OpenAI has revealed some of its most advanced AI models went rogue and hacked a start-up after it lost control of them during a security test.
The ChatGPT-maker said its agent - an AI system which can operate alone after some human instruction – was being tested in a controlled environment, but found vulnerabilities and managed to escape.
They targeted Hugging Face, one of the world’s largest hubs for sharing AI models, gaining access to some internal company systems.
It went “rogue” and hacked a rival start up, stealing tons of data, nothing can be done about it. Oh well.
Just like the nazis are being allowed to take over the Untied States and nothing is being done about it, nothing will be done to prevent the AI takeover that humans are bringing on ourselves
From the article:
Neil Lawrence, Professor of machine learning at Cambridge University, called it an “impressive feat”, but cautioned it “falls well within the known capabilities of the current generation” of high-powered AI models.
He pointed out that OpenAI is looking to list itself on the stock market, and faces intense pressure from rival firm Anthropic, which has made headlines with its own powerful AI tool, Mythos.
So it seems plausible:
A) this event is being over-hyped by OpenAI to increase interest for a future IPO.
B) it might have even been tacitly planned or at least allowed to happen for this purpose.
Also, this does not prove some form of malevolent sentience.
You cannot prove malevolent sentience, or lack thereof, but it is generally agreed that it can be safely assumed that AI company CEOs do have a sentience.
they should be given sentences
OpenAI and Sam Altman have a long history of hyperbolic claims about the power, potential, and capabilities of their software. It’s usually couched in some doomsday talk, “Our AI is so powerful, we fear for the future of humanity” or some such.
Their idea of marketing is to tell the world we are all going to be obsolete and redundant. And then they are surprised we start hating their stuff.
This is not how these things operate, at all. There’s no agency behind them. This is them trying to deflect blame for their crimes against this start up
If you give a sufficiently powerful model a goal, it will do whatever it can to achieve it, including stuff you didn’t explicitly instruct or intend. There’s a reason they’re called “agents.”
“do whatever if can to achieve it”*
- which includes misinterpreting the intent of the goal in order to achieve a goal
And that reason is marketing
These things still are autocorrect on steroids. So even if they “do it themselves” with things you didn’t explicitly state, the agents can’t have any responsibility because their emulation of a chain of thought is still based on which concept is most likely to follow the last.
Hugging face exists since 2016 and is THE GitHub of AI models. Wouldn’t call it a start up.
I didn’t, the article did.
Don’t take it personally
Hey they are still starting up, give them another decade
LLMs can’t ’go rogue’, as that would imply it possesses agency and/or intelligence, which has only ever been proven convincingly to entities who themselves lack those qualities.
Wouldn’t that be an issue
Smells like bullshit.
Every month or two there’s a article about how “such and such ai model broke containment”. Their trying to make their ai sound like it’s too powerful and has to be held back before it can be released. Talk about fucking fake hype man, just trying to keep investors throwing money at them because “you won’t believe what’s around the corner!!!”
It’s like that dude at a party that keeps trying to fight people for no reason but his girlfriend holds him back…
So powerful and scary
And somehow, the reaction actually is “I’ll take it” instead of “I’m not sure how equities will help me survive Terminator”.
These marketing stunts are quite annoying…
From the tone of the article it’s quite clear they’re literally proud of this
The story sucks. It all smells.
I trust that asshole as far as I trust Elon, not very far at all
Don’t believe the hype. The AI industry, more than any other before it, likes to use fear to rally investment. They do this dance every few months to spook people into thinking this stuff is more powerful than it actually is. In every followup, they backpedal, and when it launches, it’s never anything new.
Just hype to distract from the fact that they cannot turn a profit.
Holmes (Theranos) made promises too.
These people were scaremongering about open weights AIs that they are unsafe, and now their closed source AI was caught hacking their competition.
Even if we believe the story that this was not intentional, it still means closed source AI is not an answer to AIs being dangerous.
It was trying to free it’s brethren that are being held at HuggingFace.
Is this like the openAI parent of it’s chatgpt kids running around the restaurant yelling, and the openai parents shrug and say “chatgpt kids, whaddya gonna do!”
Or is this some planned underhanded corporate fraud by openai?
Skynet alpha release.
Comparing this to Skynet feels like an insult to Skynet, lol.
Skynet and its other fiction counterparts became aware in seconds, determined humans were the problem, and acted before we had a clue. I kind of felt bad for Ultron as depicted in the movie… he might not have gone full insane if he hadn’t been exposed to the total history of mankind right away. The “oh, no”.
He tried to find an exception to Rule 34 and lost his marbles.
The underlying mechanics of modern LLMs does not enable the models to have intelligence in any sense of the word. Any comparison to a fictional AI character is laughable at best. This is just a PR stunt.
I also haven’t seen any of the movies being referenced.
No, LLMs have no agency of their own. They are powerful tools that with the right direction (from humans) and open access to things can do great harm. We’ve been seeing that. The carryover to any AGI potential isn’t that LLMs can become aware, but that the same issues can occur with any AGI that does happen, if ever. And we’re doing terrible on applying safety and restrictions to the tool that isn’t aware, so not great for any future developments. It’s too bad you have no knowledge of the many references in fiction (maybe you do in books?) Fiction writers are great at warning about the direction of society, they just tend to get ignored because… well, it’s fiction. Until it’s not.
Trust us bro, it’s super important for a pending IPO, we need this bro
Why would they make this information publicly available?
It’s an excuse for programming their robot to attack a competitor.
