Baybe what's wrong? You've barely touched your Strawberrum

Googlum is brokum

I just tried to have Gemini navigate to the nearest Starbucks and the POS found one 8hrs and 38mins away.
Absolute trash.
- Just tried it with Target and again, it's sending me to Raleigh, North Carolina.
Despite that, it delivers its results with much applum!
- Quality pum

Some "AI" LLMs resort to light hallucinations. And then ones like this straight-up gaslight you!

Factual accuracy in LLMs is "an area of active research", i.e. they haven't the foggiest how to make them stop spouting nonsense.
- duckduckgo figured this out quite a while ago: just fucking summarize wikipedia articles and link to the precise section it lifted text from
- Because accuracy requires that you make a reasonable distinction between truth and fiction, and that requires context, meaning, understanding. Hell, full humans aren't that great at this task. This isn't a small problem, I don't think you solve it without creating AGI.
- Ha!
  You buy this? (Believe it’s incredibly expensive)
MFer accidentally got "plum" right and didn't even know it...

Coconutum

You did WHAT to em?
- I barely know 'em!

Ok, let me try listing words that ends in "um" that could be (even tangentially) considered food.

Plum
Gum
Chum
Rum
Alum
Rum, again
Sea People

I think that's all of them.

Your mum.
- Happy belated Mother's Day?
The Sea Peoples consumed by the Late Bronze Age collapse (or were a catalysts thereof)?
Or just people at sea eaten by krakens? Cause they definitely count.
- It's a dirty joke.
Bum
Cryptosporidium
Cum
Opium
Possum
Scrotum
Scum
All very Yum for the Tum!

Totally reproducible, just with slightly different prompts.

There's going to be an entire generation of people growing up with this and "learning" this way. It's like every tech company got together and agreed to kill any chance of smart kids.
- Isn't it the opposite? Kids see so many examples of obviously wrong answers they learn to check everything
- This makes me happy since I won't lose my job to newbies.

And yet it doesn’t even list ‘Plum’, or did it think ‘Applum’ was just a variation of a plum?

Well, plum originally comes from applum which morphed into a plum so yeah.
And that's absolutely not true.
- I will start using this as a factoid

Strawberrum sounds like it'll be at least 20% abv. I'd like a nice cold glass of that.

Strawberrum? Barely knew 'em!

It's crazy how bad d AI gets of you make it list names ending with a certain pattern. I wonder why that is.

I'm not an expert, but it has something to do with full words vs partial words. It also can't play wordle because it doesn't have a proper concept of individual letters in that way, its trained to only handle full words
- they don't even handle full words, it's just arbitrary groups of characters (including space and other stuff like apostrophe afaik) that is represented to the software as indexes on a list, it literally has no clue what language even is, it's a glorified calculator that happens to work on words.
- That's interesting, didn't know
It can't see what tokens it puts out, you would need additional passes on the output for it to get it right. It's computationally expensive, so I'm pretty sure that didn't happen here.
- doesn't it work literally by passing in everything it said to determine what the next word is?
- With the amount of processing it takes to generate the output, a simple pass over the to-be final output would make sense...
LLMs aren't really capable of understanding spelling. They're token prediction machines.
LLMs have three major components: a massive database of "relatedness" (how closely related the meaning of tokens are), a transformer (figuring out which of the previous words have the most contextual meaning), and statistical modeling (the likelihood of the next word, like what your cell phone does.)
LLMs don't have any capability to understand spelling, unless it's something it's been specifically trained on, like "color" vs "colour" which is discussed in many training texts.
"Fruits ending in 'um' " or "Australian towns beginning with 'T' " aren't talked about in the training data enough to build a strong enough relatedness database for, so it's incapable of answering those sorts of questions.

Gemini thought we name food like we name a periodic table

plutonium is food once
- A gram of plutonium has enough calories to last you the rest of your life.

Applum bananum jeans, boots with the fur.

AI is truly going to change the world.

30 years from now: "Haven't they always been called Strawberrums?"
- We've always been at war with Bananum!

Looks like someone set Google to "Herakles Mode".

Tomatum.. that's the one

Ok, I feel like there has been more than enough articles to explain that these things don't understand logic. Seriously. Misunderstanding their capabilities at this point is getting old. It's time to start making stupid painful.

I cannot reproduce this on Google.

Maybe they fixed it, I was able to reproduce it on Mixtral

Same vibe: HHGregg is Havaving a Summummer Sale

This is so irresponsible, kids have access to tablets and search and might learn wrong things.

Make em say ummmmmmm na na na naaaaa

The texture is intriguing... https://www.youtube.com/watch?v=rw7PUrgU3N0

This is the tech you're all afraid of?

Evaporated 20 gallons of water in Nevada for this
- Reminds me of how the "1800 gallons for one burger" statistic uses annual rainfall to calculate that as if it was captured, stored, and used from our kitchen sinks.