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

311–320 of 560 posts

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

#311

Post by yieldcrv »

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.

It’s a disgusting aspect of these revenue burning investment seeking companies noticing that sycophancy works for user engagement

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

#312

Post by recursive »

apt-apt-apt-apt wrote: post #247
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.

That looks perfect.

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

#313

Post by grogenaut »

lonelyasacloud wrote: post #260
jonplackett wrote: post #160

Earlier quoted context omitted.

I have this same problem. I’ve added a bunch of instructuons to try and stop ChatGPT being so sycophantic, and now it always mentions something about how it’s going to be ‘straight to the point’ or give me a ‘no bs version’. So now I just have that as the intro instead of ‘that’s a sharp observation’

Default is output_default = raw_model + be_kiss_a_system When that gets changed by the user to output_user = raw_model + be_kiss_a_system - be_abrupt_user Unless be_abrupt_user happens to be identical to be_kiss_a_system _and_ is applied with identical weight then it's seems likely that it's always going to add more noise to the output.

Also be abrupt is in the user context and will get aged out. The other stuff is in training or in software prompt and wont

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

#315

Post by smoe »

radarsat1 wrote: post #5

I find Gemini is also hilariously enthusiastic about telling you how amazingly insightful you are being, almost no matter what you say. Doesn't bother me much, I basically just ignore the first paragraph of any reply, but it's kind of funny.

I agree that Gemini is overly enthusiastic, but at least in my limited testing, 2.5 Pro was also the only model that sometimes does say “no.”

Recently I tested both Claude and Gemini by discussing data modeling questions with them. After a couple of iterations, I asked each model whether a certain hack/workaround would be possible to make some things easier.

Claude’s response: “This is a great idea!”, followed by instructions on how to do it.

Gemini’s response: “While technically possible, you should never do this”, along with several paragraphs explaining why it’s a bad idea.

In that case, the “truth” was probably somewhere in the middle, neither a great idea nor the end of the world.

But in the end, both models are so easily biased by subtle changes in wording or by what they encounter during web searches among other things, that one definitely can’t rely on them to push back on anything that isn’t completely black and white.

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

#316

Post by smeej »

I think the developers want these AI tools to be likable a heck of a lot more than they want them to be useful--and as a marketing strategy, that's exactly the right approach.

Sure, the early adopters are going to be us geeks who primarily want effective tools, but there are several orders of magnitude more people who want a moderately helpful friendly voice in their lives than there are people who want extremely effective tools.

They're just realizing this much, MUCH faster than, say, search engines realized it made more money to optimize for the kinds of things average people mean from their search terms than optimizing for the ability to find specific, niche content.

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

#317

Post by cruffle_duffle »

incone123 wrote: post #239
corytheboyd wrote: post #205

Earlier quoted context omitted.

As part of the AI insanity $employer forced us all to do an “AI training.” Whatever, wasn’t that bad, and some people probably needed the basics, but one of the points was exactly this— “use negative prompts: tell it what not to do.” Which is exactly an approach I had observed blow up a few times already for this exact reason. Just more anecdata suggesting that nobody really knows the “correct” workflow(s) yet, in th…

That your firm purchased training that was clearly just some chancers doing whatever seems like an even worse approach than just giving out access to a service and telling everyone to give it a shot. Do they also post vacancies asking for 5 years experience in a 2 year old technology?

To be fair this shit is so new and constantly changing that I don’t think anybody truly understands what is going on.

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

#319

Post by cbracketdash »

Here are my instructions to Claude.

"Get straight to the point. Ruthlessly correct my wrong assumptions. Do not give me any noise. Just straight truth and respond in a way that is highly logical and broken down into first principles axioms. Use LaTeX for all equations. Provide clear plans that map the axioms to actionable items"

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

#320

Post by marcosdumay »

justusthane wrote: post #264
tho24i234234 wrote: post #144

Earlier quoted context omitted.

It most definitely is a American thing - this is why non-native speakers often come out as rude or unfriendly or plain stupid. We don't appreciate how much there is to language.

> We don't appreciate how much there is to language. This can’t possibly be true, can it? Every language must have its own nuance. non native English speakers might not grasp the nuance of English language, but the same could be said for any one speaking another language.

Language barriers are cultural barriers.

It's as simple as that. Most people do not expect to interact the way that most native English speakers expect.

Post reply on HN