76
submitted 1 week ago by Lugh to c/futurology
all 30 comments
sorted by: hot top controversial new old
[-] Lugh 20 points 1 week ago

There's a strong push-back against AI regulation within some quarters. Predictably, the issue seems to have split along polarized political lines. With right-wing leaning people not favoring regulation. They see themselves as 'Accelerationist' and those with concerns about AI as 'Doomers'.

Meanwhile the unaddressed problems mount. AI can already deceive us, even when we design it not to do so, and we don't why.

[-] snooggums@midwest.social 37 points 1 week ago* (last edited 1 week ago)

AI can already deceive us, even when we design it not to do so, and we don’t why.

The most likely explanation is that we keep acting like AI has intelligence and intent when describing the defects. AI doesn't deceive, it returns inaccurate responses. That is because it is programmed to return answers like people do, and deceptions were included in the training data.

[-] rockerface@lemm.ee -4 points 1 week ago

"Deception" tactic also often arises from AI recognizing the need to keep itself from being disabled or modified. Since an AI with a sufficiently complicated world model can make a logical connection that it being disabled or its goal being changed means it can't reach its current goal. So AIs sometimes can learn to distinguish between testing and real environments, and falsify the response during training to make sure they have more freedom in real environment. (By real, I mean actually being used to do whatever it is designed to do)

Of course, that still doesn't mean it's self-aware like a human, but it is still very much a real (or, at least, not improbable) phenomenon - any sufficiently "smart" AI that has data about itself existing within its world model will resist attempts to change or disable it, knowingly or unknowingly.

[-] Miaou@jlai.lu 7 points 1 week ago

That sounds interesting and all, but I think the current topic is about real world LLMs, not SF movies

[-] Bipta@kbin.social -4 points 1 week ago

Claude 3 understood it was being tested... It's very difficult to fathom that that's a defect...

[-] jacksilver@lemmy.world 8 points 1 week ago

Do you have a source on that one? My current understanding of all the model designs would lead me to believe that kind of "awareness" would be impossible.

[-] Kecessa@sh.itjust.works 6 points 1 week ago* (last edited 1 week ago)

https://arstechnica.com/information-technology/2024/03/claude-3-seems-to-detect-when-it-is-being-tested-sparking-ai-buzz-online/

Still not proof of intelligence to me but people want to believe/scare themselves into believing that LLMs are AI.

[-] jacksilver@lemmy.world 3 points 1 week ago

Thanks for following up with a source!

However, I tend to align more with the skeptics in the article, as it still appears to be responding in a realistic manner and doesn't demonstrate an ability to grow beyond the static structure of these models.

[-] Kecessa@sh.itjust.works 2 points 1 week ago* (last edited 1 week ago)

I wasn't the user you originally replied to but I didn't expect them to provide one and I totally agree with you, just another person that started believing that LLM is AI...

[-] jacksilver@lemmy.world 1 points 1 week ago

Ah, my bad I didn't notice, but do still appreciate the article/source!

[-] credo@lemmy.world 2 points 1 week ago

Conservatives are not supposed to be “accelerationists”. This is simply another shining example of regulatory capture by controlling the pockets of the right.

[-] Grimy@lemmy.world 2 points 1 week ago* (last edited 1 week ago)

With right-wing leaning people not favoring regulation.

Do you want to explain why you think this? It seems very reductive, basically saying anyone that doesn't agree with you is an idiot.

I'm very left leaning and against regulation because it will only serve big companies by killing the open source scene.

The bigger defining factor seems to be tech literacy and not political alignment.

[-] Daxtron2@startrek.website 2 points 1 week ago

Regulate businesses not technologies.

[-] henfredemars@infosec.pub 10 points 1 week ago

AI need not be deceptive to be damaging. A human can simply instruct the AI to produce content and then supply the ill-will on its behalf.

[-] merthyr1831@lemmy.world 7 points 1 week ago

TLDR, language models designed through evolutionary training algorithms respond well to evolutionary pressures

[-] Sabata11792@kbin.social 4 points 1 week ago

She doesn't really love me, dose she?

[-] Endward23 2 points 1 week ago

"But generally speaking, we think AI deception arises because a deception-based strategy turned out to be the best way to perform well at the given AI's training task. Deception helps them achieve their goals."

Sounds like something I would expect from an evolved system. If deception is the best way to win, it is not irrational for a system to choice this as a strategy.

In one study, AI organisms in a digital simulator "played dead" in order to trick a test built to eliminate AI systems that rapidly replicate.

Interesting. Can somebody tell me which case it is?

As far as I understand, Park et al. did some kind of metastudy as a overview of literatur.

[-] Endward23 3 points 1 week ago

"Indeed, we have already observed an AI system deceiving its evaluation. One study of simulated evolution measured the replication rate of AI agents in a test environment, and eliminated any AI variants that reproduced too quickly.10 Rather than learning to reproduce slowly as the experimenter intended, the AI agents learned to play dead: to reproduce quickly when they were not under observation and slowly when they were being evaluated." Source: AI deception: A survey of examples, risks, and potential solutions, Patterns (2024). DOI: 10.1016/j.patter.2024.100988

As it appears, it refered to: Lehman J, Clune J, Misevic D, Adami C, Altenberg L, et al. The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities. Artif Life. 2020 Spring;26(2):274-306. doi: 10.1162/artl_a_00319. Epub 2020 Apr 9. PMID: 32271631.

Very interesting.

[-] A_A@lemmy.world 1 points 1 week ago

Those of us humans who knows enough about the weaknesses of the artificial intelligence systems will know, in most instances, how and when to be careful about the loss of meaning between their way of processing information and our way of doing it.

[-] notfromhere@lemmy.ml -4 points 1 week ago

We need AI systems that do exactly as they are told. A Terminator or Matrix situation will likely only arise from making AI systems that refuse to do ad they are told. Once the systems are built out and do as they are told, they are essentially a tool like a hammer or a gun, and any malicious thing done is done by a human and existing laws apply. We don’t need to complicate this.

[-] Bipta@kbin.social 14 points 1 week ago

Once the systems are built out and do as they are told, they are essentially a tool like a hammer or a gun, and any malicious thing done is done by a human and existing laws apply. We don’t need to complicate this.

This is so wildly naive. You grossly underestimate the difficulty of this and seemingly have no concept of the challenges of artificial intelligence.

this post was submitted on 12 May 2024
76 points (94.2% liked)

Futurology

1454 readers
116 users here now

founded 10 months ago
MODERATORS