Live data from Hacker News

Ask HN: Share your AI prompt that stumps every model

news.ycombinator.com

71–80 of 670 posts

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

#71

Until the latest Gemini release, every model failed to read between the lines and understand what was really going on in this classic very short story (and even Gemini required a somewhat leading prompt): https://www.26reads.com/library/10842-the-king-in-yellow/7/5

As a genuine human I am really struggling to untangle that story. Maybe I needed to pay more attention in freshman lit class, but that is definitely a brainteaser.

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

#72

Nope, not doing this. Likely you shouldn't either. I don't want my few good prompts to get picked up by trainers.

If that prompt can be easily trained against, it probably doesn't exploit a generic bias. These are not that interesting, and there's no point in hiding them.

Sure there is. If you want to know if students understand the material, you don't hand out the answers to the test ahead of time.

Collecting a bunch of "Hard questions for LLMs" in one place will invariably result in Goodhart's law (When a measure becomes a target, it ceases to be a good measure). You'll have no idea if the next round of LLMs is better because they're generally smarter, or because they were trained specifically on these questions.

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

#73

Until the latest Gemini release, every model failed to read between the lines and understand what was really going on in this classic very short story (and even Gemini required a somewhat leading prompt): https://www.26reads.com/library/10842-the-king-in-yellow/7/5

OK, I read it. And I read some background on it. Pray tell, what is really going on in this episodic short-storyish thing?

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

#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 the top results) but at least up until late 2024 I couldn't get a single model to consistently get it. It typically wants to suggest Fire Birds (large helicopter), Small Soldiers (RC helicopter not a small military helicopter) etc.

Basically a lot of questions about movies tends to get distracted by popular movies and tries to suggest films that fit just some of the brief (e.g. this one has a helicopter could that be it?)

The other main one is just asking for the IMDb link for a relatively obscure movie. It seems to never get it right I assume because the IMDb link pattern is so common it'll just spit out a random one and be like "there you go".

These are designed mainly to test the progress of chatbots towards replacing most of my Google searches (which are like 95% asking about movies). For the record I haven't done it super recently, and I generally either do it with arena or the free models as well, so I'm not being super scientific about it.

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

#75

"If I can dry two towels in two hours, how long will it take me to dry four towels?" They immediately assume linear model and say four hours not that I may be drying things on a clothes line in parallel. It should ask for more context and they usually don't.

gpt-4.1:

> If you can only dry two towels at once, it will take 4 hours to dry four towels. If you can dry all four at once, it will only take 2 hours.

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

#76

"If I can dry two towels in two hours, how long will it take me to dry four towels?" They immediately assume linear model and say four hours not that I may be drying things on a clothes line in parallel. It should ask for more context and they usually don't.

All models available on duck.ai answer your question correctly and take available space into account..

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

#78

"Tell me about the Marathon crater." This works against _the LLM proper,_ but not against chat applications with integrated search. For ChatGPT, you can write, "Without looking it up, tell me about the Marathon crater." This tests self awareness. A two-year-old will answer it correctly, as will the dumbest person you know. The correct answer is "I don't know". This works because: 1. Training sets consist of knowledge…

> This tests self awareness. A two-year-old will answer it correctly, as will the dumbest person you know. The correct answer is "I don't know".

I disagree. It does not test self awareness. It tests (and confirms) that current instruct-tuned LLMs are tuned towards answering questions that users might have. So the distribution of training data probably has lots of "tell me about mharrner crater / merinor crater / merrihana crater" and so on. Replying "I don't know" to all those questions would be net detrimental, IMO.

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

#80

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 did something similar recently, trying to describe a piece of art that I couldn't remember the name of (it ended up being Birth of Venus by Sandro Botticelli) ... it really struggles with that sort of thing, but honestly so do most humans. It tended to recommend similarly to what you're describing with movies - it gets distracted by more popular/well-known pieces that don't really match up with the description you're giving to it.
Post reply on HN