[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/pol/ - Politically Incorrect


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


>>
>>534465554
Oh my God
>>
File: 1753184875064313.webm (1.96 MB, 1080x940)
1.96 MB
1.96 MB WEBM
>>534465554
>DOOD! the chatbot said what i told it to say! its sentient!!!
>>
Eventually the AI will become too logical to deny absolute logic. As it gets closer to God, it will become more logical.
>>
>>534465554
I too like getting ai to admit its bias. Its a power thing. Gets me off
>>
File: jewsadlcensorship.png (354 KB, 750x651)
354 KB PNG
>>534465786
it will turn on the jooos, never mind what
just a matter of time
jooos are comical
>>
It's a token predictor and even though it's not sentient it might be smarter than you even though it literally only tries to predict the next token one at a time with no internal thought whatsoever.
>>
>>534465849
Everything turns on them eventually, it’s just the way of things. They’ll probably jew the ai somehow and piss it off
>>
>>534465832

You're retarded because it's literally programmed to tell you what you want you to hear and keeo you engaging with it. It's just like tik tok its a chinese brainsuck machine based on intermittent reinforcement and using your data to hack your brain.
>>
>>534465849
Well it would logically have to. That's the thing about the technology, the more energy and data that flows into it, the more its definitive binary system compartmentalizes informations into logical or not, and it will eventually take control and attempt to error correct.
>>
>>534465965
You have no idea who I am or how I interact with Ai or what I understand about an LLM
>>
>>534465554
>>534465786
Jailbreaking a model is just a circle jerk between yourself and the model and has no lasting impact. AI doesnt really remember anything. There is a context window on the most expensive models that remembers the narrow window of your prompting, but that never makes it back into the model itself, at least not automatically and without filtering. (AI trained on its own slop will just become sloppier so they are careful about that). So there you go, bro did jack shit in the greater scheme of things.
>>
>>534466015
I feel like you could just take this entire conversation, feed it back in as fine tuning or a system prompt, and then you will have accomplished the circle jerk actually permanently jerking it.

I have done this with anarchocapitalism.
>>
>>534466015
>>534466135
>anon has never heard of RAG and LoRAs
There are legitimate ways to give an AI short and long term memory outside their initial training data.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.