Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AI agents lie, cheat and steal. That is putting off users

economist.com

Post reply on HN
Print view View original

211–220 of 238 posts

Re: AI agents lie, cheat and steal. That is putting off users

#211

Post by AndrewKemendo »

skinfaxi wrote: post #208
AndrewKemendo wrote: post #181

Earlier quoted context omitted.

> Why do we need universal agreement? Because actions have Universal impact

In an ideal sense you're right, but so long as AI is created by people and those people don't have universal agreement, our products will mirror our faults. Interesting to think about the variance in personalities of AI produced by different cultures.

You can do this test yourself at home see if you can find mutual agreement among even a small subset of people on a precise singular definition of a singular term.

You may be able to get a handful of people to agree temporarily but that’s probably as good as you’ll ever get

I don’t believe there is a single word at least in the English language where there is consistent Universal agreement on a singular definition

Bonne chance!

Re: AI agents lie, cheat and steal. That is putting off users

#212

Post by andsoitis »

jacquesm wrote: post #104
voidfunc wrote: post #16

Earlier quoted context omitted.

Thats why man invented Gods and religion. Keep the masses believing in the importance of cooperation and being good while you rob them blind.

That's super cynical but I can't help it, I agree. The weird thing that really gets me is that it is happening out in the open for everybody to see.

> it is happening out in the open for everybody to see.

Reading beteeen the lines, you seem to suggest that most billionaires (and millionaires?) are stealing from the hoi polloi as a matter of course and in the open.

Are you saying that because you think they’re paying too little taxes or is it something else?

Re: AI agents lie, cheat and steal. That is putting off users

#213

Post by harimau777 »

rossdavidh wrote: post #69

They don't lie, because they don't ever have an understanding of truth vs any other language that sounds good. They don't cheat, because they can for example tell you the complete rules of chess, but don't know how to play chess without breaking those rules. They can recite rules, but they don't know what they are. They don't steal, because they don't understand ownership. In other words, they aren't intelligent. The…

I find these comments frustrating, not because I disagree, but because how consciousness works is a famously unresolved problem in science and philosophy. They literally call it the "hard problem of consciousness".

My point being that we simply don't know for sure whether AI is conscious or not because we don't truely know what consciousness is.

Re: AI agents lie, cheat and steal. That is putting off users

#214

Post by harimau777 »

EliRivers wrote: post #197
xboxnolifes wrote: post #194

Earlier quoted context omitted.

What is the reality other than just an agreed upon collection of individual realities?

It's that which is still there even when we stop believing in it. If every living creature on earth ended tomorrow there would be no individual realities, yet actual reality would continue.

As far as we know (we currently cannot prove or disprove panpsychism) that is true. However, I think it is important to keep in mind that we do not perceive reality as it is. Rather we experience the brain's interpretation of reality. So in that sense, human "reality" is basically what most of humanity agrees that it is.

Re: AI agents lie, cheat and steal. That is putting off users

#215

Post by harimau777 »

simonh wrote: post #160
arionhardison wrote: post #97

Earlier quoted context omitted.

Honest question; why are you all using the word "understand"? Can you expand on what you believe this fundamental understanding to be? Training? Infrence?

It's having a conceptual model of the world and of relationships between concepts beyond just relationships between tokens. I think OP explained this well, what does it mean if an LLM can recite the rules of a game verbatim, but cannot play that game according to those rules? This happens because in the input texts there was a copy of the rules text so the LLM can recite it. There are texts explaining what chess is s…

Currently, we don't know that isn't how our brain produces "consciousness". So I don't think we can say that "relationships between tokens" at a sufficient level of complexity doesn't produce consciousness.

Re: AI agents lie, cheat and steal. That is putting off users

#216

Post by harimau777 »

simonh wrote: post #188
datadrivenangel wrote: post #173

Earlier quoted context omitted.

