How many legs do ten elephants have, if two of them are legless?
How many legs do ten elephants have, if two of them are legless?
There is a discussion on Hacker News, but feel free to comment here as well.
Bing got it right. I'm kind of surprised.
Citing WebMD as a source that elephants have 4 legs is kinda hilarious ngl
Last time I tried bard it told me about a time in the early 2000s when computers had spoon slots to store your spoon
Where else would you put your spoon?
In the coffee cup in the coffee cup holder obviously
I grew up in the era of analogue spoon slots for storing all spoon like utensils. It was a simpler time.
Beside your poop knife, obviously.
LAN events often had food and a good host would serve food that doesn't cause greasy fingers. It was common in the early 90s to have chili while using the Internet, back when it was called 'the world wide web' by very serious news anchors. (There was even a Futurama joke about chili on the screen after Scruffy was eating while browsing porn!)
Spoon slots fit in the 5 1/4 slot and had a manual ejecting tray. It was exciting to see cup holders get automatic doors and I was waiting for spoon slots to get some sweet upgrades but they faded into obscurity before the internet took off and no one seems to remember. Or worse, the say that it never happened. Like I would forget my favorite part of PC gaming!
Bard was only wrong about the year though. Spoons disappeared as finger foods took over gaming. It was fun while that weird little moment in history lasted.
Actual old person here: the above is all made up.
The slots were for sporks.
I was trying to trigger the same output from Bard, and it answered correctly; they probably addressed that very specific case after it got some attention. Then of course I started messing around, first replacing "elephants" with "snakes" and then with "potatoes". And here's the outcome of the third prompt:
My sides went into orbit. The worst issue isn't even claiming that snakes aren't animals (contradicting the output of the second prompt, by the way), but the insane troll logic that Bard shows when it comes to hypothetical scenarios. In an imaginative scenario where potatoes have legs, the concept of "legless" does apply to them, because they would have legs to lose.
Just went into a coughing fit from this, its so laughable. So are snakes just not in the animal kingdom at all?? I bet they are in the same class as viruses.
So are snakes just not in the animal kingdom at all??
Of course they aren't! Snakes, fish, slugs, they're all moving fungi. (On another matter, each spider is 2x more of an animal than your typical mammal. 4x than birds and us.) /s
Potatoes, not being biological, are therefore mineral. This concludes today's lesson.
I corrected it (incorrectly) and it did slightly better. Then I corrected it again and it decided 10 elephants with proper leg counts should be 32.
How many legs could a legless potato have if a potato could be legless?
[I'll answer this seriously but dunno if you're being serious or just playing along the ridiculous setting.]
In this scenario, the matter here would be the other potatoes. Because, once you say that two of the four potatoes are legless, you're implying that the other two have legs; as such, the sum of the legs of all four potatoes would not be zero.
Got a fantastic response when asking it nonsense questions just now. This is actually incredibly impressive.
And here's GPT-4:
https://chat.openai.com/share/7a8eb92f-57f1-47fc-9a53-55187e25d892
Why it has un compiled latex?
It doesn't. It compiles.
The latex renders correctly in my browser
[Double reply to avoid editing my earlier comment]
From the HN thread:
It's a good example how this models are not answering based on any form of understanding and logic reasoning but probabilistic likelihood in many overlapping layers. // Through this also may not matter if this creates a good enough illusion of understanding and intelligence.
I think that the first sentence is accurate, but I disagree with the second one.
Probabilistic likelihood is not enough to create a good illusion of understanding/intelligence. Relying on it will create situations as in the OP, where the bot outputs nonsense because of an unexpected prompt.
To avoid that, the model would need some symbolic (or semantic, or conceptual) layer[s], and handle the concepts being conveyed by the tokens, not just the tokens themselves. But that's already closer to intelligence than to prob likelihood.
Could someone ask a todler the same question and report back?
If the elephants are octopi with legs that grow back, perhaps.