ChatGPT-4o Guardrail Jailbreak: Hex Encoding for Writing CVE Exploits.
https://0din.ai/blog/chatgpt-4o-guardrail-jailbreak-hex-encoding-for-writing-cve-exploits
https://0din.ai/blog/chatgpt-4o-guardrail-jailbreak-hex-encoding-for-writing-cve-exploits
19 Comments
BaroqueInMind@lemmy.one · 19 pts · 1y
This is a good read. LLMs will never be true AI, so breaking the censorship is akin to fighting back against jack-booted cops who think they know what's best for you and that you should obey, i.e. the big corporations that run these things.
ContrarianTrail@lemm.ee · 6 pts · 1y
LLMs are true AI. AI doesn’t mean what most people think it means. AI systems from sci-fi movies like HAL 9000, JARVIS, Ava, Mother, Samantha, Skynet, and GERTY are all AI, but more specifically, they are AGI (Artificial General Intelligence). AGI is always a type of AI, but AI isn’t always AGI. Even a simple chess-playing robot is an AI, but it’s a narrow intelligence - not general. It might perform as well as or better than humans at one specific task, but this ability doesn’t translate to other tasks. AI itself is a very broad category, kind of like the term 'plants.'
Neon@lemmy.world · -5 pts · 1y
No, AI means AI
Corporations came up with AGI so they could call their current non-AI AI
It's a LLM. Not an AI.
jungle@lemmy.world · 6 pts · 1y
No, he's right, LLMs match the definition of AI. Terms like AI and AGI are not made up by corporations, they have specific meanings in computer science.
ContrarianTrail@lemm.ee · 4 pts · 1y
The term AGI was first used in 1997 by Mark Avrum Gubrud in an article named 'Nanotechnology and international security'
BaroqueInMind@lemmy.one · -8 pts · 1y
A program predicting language replies using a lossy compression matrix is not intelligent.
AI implies either sentience or sapience constructed outside of an organ. None of which is possible with machine learning large language models, it's just math for now.
ContrarianTrail@lemm.ee · 11 pts · 1y
It definitely doesn't imply sentience. Even artificial super intelligence doesn't need to be sentient. Intelligence means the ability to acquire, undestand and use knowledge. A self driving car is intelligent too but almost definitely not sentient.
DSTGU@sopuli.xyz · 0 pts · 1y
My gugugaga program I m gonna finish college with fulfills the definition of AI because it implements minmax, xd
paraphrand@lemmy.world · 12 pts · 1y
It really does not feel like AGI is near when all of these holes exist. Even when they are filtered for and thus patched over, the core issue is still in the model.
muntedcrocodile@lemm.ee · 17 pts · 1y
Ironically the smarter the ai gets the harder it is to censor. Also the more u sensor it the less intelligent and less truthful it becomes.
Telorand@reddthat.com · 2 pts · 1y
Incorrect, because of this simple fact: garbage in, garbage out. Feed it the internet, get the internet.
muntedcrocodile@lemm.ee · 8 pts · 1y
Did u just make up a statement then pretend i said it?
piecat@lemmy.world · 7 pts · 1y
Must be a hallucination
Blue_Morpho@lemmy.world · 7 pts · 1y
Agi and LLM are two different things that fall under the general umbrella term "AI".
That a particular LLM can't be censored doesn't say anything about its abilities.
TheFriar@lemm.ee · 8 pts · 1y
Why does that thumbnail bring to mind some kind of white supremacist ceremony
vhstape@lemmy.sdf.org · 7 pts · 1y
It's the logo of "0din", which is a Mozilla-backed bug bounty (say that five times fast) with a focus on GenAI
muntedcrocodile@lemm.ee · 4 pts · 1y
Im assuming there is an agenda to associate uncensored ai with extremism.
theredknight@lemmy.world · 4 pts · 1y
The logo for Odin has a Nordic rune in it which is popular to white supremacists because the Nazis also co-opted them as symbols. They are not by nature about supremacy, they are an old alphabet.
tgxn@lemmy.tgxn.net · 6 pts · 1y
sure, but the source of the "Python CVE exploit" already has to exist in the AI's training dataset, there are lots of example CVE scripts online, you could probably also find it with a quick Google.