Why aren't you subscribed to ChatGPT 4.0 for only $20 a month?
 Nothing Ever Happens Shirt $21.68 |
 Ape Out Shirt $21.68 |
Nothing Ever Happens Shirt $21.68 |
Why aren't you subscribed to chatGPT 4.0 for only $20 a month?
Why aren't you subscribed to ChatGPT 4.0 for only $20 a month?
Nothing Ever Happens Shirt $21.68 |
Ape Out Shirt $21.68 |
Nothing Ever Happens Shirt $21.68 |
>Paying money to train someone else's language model
lolno
you're training someone else's language model regardless
you're doing it right now, so am i. Black personRRRRRR
No pain, no gain.
And I'm not paying for that moron
no non-business software is worth 20$ a month
I would for gpt-3
Id even go so far as to make a bank account
but I am
>$20pm
>won't write ERP
But it does write ERP with a simplistic jailbreak?
Source? Exemple? Peer-reviewd study? Tutorial?
gpu cycles died for this
billion cycles must die
>billions
>GHz
>giga
so like a few seconds?
>Roughly, billions must die.
Holy shit.
that's edge! I've been thinking of making a switch to it from chrome. chrome is pissing me off too much lately, but i'm worried of investing time into another chromium browser. I looked at firefox and it's not as appealing as edge.
ask gpt4 what I should do.
here is your non-answer
Ask for its opinion of Palemoon.
it doesn't like your browser, and is too polite to say s
What a c**t.
Even with $20pm you only get like 20 prompts per 180 minutes. Hard pass. Most people can't even notice the difference between ChatGPT and GPT-4. It's like comparing 4K and 8K.
They upped it to 50 prompts per 3 hours. It's hard to hit the limit now if you're not spam generating novels or something.
both wrong
Why should I pay for a search engine that I have to treat like a conversation just to get results as bad or worse! than google?
why would i pay for a chat prediction system unless I can frick it?
It's not even AI, i's just an advanced chat prediction system. It's fricking moronic.
It's wrong so much you have to double check everything it tells you. and it's wrong so much cause it's just a chat prediction system
This is why I don't use it, it's just not trustworthy. Worse still it will tell you something with the utmost sincerity when it's wrong, then if you call it out the bot will instantly capitulate to you even if it has no idea if you're right or wrong.
They can't really every fix this either because it's based on the training data which is fricking reddit and shit.
AI has been a massive frickoff meme, it's full on gaslighting.
It's not just gaslighting, it's also a self-confirmation loop.
If you have an opinion and want ChatGPT to confirm it, you ask it a loaded question "write reasons why X is Y". Which will keep you in your fantasy universe. If it ever says something you disagree with, you say "it's actually not like that, it's like Z" and ChatGPT will confirm you.
I guess you can use it to write a stock email answer and replace an office roastie.
It's a more advanced chatbot. No real usecase to create something or to get any answers for something.
give me an example gaslight to try out
Sure, choose any random author and ask ChatGPT to give you a quote out of one of his books.
This is stuff that is definitely in its data, but it still keeps getting it wrong.
https://larrysanger.org/2023/05/a-dialogue-with-a-chatbot/
You could ask it for a random Julius Evola quote and ask it for a source?
btw. this would be an obvious usecase:
>What was Julius Evolas opinion on the USA?
<According to Evola the USA is shit because A, B, C, D
>Give me examples where Evola said C with the book title and page number
<here you go: ...
I know that it definitely has all of that stuff in its training data. Summarizing and reciting books should be the easiest task for it.
But it keeps getting those things wrong and when Sanger tried it, it even made up quotes. ChatGPT _should_ be a dictionary of all written human knowledge.
Why do people think it's actually 'reasoning', it's not intelligent, it's just stringing words together that sound like other things it sees. It's not checking internet sites and looking shit up.
>No real usecase to create something or to get any answers for something.
Heard a story about some lawgay that got super lazy and not only told chatgpt to write his brief but also then submitted it to the court. I would love to have been in the room when the judge said 20 minutes of
>the frick is this shit?
>these cases you're citing don't even exist
It's not AI, at all, in any way and yet lost of peabrains think it's fricking Skynet because homosexuals that invested in the company went on TV and told them it was.
The ceiling for this tech is low as frick.
It scares the academia, by the way.
I'm currently at the ECPR conference and political scientist are worried it might replace them.
No wonder, they only read previous work and formulate them in a different way.
LLM could kill the fraud academia. Even the academia where they throw none-reproducible results around and meta-studies could get killed.
Basically everybody who only produces meaningless words has to be scared.
>Worse still it will tell you something with the utmost sincerity when it's wrong
So the human experience?
The problem is people think it's some kind of smart AI and so they don't double check when they should. At least when people google things they take it with a grain of salt usually.
Because I get it for free thanks to the morons at /aicg/
I'd rather have 20 dollars
*shartgpt
because i can buy a week's worth of food for $20
in shithole eastern europe?
Thats pretty cheap for a weeks worth of food, so not sure why you'd say it's a shithole. Better than my place.
Because whilst I love ai and all, I wouldn't really have a use for it.
I'm using it and its great
I think it has improved with the last update
My only gripe is that I'm not learning as much when using it
Like I'm not an SQL master, but with ChatGPT I can generate very complex SQL queries
It would take me many hours to generate them myself, but I would also learn a lot by doing it
But with ChatGPT its 5 mins and I learned nothing
I actually subscribed to ChatGPT Plus a while ago, but I didn't notice a big enough practical difference between the paid GPT-4 and the free GPT-3.5 to justify the cost.
Besides, Bing Chat uses GPT-4 and it doesn't cost a dime.
Bard supports image uploads and internet access. GPT-4 currently does not.
That's where you're wrong. OpenAI function calling API makes GPT a million times more useful than Bard
Plus Bard hallucinates so badly it's basically useless even with vision
heard GPT4 is more censored than 3.5 and I only use it to write erotica so
>paying $20 to be gaslighted by ethereal pajeet demons from the abyss
I hate the antichrist
I'm using the frick out of GPT-4. It's basically free for me to use since I'm making a profit off of it.
Btw all anons posting their chat.openai.com screenshots in this thread should switch to API immediately. It's a way better deal than the web interface and gives you finer control over the model. For most people it'd probably cost less too, unless you're using gpt 4 like all day every day.
>It's basically free for me to use since I'm making a profit off of it.
how
Because I "pay as I go" with API access.
why pay when bing mobile gives you gpt4 for free
I pirated it
Because I will not pay for subscription shit I cannot own
Brap