Skip to content

Comment on Anthropic moves to close loopholes that allow Chinese access to Claudeparent

Comments

In this logic it seems reasonable to allow some Chinese version of Claude to talk about self-harm topic without censourship. Let those not friendly counterparties to experience AI not just without some proper functions but also loaded with improper ones. Exactly as stratagem 25 tells "Replace strong piles with rotten ones".

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.