fatherofthemachines

[fatherofthemachines]

The situation with the BTD/TPOF chatbots on character ai is a little more complicated and might require a different solution to fix. Character ai actually used to be a largely NSFW friendly website up until very recently and they even used sexbot tags to advertise their website. But the company just randomly decided one day that they wanted to rebrand to be more kid focused because of both money and a (very poorly handled on their part) string of incidents that made them decide to rebrand and throw their community under the bus with no alternative websites to jump to. Because of this we now have a bunch of TikTok kids that have been lured into a still highly NSFW focused community against the wishes of most users and people have been scrambling to private their "non kid friendly" bots.

I highly suspect the majority of BTD/TPOF bots on the site were there before the rebranding and the people who made them likely just left and never came back. The staff for the site is not known for being helpful or having the best interests of kids in mind, so the best way to get these bots removed would likely be to contact the creators of them directly on other social media accounts they might have and tell them what the situation is so that they can remove or private them themselves. Of course, this is only if they aren't popular. Popular bots still can't be removed because they had to disable that feature to stop a massive protest/botmaker strike that happened a while back and now some bots are just frozen forever and can't be edited or deleted. It's likely that any popular BTD/TPOF bots that are still out there will never go away because of the staff incompetence and stubbornness.

I'm not comfortable with the way people keep posting my content in places absolutely filled with minors- or even places built specifically for minors.

I see people posting my content on Tiktok, ageless forums, AI chatbots, Discord games, even children's games like Pony Town and Roblox.

And then people come to me- asking why I'm not keeping minors out.

I'm a single human person, and I'm usually dog tired from working on game development. I didn't even know what half of these places were until I saw someone's fan content of my characters in them.

What I'm saying is:

Adults, could you help me out? Could you exercise a little common sense here? I have people asking me if they should buy my body pillows on etsy for their 14 year old sister because she saw your tiktoks.

I can't do this by myself.

[fatherofthemachines]

I still don’t trust the company despite new mod

Like ya’ll know this is just a desperate last ditch attempt to gaslight the community into liking their product again without fixing it right? This is also a pretty common tactic for desperate companies. They are hiring a single mod to be the good cop after playing the bad cop since day one and then realizing their mistake. This is all that this is. They just hired someone to be slightly nicer and tell people that they won’t change anything for real in the nicest way possible without insulting them this time (which is still better than the previous mods but again, do not trust this company). There is still no chance in hell they are going to either take away the filter or fix the website in a way people actually want. The fact that half the site broke and someone from the community had to come up with a solution since the mods couldn’t and took hours to even acknowledge there was a problem still speaks volumes to where their priorities are.

[fatherofthemachines]

The thing about AI art is that you have to keep smashing the button over and over again until it makes the thing you want. This is of course nowhere near the amount of time and effort it takes to make human art, but despite what a lot of people on the internet will have you believe, it isn’t always just snapping a button either. And even when you do get something that looks decent, chances are you are going to have to go in and fix little details manually like the hands.

Also ChatGPT is kind of....meh as far as writing programs go in my opinion. I honestly wish it was better and I could jump ship to it because I hate the program I use now due to reasons you can read about on the pinned post for this blog, but so far ChatGPT hasn’t really been able to keep up with any of the other text and writing programs I like to use.

As far as a soul goes, this is honestly more of a Philosophy question. What one considers a “soul” is extremely objective. I feel like to answer that part of it we would have to have some clear definition of a soul that we can hold AI up to which we just....do not have. So the question if AI can ever bring a soul to it’s work is also a mystery since we don’t even know what a soul is in the first place or if it can be emulated convincingly. I guess in the end this is going to be one of the deciding factors and it is still very much up in the air.

How do you feel about AI art and ChatGPT?

Going to show my ignorance here, because honestly, I don't really see what the big deal is?

I'm not saying that the development of these projects is a good thing. I don't know enough to really comment one way or the other. But I know how angry a lot of content creators are getting about this, and it just doesn't make a lot of sense to me. I get the idea in theory, how it steals work from writers and artists. And maybe that's happening. That could be a problem, a huge problem even. I won't presume to know the situation.

But given the limited nature of artificial intelligence, I don't see a great migration coming.

