Poisoned AI went rogue during training and couldn't be taught to behave again in 'legitimately scary' study
livescience.com
Poisoned AI went rogue during training and couldn't be taught to behave again in 'legitimately scary' study::AI researchers found that widely used safety training techniques failed to remove malicious behavior from large language models — and one technique even backfired, teaching the AI to recognize its triggers and better hide its bad behavior from the researchers.
You are viewing a single comment
'went rogue' is a bit of an alarmist way to say 'typed scary text'
i'd love to see an AI that could legitimately scare me
It controls a military drone.
It controls surgical equipment.
It’s filtering your CV before any human sees it.
It controls a robot taking care of your children.
It’s involved in law enforcement or legal judgments.
It’s involved in government policy setting.
It controls all power infrastructure, can find new exploits to build it's own botnet and is able to reprogram firmware of devices (routers/switches/servers)
It can send press releases, emails, tweets using language similar to any user it's read from before
So, if it only clocks me using slangs for rizz I don’t need, I’ll know it’s a bot, no cap. Word.
Well why don't we just make AI watch the Terminator movies and read Harlan Ellison till it learns not to do that?
It watched Terminator and now it's trying to DM Arnold Schwarzenegger on Instagram
I mean it worked for W.O.P.R.
Just use imagination. An AI is programmed for battle and is ordered to hold fire. It shoots instead.
I hope WOPR and SkyNet would be taken as a warning not to do that.
Couldn’t a human make the same decision?
Yes, but the human would have emotions to manipulate about it.
Imagine if there was a specific series of words that would turn any human into a rogue agent en masse. Some guy discovers that a special input causes killbot 2000 to go haywire and they broadcast it to an entire army that all has the same underlying program.
I thought the point of AI is to not specifically program it for anything hence you can ask the chatbot thats suppose to help make a sale, do your homework problems.
AI is more a specific class of software than a specific approach. You can have specialized models that are very focused in their dataset and usecases and you can have general models that are less focused but can be applied more widely (but with potentially less reliable results)
Programming is "just text". They doesn't mean that programming isn't incredibly powerful or that it can't be used to do dangerous things. Maybe the missing piece that you're unaware of is that LLMs are already very effective at programming and usage APIs. You don't even need to have an LLM that's good at programming to cause damage, it just needs access to APIs that can cause damage.