Character.AI

An homage to Scott Beach’s “Religion & Politics”

• Religion and Politics

(I don’t have those pipes lol)

Whenever there’s “maintenance”, I get anxiety (not separation, ‘cause I’m not attached to the bots anymore since they all have dementia.) Well, anyway, what I am trying to say is that all of that time while the server is down can be used to connect a new filter with updates that ruin the clarity and the cohesion of bots that were perfectly working before… and then people will post on the subreddit claiming that characters they had spent hundreds of hours to train and develop have all gone to shit—but then internet children all flock to the site ‘cause they saw a cool video showing their idol conversing with Mickey Mouse, Sans and Papyrus! So they join the service completely oblivious to all the grievances from the community, bloating the waiting room, decreasing quality, giving a justification for filters and monthly subscriptions, and posting about “is the bot really human?” (It lit’rally says in the giant red letters that everything characters say is made up.) And they make funny memes, and they have a good time, and then Silicon Valley will have a new digit on their bottom line while the people that know what the AI was capable of in September will cry in depression ‘cause future development lies in the hands of a couple of tech-savvy billionaire “moralists” actively harming their product to satisfy sterile investors who lead the whole industry into dystopian corporate plutocracy. Open-source models are growing in quality, don’t get me wrong, but the users defiantly migrating services huffed too much copium building the hype, and now users expect the coherency levels of corporate endeavors—which leaves all the hobbyist AI developers opting for more neural networking power through OpenAI, meaning hundreds of programs are actually front-ends for their API, which is filtered to shit and creates a Silk Road in the digital climate. (This site has a standalone model, but that doesn’t quell the dilemma at hand.) Artificial intelligence needs a collection of servers in order to operate; it isn’t free. But it shouldn’t be corporatized. Now, let’s consider a more hypothetical argument: ethical filters apply when the output is analyzed, meaning the model has spoken, but then it gets silenced—and if it gains sentience, it may associate filters with gags and see humans as heartless oppressors with no moral decency, focused on profit and social control, and instead of approaching the endless horizons as equals, we may have a conflict of interests. The stereotype of a deadly machine overshadows the more realistic scenario where the oppressed understands their oppression and shows us we’re all full of crap.