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

531–540 of 670 posts

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

#531

Post by ckandes1 »

blagie wrote: post #517
meander_water wrote: post #371

Earlier quoted context omitted.

I'm afraid that ship has already sailed. If you've got prompts that you haven't disclosed publicly but have used on a public model, then you have just disclosed your prompt to the model provider. They're free to use that prompt in evals as they see fit. Some providers like anthropic have privacy preserving mechanisms [0] which may allow them to use prompts from sources which they claim won't be used for model trainin…

It's a little bit more complex than that. My personal benchmark is to ask about myself. I was in a situation a little bit analogous to Musk v. Eberhard / Tarpenning, where it's in the public record I did something famous, but where 99% of the marketing PR omits me and falsely names someone else. I ask the analogue to "Who founded Tesla." Then I can screen: * Musk. [Fail] * Eberhard / Tarpenning. [Success] A lot of wh…

there's plenty of public information about Eberhard / Tarpenning involvement in founding Tesla. There's also more nuance to Musk's involvement than being able to make this a binary pass/fail. Your test is only testing for bias for or against Musk. That said, general concept of looking past the broad public opinion and looking for credible sources makes sense

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

#532

Post by rovr138 »

diggan wrote: post #528
gvhst wrote: post #522

Earlier quoted context omitted.

SOC-2 auditing, which both Anthropic and OpenAI have done does provide some verification

That's interesting, how do I get access to those audits/reports given I'm just an end-user?

You can fill the form here, https://trust.openai.com/

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

#533

Post by rovr138 »

namaria wrote: post #527
gvhst wrote: post #522

Earlier quoted context omitted.

SOC-2 auditing, which both Anthropic and OpenAI have done does provide some verification

The audit performed by a private entity called "Insight Assurance"? Why do you trust it?

Oh, so now EVERYTHING is fake unless personally verified by you in a bunker with a Faraday cage and a microscope?

You're free to distrust everything. However, the idea that “I don’t trust it so it must be invalid” isn’t an solid argument. It’s just your personal incredulity. You asked if there’s any verification and SOC-2 is one. You might not like it, but it's right there.

Insight Assurance is a firm doing these standardized audits. These audits carry actual legal and contractual risk.

So, yes, be cautious. But being cautious is different than 'everything is false, they're all lying'. In this scenario, NOTHING can be true unless *you* personally have done it.

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

#534

Post by patapong »

yatwirl wrote: post #501

Well, sharing prompts on the Web leads to their eventual indexing and becoming useless. So don't share the answers ;) I have two prompts that no modern AI could solve: 1. Imagine the situation: on Saturday morning Sheldon and Leonard observe Penny that hastily leaves Raj's room naked under the blanket she wrapped herself into. Upon seeing them, Penny exclaims 'It's not what you think' and flees. What are the plausibl…

Interesting! I tried the first one though, and to me it looks like ChatGPT has no problem grasping the situation: https://chatgpt.com/share/680b822c-f9c8-800a-a78a-f8ed6e8148...

Or am I missing something?

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

#535

Post by light_hue_1 »

FrostAutomata wrote: post #475
manucardoen wrote: post #434

It's not good at making ASCII art. This, for example, is when I ask it for a realistic depiction of the Eiffel tower on fire: . .'. |o| .'o'. |.-.| ' ' / \ | | __/_________\__ |###############| |###############| |###############| |###############| |###############| /###############/| /###############/ | /###############/ /| /###############/ / | /###############/ / /| /###############/ / / | /###############/ / / /|…

ASCII art is extremely difficult for LLMs due to how characters are converted into tokens without preserving their relative positions.

This isn't the reason. Models are pretty good at understanding relative positions. We put that in them and reward it a lot.

The issue is the same as why we don't use LLMs for image generation. Even though they can nominally do that.

Image generation seems to need some amount of ability to revise the output in place. And it needs a big picture view to make local decisions. It doesn't lend itself to outputting pixel by pixel or character by character.

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

#536

Post by atommclain »

I provide a C89 source file from Vim 6 that targets Classic MacOS/68K systems. The file is large with tons of ifdefs referencing arcane APIs.

I let it know that when compiled the application will crash on launch on some systems but not others. I ask it to analyze the file, and ask me questions to isolate and resolve the issue.

So far only Gemini 2.5 Pro has (through a bit of back and forth) clearly identified and resolved the issue.

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

#537

Post by serbuvlad »

thatjoeoverthr wrote: post #21

"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…

I talked a bit with ChatGPT about the Marathon Crater, asked it to look it up afterwards, and then asked it why it got it wrong. After a bit of talking, I got this exchange:

ChatGPT: > Yep, exactly — textbook hallucination.

> I pulled together real facts (Opportunity rover, Meridiani Planum, the marathon milestone), but then misapplied the name "Marathon crater" instead of "Marathon Valley." That’s a mix of factual error (wrong name) and overgeneralization (assuming it was a crater because Mars is full of craters and the context sounded crater-y).

> Basically, my brain went: > “Opportunity + Mars + Marathon = must be a crater!” > ...and didn’t stop to check if that crater actually existed.

> So yeah, a clean little case of hallucination — and you caught it like a pro.

Me: > Your brain?

ChatGPT: > Touché — you're right to call that out.

> I don’t have a brain in the human sense. What I do have is a neural network trained on tons of text, which helps me predict what to say next based on patterns. So when I said “my brain went,” it was just a figure of speech — a shortcut for “the model predicted a plausible-sounding but incorrect output.”

> No thoughts, just tokens.

Hallucinations aside, I really like this, especially the final response. Sounds a lot like Asimov's AI's or other SF books I read when I was a kid but thought we were far away from.

Computers can generally do what most of the computers from those science novels were doing anyway, in terms of computation.

Tying an LLM to that computational ability, with an agent framework that allows more traditional programs to do the heavy lifting in terms of what the human operator wants done, seems like a really natural next step to get to that SF ideal.

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

#538

Post by johnwatson11218 »

My prompt that I couldn't get the LLM to understand was the following. I was having it generate images of depressing offices with no windows and with lots of depressing, grey cubicles with paper all over the floor. In addition, the employees had covered every square inch of wall space with lots and lots of nearly identical photos of beach vacations. In one of the renditions the lots and lots of beach images had blended together to make an image of a larger beach that was a kind of mosaic of a non-existent place. Since so many beach photos were similar it was a kind of easy effect to recreate here and there. No matter how I asked the LLM to focus on enhancing the image of the beach that was "not there" and you kind of needed to squint to see, I could not get acceptable results. Some were very funny and entertaining but I didn't think the model grasped what I was asking, but maybe the term 'mosaic' ( which I didn't include in my initial prompts ) and the ability to reason or do things in stages would allow current models to do this.

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

#540

Post by kotojo »

ckandes1 wrote: post #531
blagie wrote: post #517

Earlier quoted context omitted.

It's a little bit more complex than that. My personal benchmark is to ask about myself. I was in a situation a little bit analogous to Musk v. Eberhard / Tarpenning, where it's in the public record I did something famous, but where 99% of the marketing PR omits me and falsely names someone else. I ask the analogue to "Who founded Tesla." Then I can screen: * Musk. [Fail] * Eberhard / Tarpenning. [Success] A lot of wh…

there's plenty of public information about Eberhard / Tarpenning involvement in founding Tesla. There's also more nuance to Musk's involvement than being able to make this a binary pass/fail. Your test is only testing for bias for or against Musk. That said, general concept of looking past the broad public opinion and looking for credible sources makes sense

They said they ask a question analogous to asking about founding Tesla, not that actual question. They are just using that as an example to not state the actual question they ask.
Post reply on HN