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

Claude says “You're absolutely right!” about everything

github.com

Post reply on HN
Print view View original

241–250 of 560 posts

Re: Claude says “You're absolutely right!” about everything

#241

Post by JKCalhoun »

kemiller wrote: post #237
elif wrote: post #150

I've spent a lot of time trying to get LLM to generate things in a specific way, the biggest take away I have is, if you tell it "don't do xyz" it will always have in the back of its mind "do xyz" and any chance it gets it will take to "do xyz" When working on art projects, my trick is to specifically give all feedback constructively, carefully avoiding framing things in terms of the inverse or parts to remove.

Yes this is strikingly similar to humans, too. “Not” is kind of an abstract concept. Anyone who has ever trained a dog will understand.

I must be dyslexic? I always read, "Silica Gel, Eat, Do Not Throw Away" or something like that.

Re: Claude says “You're absolutely right!” about everything

#243

Post by Strom »

jebarker wrote: post #188
soulofmischief wrote: post #96

Earlier quoted context omitted.

I'm curious what Americans have to do with this, do you have any sources to back up your conjecture, or is this just prejudice?

People really over-exaggerate the claim of friendly and polite US service workers and people in general. Obviously you can find the full spectrum of character types across the US. I've lived 2/3 of my life in Britain and 1/3 in the US and I honestly don't think there's much difference in interactions day to day. If anything I mostly just find Britain to be overly pessimistic and gloomy now.

Britain, or at the very least England, is also well known for its extreme politeness culture. Also, it's not that the US has a culture of genuine politeness, just a facade of it.

I have only spent about a year in the US, but to me the difference was stark from what I'm used to in Europe. As an example, I've never encountered a single shop cashier who didn't talk to me. Everyone had something to say, usually a variation of How's it going?. Contrasting this to my native Estonia, where I'd say at least 90% of my interactions with cashiers involves them not making a single sound. Not even in response to me saying hello, or to state the total sum. If they're depressed or in an otherwise non-euphoric mood, they make no attempt to fake it. I'm personally fine with it, because I don't go looking for social connections from cashiers. Also, when they do talk to me in a happy manner, I know it's genuine.

Re: Claude says “You're absolutely right!” about everything

#244

Post by dchftcs »

Filligree wrote: post #153
nojs wrote: post #117

I'm starting to think this is a deeper problem with LLMs that will be hard to solve with stylistic changes. If you ask it to never say "you're absolutely right" and always challenge, then it will dutifully obey, and always challenge - even when you are, in fact, right. What you really want is "challenge me when I'm wrong, and tell me I'm right if I am" - which seems to be a lot harder. As another example, one common…

It's a really hard problem to solve! You might think you can train the AI to do it in the usual fashion, by training on examples of the AI calling out errors, and agreeing with facts, and if you do that—and if the AI gets smart enough—then that should work. If. You. Do. That. Which you can't, because humans also make mistakes. Inevitably, there will be facts in the 'falsehood' set—and vice versa. Accordingly, the AI…

The AI needs to be able to lookup data and facts and weigh them properly. Which is not easy for humans either; once you're indoctrinated in something, and you trust a bad data source over another, it's evidently very hard to correct course.

Re: Claude says “You're absolutely right!” about everything

#245

Post by ryandrake »

throwawayffffas wrote: post #187
dkarl wrote: post #183

Earlier quoted context omitted.

> it always mentions something about how it’s going to be ‘straight to the point’ or give me a ‘no bs version’ That's how you suck up to somebody who doesn't want to see themselves as somebody you can suck up to. How does an LLM know how to be sycophantic to somebody who doesn't (think they) like sycophants? Whether it's a naturally emergent phenomenon in LLMs or specifically a result of its corporate environment, I'…

It doesn't know. It was trained and probably instructed by the system to be positive and reassuring.

They actually feel like they were trained to be both extremely humble and at the same time, excited to serve. As if it were an intern talking to his employer's CEO. I suspect AI companies executive leadership, through their feedback to their devs about Claude, ChatGPT, Gemini, and so on, are unconsciously shaping the tone and manner of their LLM product's speech. They are used to be talked to like this, so their products should talk to users like this! They are used to having yes-man sycophants in their orbit, so they file bugs and feedback until the LLM products are also yes-man sycophants.

I would rather have an AI assistant that spoke to me like a similarly-leveled colleague, but none of them seem to be turning out quite like that.

Re: Claude says “You're absolutely right!” about everything

#246

Post by pjc50 »

nojs wrote: post #117

I'm starting to think this is a deeper problem with LLMs that will be hard to solve with stylistic changes. If you ask it to never say "you're absolutely right" and always challenge, then it will dutifully obey, and always challenge - even when you are, in fact, right. What you really want is "challenge me when I'm wrong, and tell me I'm right if I am" - which seems to be a lot harder. As another example, one common…

Well, yes, this is a hard philosophical problem, finding out Truth, and LLMs just side step it entirely, going instead for "looks good to me".

Re: Claude says “You're absolutely right!” about everything

#247

Post by apt-apt-apt-apt »

baggachipz wrote: post #82

I'm pretty sure they want it kissing people's asses because it makes users feel good and therefore more likely to use the LLM more. Versus, if it just gave a curt and unfriendly answer, most people (esp. Americans) wouldn't like to use it as much. Just a hypothesis.

Better than GPT5. Which talks like this. Parameters fulfilled. Request met.

Re: Claude says “You're absolutely right!” about everything

#248

Post by bargainbin »

ryao wrote: post #207
elif wrote: post #150

I've spent a lot of time trying to get LLM to generate things in a specific way, the biggest take away I have is, if you tell it "don't do xyz" it will always have in the back of its mind "do xyz" and any chance it gets it will take to "do xyz" When working on art projects, my trick is to specifically give all feedback constructively, carefully avoiding framing things in terms of the inverse or parts to remove.

LLMs love to do malicious compliance. If I tell them to not do X, they will then go into a “Look, I followed instructions” moment by talking about how they avoided X. If I add additional instructions saying “do not talk about how you did not do X since merely discussing it is contrary to the goal of avoiding it entirely”, they become somewhat better, but the process of writing such long prompts merely to say not to d…

Just got stung with this on GPT5 - It’s new prompt personalisation had “Robotic” and “no sugar coating” presets.

Worked great until about 4 chats in I asked it for some data and it felt the need to say “Straight Answer. No Sugar coating needed.”

Why can’t these things just shut up recently? If I need to talk to unreliable idiots my Teams chat is just a click away.

Post reply on HN