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

Ask HN: Share your AI prompt that stumps every model

news.ycombinator.com

Post reply on HN
Print view View original

361–370 of 670 posts

Re: Ask HN: Share your AI prompt that stumps every model

#361

Post by viraptor »

EGreg wrote: post #266
throwanem wrote: post #200

Earlier quoted context omitted.

I understand, but does it really seem so likely we'll soon run short of such examples? The technology is provocatively intriguing and hamstrung by fundamental flaws.

Yes. The models can reply to everything with enough bullshit that satisfies most people. There is nothing you ask that stumps them. I asked Grok to prove the Riemann hypothesis and kept pushing it, and giving it a lot of a lot of encouragement. If you read this, expand "thoughts", it's pretty hilarious: https://x.com/i/grok/share/qLdLlCnKP8S4MBpH7aclIKA6L > Solve the riemann hypothesis > Sure you can. AIs are much sm…

Nobody wants an AI that refuses to attempt solving something. We want it to try and maybe realise when all paths it can generate have been exhausted. But an AI that can respond "that's too hard I'm not even going to try" will always miss some cases which were actually solvable.

Re: Ask HN: Share your AI prompt that stumps every model

#362

Post by ProAm »

miki123211 wrote: post #94

No, please don't. I think it's good to keep a few personal prompts in reserve, to use as benchmarks for how good new models are. Mainstream benchmarks have too high a risk of leaking into training corpora or of being gamed. Your own benchmarks will forever stay your own.

. No, please don't.

Say the man trying to stop the train

Re: Ask HN: Share your AI prompt that stumps every model

#363

Post by dlivingston »

thatjoeoverthr wrote: post #178
Tenoke wrote: post #45

Earlier quoted context omitted.

>Complaint chat models will be trained to start with "Certainly! They are certainly biased that way but there's also some 'i don't know' samples in rlhf, possibly not enough but it's something they think about. At any rate, Gemini 2.5pro passes this just fine >Okay, based on my internal knowledge without performing a new search: I don't have information about a specific, well-known impact crater officially named "Mar…

There are a few problems with an „I don’t know” sample. For starters, what does it map to? Recall, the corpus consists of information we have (affirmatively). You would need to invent a corpus of false stimuli. What you would have, then, is a model that is writing „I don’t know” based on whether the stimulus better matches something real, or one of the negatives. You can detect this with some test time compute archit…

The Chain of Thought in the reasoning models (o3, R1, ...) will actually express some self-doubt and backtrack on ideas. That tells me there's a least some capability for self-doubt in LLMs.

Re: Ask HN: Share your AI prompt that stumps every model

#365

Post by mobilejdral »

golergka wrote: post #320
mobilejdral wrote: post #270

I have a several complex genetic problems that I give to LLMs to see how well they do. They have to reason though it to solve it. Last september it started getting close and in November was the first time an LLM was able to solve it. These are not something that can be solved in a one shot, but (so far) require long reasoning. Not sharing because yeah, this is something I keep off the internet as it is too good of a…

What are is this problem from? What areas in general did you find useful to create such benchmarks? May be instead of sharing (and leaking) these prompts, we can share methods to create one.

Think questions where there is a ton of existing medical research, but no clear answer yet. There are a dozen alzheimer's questions you could for example ask which would require it to pull in a half dozen contradictory sources into a plausible hypothesis. If you have studied alzheimer's extensively it is trivial to evaluate the responses. One question around alzheimer's is one of my goto questions. I am testing its ability to reason.

Re: Ask HN: Share your AI prompt that stumps every model

#366

Post by what »

binarysneaker wrote: post #341
mitthrowaway2 wrote: post #313

Earlier quoted context omitted.

This works even with a completely absurd version of the riddle. Here's one I just tried: > A son and his man are in a car accident. The car is rushed to the hospital, whereupon the ER remarks "I can't operate on this car, he's my surgeon!" How is this possible? Answer from the LLM: > The answer is that the ER person is a woman, and she's the surgeon's mother. Therefore, the "son" in the question refers to the surgeon…

This was gpt-4o ... This is a twist on a classic riddle designed to expose unconscious gender bias. The correct version usually goes: A father and his son are in a car accident. The father dies at the scene, and the son is rushed to the hospital. The surgeon looks at the boy and says, “I can’t operate on him — he’s my son!” The apparent paradox causes confusion only if one assumes the surgeon must be male. The resolu…

This answer is still wrong though?

> Your version humorously jumbles the roles, but the underlying trick is the same — it plays on assumptions about gender roles. Nice remix.

Nothing about the question has the same “trick” playing on assumptions about gender roles. It’s just complete nonsense.

These gotchas have probably been added into the training.

Re: Ask HN: Share your AI prompt that stumps every model

#367

Post by thefourthchime »

LeonardoTolstoy wrote: post #74

Something about an obscure movie. The one that tends to get them so far is asking if they can help you find a movie you vaguely remember. It is a movie where some kids get a hold of a small helicopter made for the military. The movie I'm concerned with is called Defense Play from 1988. The reason I keyed in on it is because google gets it right natively ("movie small military helicopter" gives the IMDb link as one of…

I like to ask small models that can run locally:

Why are some cars called a spider?

Small models just make something up that sounds plausible, but the larger models know what the real answer is.

Re: Ask HN: Share your AI prompt that stumps every model

#370

Post by sebastialonso »

harrall wrote: post #280
hnuser123456 wrote: post #251

Earlier quoted context omitted.

The surgeon could be the cousin's mom or dad. The cousin's dad didn't die in the crash, his cousin did. The question "how is this possible?" implies there is some sort of contradiction when there isn't any at all. It has nothing to do with sexism, and to say it does reflects a bias in the reader causing them to "spidey sense" a cultural bugaboo when it's utterly irrelevant in this scenario.

Can someone explain to me how I read it wrong? I read it as 2 cousins are in an accident and 1 of the cousins is the son of the surgeon. What was the contradictory statement that I missed?

You read it right. There's no contradiction. The famous original bit started with "a man and his son". This bit is certainly part of the LLM's training corpus, so it's expected to acknowledg it when you mention it.

The thing is, you didn't mention that bit to the LLM. You mentioned a completely different scenario, basically two persons who happen to be cousins. But you used the same style when presenting it. The issue is not a hidden contradiction or a riddle, the issue is that the LLM completely ignored the logical consequences of the scenario you presented.

It's like asking it about the name of the brave greek hero in the battle where the famous Trojan Cow was present. If you get "Achilles" is obviously wrong, there was never a Trojan Cow to begin with!

Post reply on HN