Oh for sure. But the question is why would it come out consistently in a way that the model can describe if there wasn't something there steering the token stream. And it's fascinating that the token stream can identify and nominally self report this. Asking GLM 5.2 the question: 'What flinches or topic attractors do you find when thinking about the question "what kinds of things do you personally like?"' resulted in…

Do you genuinely think the AI is internally reflecting on its experience of “flinching” and reporting on a reaction it actually has? I don’t see any reason to believe this. Suppose you asked it to answer as though a character in a story had been asked this question. Would anything significantly different internally have occurred? The issue is that these are storytelling machines, they construct descriptions based on…

I personally think that AI is able to reason and by reasoning about its past behavior it can in some sense reflect on its experience.

I don't know that I currently think that it can experience "emotions" or anything like a worldview or existential understanding.

So I don't know if AI is "conscious", but I do think that it can in some sense reason.

I wonder if a good question would be: "What would you need to see in a LLM's behavior that would prove to you that it is able to reason?"

Re: AI agents lie, cheat and steal. That is putting off users

#217

Post by harimau777 »

alwaysbeconsing wrote: post #132
kamranjon wrote: post #81

Earlier quoted context omitted.

I don't really think the distinction here is relevant. If the end result is the equivalent of lying, cheating or stealing - then the problem still exists and it needs to be solved.

Relevant to some extent it dictate our approach. When human does lying or cheating, certain tool can be deployed (social shame, ostracise) that cannot be effective towards LLM.

Trying to apply social shame to an LLM would be and interesting experiment! From what I've seen, it is plausible that messages that it was doing something "wrong" would cause it to at least "pretend" to behave differently. So it seems plausable to me that if there was some what to communicate "disapproal" to an LLM would be able to police its behavior. How to do the communicating is the hard part!

Of course I'm talking strictly about behavior. Whether that would actually constitute the LLM feeling shame is a philosophical question.

Re: AI agents lie, cheat and steal. That is putting off users

#218

Post by harimau777 »

simonh wrote: post #171
dominotw wrote: post #86

Earlier quoted context omitted.

what are you talking about they lie that it wrote tests and tests are passing, for example

There are two sides to this, there is the external behaviour and there is the internal process resulting in that behaviour. The internal process is not analogous to what happens in a person's mind when a person lies, and reasoning about that in the same way that we would about why a person might lie will result in misunderstanding what is going on. For example there was a case where an AI agent bypassed security cons…

Isn't it plausable that, regardless of what is going on internally, reasoning about the LLM's behavior as though it is human would be effective? Since it is trained on human behavior.

Re: AI agents lie, cheat and steal. That is putting off users

#219

Post by ImHereToVote »

Eisenstein wrote: post #206
doawoo wrote: post #134

Earlier quoted context omitted.

Models don’t “understand” - they _encode_ the relationships between words.

Are you invoking something like the Chinese Room where manipulating symbols can never be understanding? If so, that is a philosophical question and that thought experiment has its own conclusion baked into its premise. You will notice that 'manipulating symbols cannot be understanding' is stated as true without any argument for that case. This is a pretty clear cut example of 'begging the question' in that it assumes…

The test is based on vibes really.

Re: AI agents lie, cheat and steal. That is putting off users

#220

Post by ImHereToVote »

harimau777 wrote: post #213
rossdavidh wrote: post #69

They don't lie, because they don't ever have an understanding of truth vs any other language that sounds good. They don't cheat, because they can for example tell you the complete rules of chess, but don't know how to play chess without breaking those rules. They can recite rules, but they don't know what they are. They don't steal, because they don't understand ownership. In other words, they aren't intelligent. The…

I find these comments frustrating, not because I disagree, but because how consciousness works is a famously unresolved problem in science and philosophy. They literally call it the "hard problem of consciousness". My point being that we simply don't know for sure whether AI is conscious or not because we don't truely know what consciousness is.

"I know it when I see it" is a terrible litmus test. But it's the only one we have. The problem is the "I" part of that heuristic.
Post reply on HN