I got a couple of good ones:
When I first opened it, deepseek also had a correct clock, but I accidentally refreshed the page when I scrolled up to double check its time, second time through only these two were right
Ed. Names are below the clock, the top one was by the clock drawing champion Kimi K2
This is my favorite
gaslight clock
So the prompt is intentionally breaking things…?
Edit: I guess this was somehow the chatty output doing who know wtf
Is it just me or the clocks frequently break or change appearance without the page being refreshed?
Edit: nevermind, I skipped past the sentence explaining that every minute, the site prompts LLMs for a new solution. This is hilariously sad how LLMs aren’t able to be consistent from one prompt to another.
It’s the expected result if your big ol’ artificial intelligence wannabe is ultimately just a stochastic word combinator.
if every single token is, at the end, chosen by random dice roll (and they are) then this is exactly what you’d expect.
that’s a massive oversimplification
not really. If the system outputs a probability distribution, then by definition, you’re picking somewhat randomly. So not really a simplification
This is hilariously sad how LLMs aren’t able to be consistent from one prompt to another.
Typically that’s configurable. Like for a chatbot, you’d want it to give the same/similar results for a given question, where with a character creator, you might want the results to vary so you can re-run until you get something you like.
Of course that wouldn’t be as funny here.
You can click on the button in the top right corner (with a question mark) to have explanations. The clocks are refreshed every minute
Removed by mod
that’s scary how dementia works :'(
This kinda freaked me out: AI models fed their own outputs as training data will quickly start making distorted images that look spookily like human painting made under the progression of mental illness or drugs.
well that was terrifying
Thanks for sharing this! I really think that when people see LLM failures and say that such failures demonstrate how fundamentally different LLMs are from human cognition, they tend to overlook how humans actually do exhibit remarkably similar failures modes. Obviously dementia isn’t really analogous to generating text while lacking the ability to “see” a rendering based on that text. But it’s still pretty interesting that whatever feedback loops did get corrupted in these patients led to such a variety of failure modes.
As an example of what I’m talking about, I appreciated and generally agreed with this recent Octomind post, but I disagree with the list of problems that “wouldn’t trip up a human dev”; these are all things I’ve seen real humans do, or could imagine a human doing.
such a variety of failure modes
What i find interesting is that in both cases there is a certain consistency in the mistakes too - basically every dementia patient still understands the clock is something with a circle and numbers and not a square with letters for example. LLMs can tell you cokplete bullshit, but still understands it has to be done with perfect grammar in a consistant language. So much so it struggles to respond outside of this box - ask it to insert spelling errors to look human for example.
the ability to “see”
This might be the true problem in both cases, both the patient and the model can not comprehend the bigger picture (a circle is divided into 12 segments, because that is how we deconstructed the time it takes for the earth to spin around it’s axis). Things that seem logical to use, are logical because of these kind of connections with other things we know and comprehend.
… what
Basically this: https://www.psychdb.com/cognitive-testing/clock-drawing-test
I guess that test is going to become less useful as generation Z age
Thanks.
What I still didn’t figure out about the comment I replied to is:
- What is each row? They’re labeled I, II, III, IV. What’s being counted?
- Why did they link to a home interior design website under “via”?
Good questions. I don’t know, and I can no longer try to find out, as the mods have now removed the comment. (Sorry for the double-post–I got briefly confused about which comment you were referring to and deleted my first post, then realized I’d been frazzled and the post in question really was deleted by the mods.)
deleted by creator
Get educated
Thanks for sharing that mindfuck. I honestly would’ve thought something was wrong with my cognition if you hadn’t mentioned it was a test beforehand.
Some of these are absolutely hilarious
The last one, Kimi K2, has been consistently good as long as I’ve been looking at it. That’s pretty impressive.
The rest are hilarious!
Deepseek has a recognisable clock now and then, too. They both mix up the current time, though.
By far the best, but still off. These three were loaded in the same order as i post them:
I dig the square clock, and am now sad that the numbers can’t be put into the corners on a real clock. Unless they’re shifted from the usual position.
Cartier found a work around quite some time ago and maybe they weren’t even the first to design a square ‘clock’:
(The roman numerals are nice, but notice the ‘circle’ between the numerals and the hands, almost like the circle from the ai)
That one is pretty good, though the Roman numerals are rather busy and uneven.
This one is closer, though now I have to wonder if all non-square rectangular clocks have an old-timey whiff for me, or it’s just the border here:
This is also impressive:
I like the second one a lot, especially how the upper and bottom numerals face the floor and the left and right ones face towards the center, and to allow for that there has to be a sudden flip from 3>4 and 8>9. But the indices are not playing by the square-clock rule and unlike the cartier one form a regular oval shape.
I like how the upper one had to find a way to make clear which indice represents the numerals - it really shows the problem in projecting the circular movement of the hands into a rectangular (thanks, that’s the right word) shape.
It think most analog clocks/watches will give you an old-timey whiff much more often than not, just because there is a more new-timey alternative. I went looking for some watch faces for smart watches, but couldn’t really find any interesting one. Most are either digital numbers or a round clock on a rectangular display.
Neither of those interest me like the Cartier tank, which I find really ugly watches to be honest. It’s just this double outlined rectangle(-ish shape) which is unevenly split into 60 boxes that I like (seen below on the first, third and fifth watch).
Haha, I found myself thinking the same thing, and then caught myself, realizing all the other LLMs on this page had lowered the bar immensely for what I’m considering impressive.
I thought the same and then Kimi K2 came up with a clock that has two 12 and no 11…
It’s funny how GPT-5 is consistently the worst one, and it’s not even close.
qwen 2.5 is absolutely pants on head ridiculous compared to gpt5 when I’m looking at it right now.
I don’t even
I was surprised that both Grok and Gemini 2.5 got it right once, only to fuck it up on the refresh
Given that the AI models are basically constructing these “blindly”- using the language model to string together html and javascript without really being able to check how it looks- some of these are actually pretty impressive. But also making the AI do things it’s bad at is funny. Reminds me of all the AI ASCII art fails…
deleted by creator
Another reason why, while AI might be a fun toy, no one who is serious about getting work done will touch it with a dirty barge pole. The gratuitous hallucinations alone ought to be a sufficient deterrent.
qwen is trying her best 😭😔
So far, I’d give qwen the prize for most artistic impression of a clock.
Kimi K2 appears to consistently get it right.
And just as I typed that, Kimi made one where 9 and 10, and 11 and 12 overlapped.
What is this obsession with clocks recently?
I don’t know if it’s actually related, but I’ve read that asking people to draw a clock face is a simple way to identify some brain problems
Quick screening for dementia, according to this
Edit: I guess this means most of the AI has ‘Conceptual Deficits’, pretty accurate lol
Would be funny if AI models are generating such wildly useless “clocks” because they ingested too many dementia screening tests in their training data
There is someone training the biggest, bestest model to draw clock faces to pass that test as we speak.
I’m guessing it’s an easy metric to compare benchmarks. “Write a clock”.
Not really world clocks, they just try to use JavaScript to display the device’s time.
No JavaScript I think, it’s just html and CSS. The initial time is provided in the prompt every minute according to the description. I wonder if they’d be any better if they could use js. Probably not.
“Time is a relative of mine” / Eisenstein
Well, KIMI K2 seems to have created the working one. Others failed. I suppose that this model was optimized for this while others not.
The clocks change every minute. I’ve seen some from deepseek and qwen that looked ok. But kimi seems to be the most consistent
You know, I don’t, and what the fuck?