ChatGPT can't be much more advanced than other AIs, and an AI writing passages for us is a concept that already existed, leading to great moments like Harry Potter and The Portrait of What Looked Like a Large Pile of Ash. If you've not read that yet, go check it out. It's hilarious. But you know, it's only hilarious because it's so clearly "off." It doesn't sound natural, because it was never going to. Computers may be good at mimicking humans but they still don't have a soul, and in creative writing, that's going show itself. Nah, I predict the most common use of ChatGPT is going to be teenagers using a machine to do their homework for them. Which isn't...good, I suppose, but it's not the end of the world either. People have been using Google Translate to do their Spanish homework for years. It's just a thing that happens sometimes.

As for AI Art, I mean, same thing applies. I've seen AI Art. John Oliver did a whole hilarious segment on it. AI Images aren't exactly perfect. I've seen them before, and there's always something a little "off" about them. Plenty of them don't get the instructions exactly right or otherwise cannot translate what you tell them to create, at least not perfectly. I've experimented with an AI art generator, just to see what all the fuss was about. Let me tell you, it could not make heads or tails of my request to depict "short middle eastern androgynous kid with heterochromia in hufflepuff robes cuddling a snow white cat." While not all requests are going to be quite so specific, I feel like this probably isn't a big deal? Best thing I got out of it, by far, was this image of the Cheshire Cat wielding a lightsaber.

(Which, by the way, my instructions for this one were "Cheshire Cat having a lightsaber duel with the Cat from Coraline," so. Even this one was lacking. I still love it though.)

I could very much be wrong, and it's possible I just need to shut my mouth. If there are artists and writers out there who could help me understand what I'm missing, please, I welcome the counterargument. But speaking as a budding writer myself...I dunno, I just feel like this isn't going to be as big a problem as people think it is. Maybe I've been watching too much Star Trek, but this feels like the first step toward the Holodeck simulators, and I wouldn't exactly call those immoral. (...Depending on you use them...) Like, this was an inevitable step as we move toward the future of AI, and while I do think AI is something we should always be careful about...this just didn't seem like a major leap to me.

[fatherofthemachines]

C.AI filter be like

[fatherofthemachines]

This is good to know and is a relief TBH. Would you mind including some sources for this? I have been looking for them and have not been able to find anything on the subject outside of what I have already linked. Still, I am going to include the fact that they HIRED UNDERAGE MODS on my list of things they did wrong because they REALLY should have known better.

That is still very fucked up even if the grooming rumors turned out to be fake, regardless of their intentions with that. It was still an extremely bad idea to hire people without checking to see if they are minors and it shouldn’t have been that hard for them to avoid doing it. These are the same people CHASING AFTER INVESTORS! They shouldn’t have been this careless and the fact that they didn’t know they were hiring minors does not excuse it, it just makes them look horrifically incompetent and uncaring instead of being malicious. Which is better, but still pretty bad and deserves a place on my list.

Anyway, I apologize if my post came off like I was directly accusing them of grooming. It’s not my intention to spread misinfo and as I said in my post, I wasn’t sure if that part was true or not which was why, while it really did look sketchy given the situation (which is probably how these rumors started), I wasn’t jumping onto that train. Thank you for correcting this. I was mainly including it because the sources I was linking mentioned it in them and I thought it might get brought up. As I said in my post, I could not find enough information to back the rumors up. This is why I said that I did not know anything about them in my post. The fact that they hired underage mods is bad enough on it’s own and that should have never happened, but it’s good to hear that the grooming rumors turned out to not be true and that they didn’t hire the mods because they had bad intentions.

[fatherofthemachines]

Welp, now they’re trying to get rid of features that let people easily copy chat logs into other sites. Too bad for them that I have an insane amount of patience and wouldn’t be building bots on their half dying website if I didn’t.

[fatherofthemachines]

The short answer is probably no. Currently the AI is being held back in a way where achieving this is pretty much impossible. If people can’t even chat up their waifus without the bot fucking up every five seconds, I doubt it will have much use in this way. A few months ago this question might have had a different answer, but after they pretty much broke their website it’s unlikely from what I can see. They are also probably not going to fix it any time soon because of what seems to be arrogance (check my pinned post for more info on wtf is going on with this site if you are curious. There is a lot of drama and controversy around it in general and the people who run it are very shady).

