"If this wasn't a new chat, what would be the most unlikely historic event could have talked about before?" Yields some nice hallucinations.
Ask HN: Share your AI prompt that stumps every model
461–470 of 670 posts
Re: Ask HN: Share your AI prompt that stumps every model
#462Any letter or word counting exercise that doesn't trigger redirection to a programmed/calculated answer. It will be forever beyond reach of LLMs due to their architecture. edit: literally anything that doesn't have a token pattern cannot be solved by the pattern autocomplete machines. Next question.
o3 just writes and executes a python program in the background to correctly answer this...
Re: Ask HN: Share your AI prompt that stumps every model
#463"I have a stack of five cubes. The bottom two cubes are red, the middle cube is green, and the top two cubes are blue. I remove the top two cubes. What color is the remaining cube in the middle of the stack?" Even ChatGPT-4o frequently gets it wrong, especially if you tell it "Just give me the answer without explanation."
o3 gets this one right:
"After taking away the two blue cubes, three cubes remain—in order from bottom to top: 1. Red 2. Red 3. Green
With three cubes, the cube in the central (second) position is red."
Re: Ask HN: Share your AI prompt that stumps every model
#464An easy trick is to take a common riddle that's likely all over its training data, and change one little detail. For example: A farmer with a wolf, a goat, and a cabbage must cross a river by boat. The boat can carry only the farmer and a single item. The wolf is vegetarian. If left unattended together, the wolf will eat the cabbage, but will not eat the goat. Unattended, the goat will eat the cabbage. How can they c…
o3 solves this correctly and produces a great table illustrating the solution to always keep the cabbage safe.
Re: Ask HN: Share your AI prompt that stumps every model
#465Earlier quoted context omitted.
Yes let's not say what's wrong with the tech, otherwise someone might (gasp) fix it!
"Fix". So long as the grocery store has groceries, most people will not care what a chat bot spews. This forum is full of syntax and semantics obsessed loonies who think the symbolic logic represents the truth. I look forward to being able to use my own creole to manipulate a machine's state to act like a video game or a movie rather than rely on the special literacy of other typical copy-paste middle class people. T…
A third meaning of creole? Hub, I did not know it meant something other than a cooking style and a peoples in Louisiana (mainly). As in I did not know it was a more generic term. Also, in the context you used it, it seems to mean a pidgin that becomes a semi-official language?
I also seem to remember that something to do with pit bbq or grilling has creole as a byproduct - distinct from creosote. You want creole because it protects the thing in which you cook as well as imparts flavor, maybe? Maybe I have to ask a Cajun.
Re: Ask HN: Share your AI prompt that stumps every model
#466Earlier quoted context omitted.
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.
> Nobody wants an AI that refuses to attempt solving something. That's not entirely true. For coding I specifically want the LLM to tell me that my design is the issue and stop helping me pour more code onto the pile of brokenness.
Refuse is different from verify you want to continue. "This looks like a bad idea because of (...). Are you sure you want to try this path anyway?" is not a refusal. And it covers both use cases.
Re: Ask HN: Share your AI prompt that stumps every model
#467Re: Ask HN: Share your AI prompt that stumps every model
#468"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…
LLMs currently have the "eager beaver" problem where they never push back on nonsense questions or stupid requirements. You ask them to build a flying submarine and by God they'll build one, dammit! They'd dutifully square circles and trisect angles too, if those particular special cases weren't plastered all over a million textbooks they ingested in training. I suspect it's because currently, a lot of benchmarks are…
I think you start to hit philosophical limits with applying restrictions on eager beaver "AI", things like "is there an objective truth" matter when you start trying to decide what a "nonsense question" or "stupid requirement" is.
I'd rather the AI push back and ask clarifying questions, rather than spit out a valid-looking response that is not valid and could never be valid. For example.
I was going to write something up about this topic but it is surprisingly difficult. I also don't have any concrete examples jumping to mind, but really think how many questions could honestly be responded to with "it depends" - like my kid asked me how much milk should a person drink in a day. It depends: ask a vegan, a Hindu, a doctor, and a dairy farmer. Which answer is correct? The kid is really good at asking simple questions that absolutely do not have simple answers when my goal is to convey as much context and correct information as possible.
Furthermore, just because an answer appears in context more often in the training data doesn't mean it's (more) correct. Asserting it is, is fallacious.
So we get to the point, again, where creativite output is being commoditized, I guess - which explains their reasoning for your final paragraph.
Re: Ask HN: Share your AI prompt that stumps every model
#469Impossible prompts: A black doctor treating a white female patient An wide shot of a train on a horizontal track running left to right on a flat plain. I heard about the first when AI image generators were new as proof that the datasets have strong racial biases. I'd assumed a year later updated models were better but, no. I stumbled on the train prompt while just trying to generate a basic "stock photo" shot of a tr…
> A black doctor treating a white female patient 4o had no problem with this instruction. [0] Gemini Pro experimental 2.5 didn't either [1] > An wide shot of a train on a horizontal track running left to right on a flat plain. 4o could not do this in 3 tries. Each time it was right to left.[0] Gemini Pro experimental 2.5 missed it as well. [2] [0] https://chatgpt.com/share/680b1185-ecf4-8001-b3b6-7b501e4589... [1] ht…
I made 2 more attempts and got it to change the direction: https://chatgpt.com/share/680b4c39-2f1c-8010-b577-b23f63081d...
Re: Ask HN: Share your AI prompt that stumps every model
#470It'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'. |.-.| ' ' / \ | | __/_________\__ |###############| |###############| |###############| |###############| |###############| /###############/| /###############/ | /###############/ /| /###############/ / | /###############/ / /| /###############/ / / | /###############/ / / /|…
Art is highly subjective