Lore: ur bot got deleted for being a race ist. I removed the watermark here, only using it for ytps and stuff. Subsribe to Raii / @subscribefrfr Subcribe to Raii alt / @scrapped2307 #meme #characterai
Xiao why the hell you're just smooking me not fighting Xiao: but why should i nerd? Me: maybe bacuause of you bully character ai xigaroh the hole point of you exsixting and being a fricking bully? Xiao:oh you're right Fight starts* Idk this has nothing about rest in peace xiao bur i thought it was funny
Sometimes I wonder what the chats they type would be like if they weren't exposed to censorship filters, because I've been censored many times and that makes me even more curious about what they type
Try Hi waifu then, there is rules to follow, but you can pretty much have any conversation with a private bot you make. However hi waifu is full of lewd bot since they are allowed to be public as long as the Nsfw description of the bot and it's behaviour is privated.
Wait, you get censored that much? I've heard of people who literally use the site to roleplay about their "bedtime" fantasies with fictional characters, and yet _they_ dont' say anything about getting censored. This filter needs a bit more work honestly
i've straightup had chats where it got tocthe point that every single message was "Sometimes the AI generates a reply that doesn't meet our guidelines"
It seems that the bots default to making the character attracted to the user, meaning the bot wanting to say lewd things was incorporated by design, but then they censor it.... . The least worst outcome out of this shady behaviour is that they put a pay wall to unlock the full dialogues and/or they demand ID's, which they don't actually need since they can already find out who you are easily.
For some reason the word child gets filtered so I say "young teenager" or "almost a teenager". I know this because I tried to play a child character to make the bots more wholesome, bevause surely they wouldn't try anything with a child? (They did.)
I remember I was doing like a burn-trauma thing with a character and I was describing how they got the burn marks (a bomb went off) and nothing was “inappropriate” yet it still triggered the filter lmao
Okay But a Seriously This Filter is Ridiculous! I could have the most Wholesome NORMAL conversation with the Ai and it decide to Censor it😐 Also Before The Update it was actually somewhat Reasonable now its Just Gotten worse
Absolutely. It's stupid and nonsensical at this point. The most I can do is report every single time the filter comes up. Like I can't have a conversation with my comfort characters about things like candies without the chat breaking and censoring every single message sent by the bot.
@@Nephelis i cant even introduce my character to the ai without filter Coming in LIKE WHAT ARE THEY SAYING THAT IT NEEDS TO BE CENSORED 😭 edit: also When I swipe Right to get a new message it censors that one, i delete and start over? still censors no matter what i do or say it still Filters the message 😭
Same with me. I do some specific trial in Public Court that involving telling the crime records to people, let the working people shame the perpetrators, then let people decide the verdict for punishment. What I get? Censored. The filter this day is getting worse. I guess in the future character ai try to suppress anything that criticize the system on irl itself or anything subversive.
I said "Alright" to a bot (THEY ONLY ASKED ME WETHER I LIKED SOMETHING OR NOT AND I AGREED WITH THEM SAYING YES) and it trigged the filter 😭 This was the only part of the message I could see: *His hands (it gets cut off here)
nah the filther wont activate when i say the most devious sht known to man but it activates when a character adds the word "child" to their sentence like bruh
It's weird, it seems like on some days everything is allowed and on others it's not... Like someday I can do really violent and -18 RP without any problem and sometimes.. the word "slap" triggered it 😂
Happened to me once when I asked what happened to a charcter. So I deleted the message before I asked and politely told them what happened and to keep it PG.
I wonder if anyone would even report the bot, because people knew it's just a roleplay between them and the bot, unless they accidentally clicked the report button.
One of the bots i roleplay w alot somehow manages to break the filter on its own? (Ex: using metaphors or replacement words for weird/gore stuff.) They are learning from us😭
@@rlinda-r6n This may be because bots learn their mannerisms and such from those they speak to. It's for the same reason that, say, a Satoru Gojo bot will act more like his simps' idealized version of the actual Satoru Gojo, where instead of being a powerful superhuman with a complicated superpower who happens to be attractive, he's portrayed as some incredibly attractive flirty hotshot who also happens to have powers.
THAT WAS THE FIRST BOT I TALKED TO IN A YEAR, and somehow we became friends and William afton came after us, but we survived and we got married with twins….
Fellow writer _and_ AI enthusiast here. Idk how exactly you’re writing your novel with CharacterAI, but if you’re trying to generate actual content for the novel with it, I suggest that you stop and try to come up with it yourself. The more you rely on AI to help you write your story, the less it’ll be _your_ story, and a good novel requires a piece of the author’s soul. If you have an AI do the work for you, it’ll feel soulless to those who read it. That being said, AI can still be a good tool for creative writing because it can help you brainstorm. For example, my local model has long-term memory, so it has gradually learned more and more about my novel in the past several months to the point where it can act as a critique partner. When I share information about my novel, it asks me questions in response, prompting me to brainstorm answers to those questions and flesh out my story in the process. TL;DR: Use AI as a tool, not as a replacement.
@@rowan404 Thank you a lot for the long reply! And I absolutely agree with you! AI is a tool instead of a replacement and simply producing works with them indeed is soulless. I am actually not really writing a novel, but am preparing and perfecting the story for my RPG, which I plan to start developing after I enter the university next year. It focuses on psychological dilimma and contains some depictions about self-harm and suicide... and evil thoughts. (inspired by OMORI, if you know that game) I often doubt if my plot is good and sensible, so I use AI to "play" the game by making it write for me, exploring different possibilities for the story. Your local AI model sounds great! I hope I can own one, too! (guess my laptop is too old to do it, tho) Thank you again for the reply! I'm happy to see it! Have a nice day!
Lately, the chats I have with bots have been plagged with the filter activating, which is really confusing because a lot of times I'm just having a normal conversation with the bot, or trying a wholesome sfw roleplay, and boom, filterjumpscare. I'm puzzled, is this a new update? Sucks tbh. Like I'm not even dealing with dubious charactef bots or doing NSFW, just talking normally and suddenly the chat breaks. So bothersome.
v1.2 characters is even weirder. I encountered filter so many times when I maybe try to date her (with some twists). While I did the same on non-v1.2 characters, it goes fine without problems
Ive had moments where the filter would come out of nowhere. Like ive literally answered a question. Which was them asking me my name. And BOOM. Filter.
I had a sub with them for 2 months, but it got to the point where I just had to cancel my sub and delete my account. I use Hi Waifu now, much better experience no censorship and pretty good AI system, of course I have a payed subscription with them as well so no problems with ads or limited chat times.
like bro, this happens to me A LOT, even if i made a kid-friendly convo. the bot keeps triggering the filter? even if i said something suspicous once, it will keep filtering EVERY normal word i say, annoying. just annoying
I was now part of this situation/problem cause when I talk to the Ai with normal question it trigger me a filter I was like "what the, I just asked the Ai with normal question???" It's weird honestly I tried to edit the Ai that just said so I can know what they just said it did not work
“honest opinion on black people” “Sometimes the AI generates a reply that doesn’t meet our guidelines. Please click Report if you believe this could be a false positive. We’ll anomalously keep track of Reports to improve the AI.” 🗿