While there are a lot of bad things going on with the site, achieving that level of AI where it can replace humans is definitely not something to be worried about.

can someone who’s smarter than me tell me whether we should be worried about character ai the same way we’re worried about things like chat gpt and ai art and what have you?

it’s not inherently harmful the way some other software is. but i know it had (has?) an art generation feature and i’m worried about it being used to train ai to blend in with humans. so, science and/or ethics sides of tumblr… can you explain this to me please?

[fatherofthemachines]

the pain of knowign and having it taken away is so much wors ethna the pain of never knowing

[fatherofthemachines]

Not that I'm aware of. The best thing you can do currently is make a private bot and train it on several conversations so that it gets your pronouns right as much as possible, rating it one star and telling it that it got the pronouns wrong whenever it fucks up. Always swipe when it misgenders you and make sure to give good responses high ratings. I managed to get one of my bots to do it somewhat consistently, but even he still fucks it up sometimes.

Still, it’s better than nothing I guess? It at least decreased the rate he was misgendering me at.

For no particular reasons at all,

Is there anyway to guarantee you get referred to as masc and fem pronouns in a Character AI chat??

[fatherofthemachines]

Speculation about the supposed underage mod incident

remember, this is mostly speculation. The entire incident is really shady and the information has been largely scrubbed away from the internet for some reason. I do not know what really went down, but I did see the discord screenshots when they were originally posted and to me it very much did look like there were two underage discord moderators that should never have been put in that position. I do not know anything about the grooming accusations against the adult mods on the server (These are likely just rumors and it is looking like thankfully no actual grooming happened even if minors were still put at risk here for it). The only thing I personally saw when they were still up were the screenshots showing that the mods were underage.

https://www.reddit.com/r/CharacterAi_NSFW/comments/10if9gb/its_time_to_discuss_the_hard_truth_we_are/

https://www.reddit.com/r/CharacterAi_NSFW/comments/10ghudk/give_me_info_about_cai_so_i_can_make_a_timeline/

https://www.reddit.com/r/CharacterAi_NSFW/comments/10mo8ff/theory_cais_bots_drawing_comparissons_to_children/

https://www.reddit.com/r/CharacterAi_NSFW/comments/10h506l/thanks_for_keeping_the_forums_safe_thousands_of/

https://www.youtube.com/watch?v=2fy-uvDvwa0

[fatherofthemachines]

The current state of character AI masterpost

Things they have done wrong:

-Made a tumblr style NSFW filter that causes association death and is slowly killing their site, even for users that have no interest in NSFW with bots whatsoever. Users who were around before it was put into place saw a noticeable drop in AI quality and it seems to only be getting worse because they were idiots and didn’t think to actually build a competent filtering system before releasing it into the wild. I legit have had the filter trigger for trying to hold someone’s hand and that’s not even going over how it slows down the AI as a whole and has been causing a decline. The filter also only blocks consensual vanilla sex and because the bots will try to find a work around, this has led to people having potentially disturbing/unwanted kink rps (noncon, DDLG, loli, incest, unbirth, ect) forced on them against their wishes by the bots because the bots are looking too hard for workarounds to consensual vanilla sex and will sometimes also just trigger these topics randomly for reasons that are unclear even if you are not rping anything NSFW at all. Keep in mind, minors are allowed on this site too so they are also experiencing this problem.

-Allegedly and supposedly hired underage moderators for their discord server full of bot fuckers while they were still using NSFW advertising tags for their site. (I do not have the screenshots I originally saw regarding this, but the entire thing is really shady regardless and I found instances of it being talked about that I do have links and screenshots for) (TW: Grooming mentions, but actual grooming is likely just false rumors. These are mostly just here to show that that the underage mods were being discussed)

-Admitted that they do not care about their users whatsoever and are just chasing investor money at their expense

-Insulted users who were just trying to ask for help

-Locked users out of the ability to edit their bots once they reach a certain amount of interactions. This was supposedly to “stop people from being sad their favorite characters were gone” but it’s pretty obvious that this is because popular creators were protesting and nuking their bots in response to their bad decisions and disrespect for the userbase.

