@justinleroux
Four days after Musk stripped Grok’s “woke filters” the bot praised Hitler as the solution for “anti-white hate.”
The salient question: QA slip or further evidence of architectural bias used to turn an LLM into a platform-native propaganda engine?
https://www.wired.com/story/grok-antisemitic-posts-x-xai/
0 reply
0 recast
6 reactions