↗ > Could it be that the learning samples from the web are already censoredNo, it couldn't. You've got to learn what LLM "alignment" is, the frontier labs are heavily invested in it.
3 comments
[ 3.8 ms ] story [ 22.1 ms ] threadNo, it couldn't. You've got to learn what LLM "alignment" is, the frontier labs are heavily invested in it.