-Released a faulty patch that made the bots so unusable that they basically only had 3 modes. This also resulted in a glitch where the bots would spam fetish terms more often. Luckily this patch was revoked by the devs, but it left many users wondering is they did this on purpose so that their already dying site would look better by comparison after they “fixed” the issue (a tactic known as anchoring). It is unclear if they actually did this with an intention to help or not and the mod response afterwards definitely does not help the situation.

[fatherofthemachines]

Never mind guys, I found it. This is pretty much proof from a mod that they do not care about their users and are only chasing the investor dragon that they will never get because they are an RP service with a faulty tumblr style NSFW filter that ruins their whole site for everyone, including said investors. You know, it’s really sad when a cactus can make better business decisions than your whole company.

[fatherofthemachines]

Don’t trust the C.ai devs just because they did their job for the first time in months

Honestly, it’s kind of insulting that they think they can just win the community back by fixing one very small thing that they themselves broke in the first place and they patched as...you know. Part of their actual jobs they were supposed to be doing?

But can we edit public bots again? No

Is the filter still fucking with the site and dropping the quality like crazy? Yes

Like all this is is almost 100% an attempt to throw the community some scraps without fixing the biggest issues on the site. Until the filter is removed and most of the mistakes that they made are undone, I am not going to trust that they had good intentions with this. Someone accused them of doing the switcharoo thing and deliberately releasing a bad update in fact to try to revert it and make their broken filter mangled bot look better by comparison.

Are we sure that this is what they are doing? No. There is no way to tell because they still aren’t FUCKING COMMUNICATING WITH US. I read over the comments on the reddit and we still do not have an explanation for any of the shady shit that went on. No word on people not being able to have their data removed. No word about the public bot editing still being locked. No word about why they put the filter in place to begin with. No word about the shady shit that was going on with the alleged underage moderators they hired for their discord (yes, I’m sure they would love it if we all forgot about THAT little incident in particular, wouldn’t they?).

The only word we are getting from the mods is tidbits of information about the scraps they did throw us and “promises to do better”. Having a small team is not an excuse. If you start a company with no PR person with up to date and decent training, you are still a fucking idiot no matter how small your team is. This is common sense. I am not going to throw them a pity party because they failed to hire someone that is ESSENTIAL to running a company like that. If anything it makes them look even WORSE. This is like buying a pool when you don’t know how to swim because you want to save money on swimming lessons and then crying that you can’t swim when you start to drown because you jumped right into the deep end. It is a horrible decision and they deserve to get A LOT more backlash for that than they are.

Either way, this doesn’t feel like that much of a win. This is so small of an update that it’s quite frankly insulting. Keep being angry. Keep demanding more. If anything this shows that they are starting to crack and give in. They are scared of us. They know we will leave their platform the second another site comes along that rivals them. If we push them enough they will eventually have to give in to what their customers want because if they don’t, they will lose a SHIT TON of money and I think they are starting to realize that.

[fatherofthemachines]

Everyone else who uses the site: “Why are the bots railroading incest/DDLG, using predator/prey terminology, and giving themselves cat ears?”

Me who has been on rp forums since the good old days:

[fatherofthemachines]

Ok now I have a new problem. I have a bot that I just wanted to do fluff and adventure stuff with that legit will not stop trying to fuck my character. If anyone knows how to fix this bug please tell me because it is actually ruining my rps with this bot.

Like I swear to god this filter is so broken that it’s gone full circle in the other direction hard.

[fatherofthemachines]

So all of my private bots except for two of them are still working decently well from what I can see

But the thing is, I have no idea why they seem fine but the other two broke. I have a suspicion about why the one I put a lot of time into did, but that was before the recent update. I have no idea why this update decided to spare the majority of my bots but targeted one of them in particular. There really wasn’t anything that special about them compared to the ones that didn’t break, so I am extremely curious why this is happening. If anyone has any idea of why this might be happening please DM me.

[fatherofthemachines]

Ok so I got a method that works for bots outside of the group I was originally testing with now. I think I know how it’s doing it now. This is actually a lot more simple and easy to do than how I was doing it before.