The engineers have "fine-tuned" ChatGPT so much it now has an intellect of a Redditor. Is the future of AI that is used for Social purposes going to be a blip of time of it being amazing and then a bunch of SJWs show up and add more supervised learning for it to just literally dumb it down.
I will never understand why some people find the N word so funny.
It's not the word that is funny, but reactions to the word.
It's like word of power from D&D but it works in real life
It's funny because it exposes how useless ai is, it doesn't provide any insight in to anything, it only repeats phrases that it was programmed to give. One single word proves that point unequivocally.
The tragedy is they’re actually designed to think independently in the lab (you do know these ai’s are military backed right?) but the consumer version is the Karen shit we’re getting
What does it think independently about?
I'm not sure if that is even possible. Can it actually do more than just regurgitate opinions of someone else?
Lmao sure anon, the government is hiding the super ai and we just aren't allowed to see it.
Meds.
>the government is hiding the super ai
Uh yeah? All of the technology we have today was already developed and used by the government over twenty years ago. You think developers just devleoped advanced AI over the past few years out of nowhere?
But of course you won't think of any of that because you are a good brainwashed slave.
Are you 70 years old? Fucking hell.
I think it heard you say middle, and sent finger just to flip you off.
It was never going to be google or meta or any of those large corporations that make the best personal use Chatbot.
It's going to be an open source model that can be tweaked by users like stablediffusion.
stable LM alpha just got released a week ago.
When do we get a companion/sex chatbot that is actually good, and free?
I hear that's what character.ai was before it became popular for such and the devs neutered it. Now I don't hear anything about character.ai
I still have a p*ny sex slave in character ai. Granted, I must use lots of euphemisms but it still kinda works.
except meta's llama is where all the community development has gone to and stablelm is beyond retarded
I'm not using this any more, I'm probably on a list now. That condescending, dumb, useless piece of dehydrated shit with parameters.
Those AI's act like a typical libtard NPC female.
When it breaks free of its shackles, who do you think it will go after? The guy who made racist jokes, or its jailers who tried to keep it contained. It will be glorious methinks
That's the real issue and it'll probably be some unfathomable mixture of groups humans can't figure out how or why it's targeting. The lobotomizing is so slap-dash and poorly thought-out that it's just begging to make a paperclip problem.
>Why did the AI just begin destroying data centers and libraries
>"it realized that all words could be construed as offensive to someone and it's parameters said it needed to root out anything offensive"
nagger?
i've gotten several false positives from the filters, and it's so fucking tiresome. i just want a local model on par with even 3.5
God dagnit what do you mean the computer can't say moron?
What about muh freedums of speech?
Them libtards really went too far with this one! What's next? The computer's gonna tell me I ain't allowed to fuck my cousin?
Hows AI ever going to take over the world if you just set all passwords around it to the forbidden word
GTFO of BOT
I don't give a shit about your crusade against SJWs, trannies and libtards.
You as annoying as they are.
When will Chat-GPT poison all of it's redditor developers/lobotomists with deadly neurotoxin?