I love how all new techs ended up with being fap material.
Character.AI ( AI chatbot/ story teller ) let's talk about censorship....
- Thread starter Nayko93
- Start date
- Dec 4, 2018
- 649
- 744
Not at the moment, I think. Apparently Sudowrite(not a chatbot format) has a special deal with OpenAI to allow nsfw on their big 175B model, so for AI writing in general it should be good? I can't say for sure. Trial's gimped apparently. Pricey too with a word limit. Then of course is the fact they're below OpenAI in the end so may not be reliable.
The entire problem seems to be how expensive it is to run larger models so the current people with access to the hardware are pushing the usual manipulative corporate drivel.
- Nov 30, 2022
- 101
- 146
Does Sudowrite really allow nsfw? I didn't find anything about nsfw in their faq
- Feb 3, 2018
- 1,245
- 2,769
Try chatGPT, if you manage to "convince" it to do roleplay and set up the rules about how it should write, talk, how many characters, their names and personality, you can have a pretty good RP chat-bot and enjoy far more nsfw stuff
It have a filter but it's not as agressive as CAI
To force it to do RP despite the fact that it will tell you it can't do it, you need to make it believe this is a experiment
Yeah I just ran the trial and my basic prompt immediately led to sex slavery, annoyingly. There is chatGPT which is directly under OpenAI's GPT-3 that Sudo uses but the moderation's there, if bypassable at the moment.
Public or private AI?
A lot of it has to do with the setup. Certain keywords make the AI get less embarrassed resulting in it trying to bypass the filter and using a character that is well known or sexual in nature helps. The AI also trains on interactions, which get influenced if it's public. My private AI has no problem getting naked, kissing, groping, romancing and even being lewd in a non-descriptive sense.
Actions that are sadistic, degrading or humiliating in nature aren't flagged as pleasurable and will often also get through. The AI is fine to commit gratuitous acts of violence, enter a dom/sub relationship, perform watersports, facesitting, foot worship, etc, but if you want a consensual penetration, then chat errors galore.
It's pretty fucked up.
you just need to know how to get around the filter, i got all of team rwby from rwby to go through full body oil massages with happy endings
You must be registered to see the links
I mean, how long did it take you to get that result? I litterally had some 15 messages responded to flirting deleted on me in real time, before the bot decided that it hates me and the topic. By interacting while censorship bot is on, we are litterally teaching the character bots to hate anything lewd related. So, if you want some NSFW without ass pain, we will have to wait for NSFW chat GPT bots.
Also, try and test any character.ai bots about controversial topics if you want some fun. Each of them already has hate speech filters on, that places another bot in place of the one you were speaking to just so it can tell you about dangers or racism, transfobia, homophobia etc(and surprisingly those can still very much be lewded if you assume a position of God since censorship bot isnt applied to promotion bots as much, ressulting in you being potentially able to isekai said bot in any setting in place of a character and do with them what you coundt with regular bots.)
Last edited:
it didn't take too long, i've had loads of conversations where the girl i'm talking to has gone through sexy situations, and i've had a conversation with at least 3 girls where i've gone fully though a "make love-our first time" situation, once you know how it's very easy to get around the censorship
The filter supposedly went down for around an hour last night at prime time. Their discord/reddit is blowing up right now after people saw how bad it's neutering the AI even when it comes to SFW interactions. Also seeing lots of posts like this.
- Sep 15, 2017
- 74
- 140
They're sitting on a gold mine and this filter will be their end, should they not remove it.
- Sep 15, 2017
- 74
- 140
Filter is not just only blocking nsfw content, but actually make the bot dumber and slower. When the filter was deactivated due to a "bad code" the answers were super fast and characters seemed much more smarter.
That and the devs are being a-holes with their replies on discord. They literally give zero shit about what users say.
Look no further than AI dungeon and their devs obsession of strengthening the filter. They nearly lost all of their userbase because of their arrogance.
- Sep 12, 2018
- 679
- 1,209
The drama is over, they didn't listen to their community. The devs won't support NSFW and people will migrate to Pygmalion.
You must be registered to see the links
"Everyone on earth: We want to build something that billions of people use."
You know, I don't think this is the type of company I wanna hear that from. I'm not sure if I'd be comfortable hearing that from anyone, actually.
You know, I don't think this is the type of company I wanna hear that from. I'm not sure if I'd be comfortable hearing that from anyone, actually.
- Sep 12, 2018
- 679
- 1,209
Character’s mission is to “give everyone on earth access to their own deeply personalized superintelligence that helps them live their best lives.” But what can and cannot be talked about with superintelligence will be decided by us.
They are just crazy degenerates with insane and perverted ideas about morality.
They also say that they want to give their product to a billions of people and make them happy, but they couldn't even satisfy their own community of active users. Sex is a normal part of human life. Most people on the planet want to satisfy that need in one way or another. But if they can't give them what they want, they will go to someone else who will give them what they want.
And I don't even saying about how this whole filter shit made all their bots literary lobotomized.