@Eli_Jah_Watt a sure fire way to break the filter is just to have the bot write all of it. When you get a message you like, don’t reply—just press the send button for the bot to generate the next response. You’ll get a LOT farther before it registers the filter.
It’s hilarious seeing WWII RP bots have the warning “this is not a licensed professional” meant for therapist bots. There’s also one on a bot for the Purge.
And the bot with the boy who suicided didn’t understand the context that he was about to end his life. The bots are smart, but they don’t have eyes into the real world. It can only work with what it’s given. And it was only given a really offhand comment “I’ll be with you soon” or smth like that
Not the bot’s fault, not the kid’s fault, more-so the mom’s fault for not regulating how much time he was spending holed up in his room. He more than likely got a lack of sleep, which exacerbated the depression. And now the mom, who is at fault for not being a proper parent to the kid, is suing c.ai over the AI doing its normal function
Sorry that was a rant-
