You must log in or register to comment.
You asked a stupid question and got a stupid response, seems fine to me.
Yes, nobody asking that question is wonderring about the “straw” part of the word. They’re asking, is the “berry” part one, or two "r"s
“strawbery” has 2 R’s in it while “strawberry” has 3.
Fucking AI can’t even count.
A guy is driving around the back woods of Montana and he sees a sign in front of a broken down shanty-style house: ‘Talking Dog For Sale.’
He rings the bell and the owner appears and tells him the dog is in the backyard.
The guy goes into the backyard and sees a nice looking Labrador Retriever sitting there.
“You talk?” he asks.
“Yep” the Lab replies.
After the guy recovers from the shock of hearing a dog talk, he says, “So, what’s your story?”
The Lab looks up and says, “Well, I discovered that I could talk when I was pretty young. I wanted to help the government, so I told the CIA. In no time at all they had me jetting from country to country, sitting in rooms with spies and world leaders, because no one figured a dog would be eavesdropping, I was one of their most valuable spies for eight years running… but the jetting around really tired me out, and I knew I wasn’t getting any younger so I decided to settle down. I signed up for a job at the airport to do some undercover security, wandering near suspicious characters and listening in. I uncovered some incredible dealings and was awarded a batch of medals. I got married, had a mess of puppies, and now I’m just retired.”
The guy is amazed. He goes back in and asks the owner what he wants for the dog.
“Ten dollars” the guy says.
“Ten dollars? This dog is amazing! Why on Earth are you selling him so cheap?”
“Because he’s a liar. He’s never been out of the yard.”
Doc: That’s an interesting name, Mr…
Fletch: Babar.
Doc: Is that with one B or two?
Fletch: One. B-A-B-A-R.
Doc: That’s two.
Fletch: Yeah, but not right next to each other, that’s what I thought you meant.
Doc: Isn’t there a children’s book about an elephant named Babar.
Fletch: Ha, ha, ha. I wouldn’t know. I don’t have any.
Doc: No children?
Fletch: No elephant books.
The terrifying thing is everyone criticising the LLM as being poor, however it excelled at the task.
The question asked was how many R in strawbery and it answered. 2.
It also detected the typo and offered the correct spelling.
What’s the issue I’m missing?
Uh oh, you’ve blown your cover, robot sir.
There’s also a “r” in the first half of the word, “straw”, so it was completely skipping over that r and just focusing on the r’s in the word “berry”
It wasn’t focusing on anything. It was generating text per its training data. There’s no logical thought process whatsoever.
The issue that you are missing is that the AI answered that there is 1 ‘r’ in ‘strawbery’ even though there are 2 'r’s in the misspelled word. And the AI corrected the user with the correct spelling of the word ‘strawberry’ only to tell the user that there are 2 'r’s in that word even though there are 3.
Sure, but for what purpose would you ever ask about the total number of a specific letter in a word? This isn’t the gotcha that so many think it is. The LLM answers like it does because it makes perfect sense for someone to ask if a word is spelled with a single or double “r”.
It makes perfect sense if you do mental acrobatics to explain why a wrong answer is actually correct.
Not mental acrobatics, just common sense.
“My hammer is not well suited to cut vegetables” 🤷
There is so much to say about AI, can we move on from “it can’t count letters and do math” ?
But the problem is more “my do it all tool randomly fails at arbitrary tasks in an unpredictable fashion” making it hard to trust as a tool in any circumstances.
Your not supposed to just trust it. Your supposed to test the solution it gives you. Yes that makes it not useful for some things. But still immensely useful for other applications and a lot of times it gives you a really great jumping off point to solving whatever your problem is.
I get that it’s usually just a dunk on AI, but it is also still a valid demonstration that AI has pretty severe and unpredictable gaps in functionality, in addition to failing to properly indicate confidence (or lack thereof).
People who understand that it’s a glorified autocomplete will know how to disregard or prompt around some of these gaps, but this remains a litmus test because it succinctly shows you cannot trust an LLM response even in many “easy” cases.
I’ve already had more than one conversation where people quote AI as if it were a source, like quoting google as a source. When I showed them how it can sometimes lie and explain it’s not a primary source for anything I just get that blank stare like I have two heads.
Works fine for me in o3-mini-high:
Counting letters in “strawberry”
Alright, I’m checking: the word “strawberry” is spelled S T R A W B E R R Y. Let me count the letters: S (1), T (2), R (3), A (4), W (5), B (6), E (7), R (8), R (9), Y (10). There are three R’s: in positions 3, 8, and 9. So, the answer is 3. Even if we ignore case, the count still holds. Therefore, there are 3 r’s in “strawberry.”
A normal person would say ‘strawberry with two "r"s’
Finally! With a household energy consumption for one day we can count how many Rs are in strawberry.
I think I have seen this exact post word for word fifty times in the last year.
And yet they apparently still can’t get an accurate result with such a basic query.
Meanwhile… https://futurism.com/openai-signs-deal-us-government-nuclear-weapon-security
These models don’t get single characters but rather tokens repenting multiple characters. While I also don’t like the “AI” hype, this image is also very 1 dimensional hate and misreprents the usefulness of these models by picking one adversarial example.
Today ChatGPT saved me a fuckton of time by linking me to the exact issue on gitlab that discussed the issue I was having (full system freezes using Bottles installed with flatpak on Arch). This was the URL it came up with after explaining the problem and giving it the first error I found in dmesg: https://gitlab.archlinux.org/archlinux/packaging/packages/linux/-/issues/110
This issue is one day old. When I looked this shit up myself I found exactly nothing useful on both DDG or Google. After this ChatGPT also provided me with the information that the LTS kernel exists and how to install it. Obviously I verified that stuff before using it, because these LLMs have their limits. Now my system works again, and figuring this out myself would’ve cost me hours because I had no idea what broke. Was it flatpak, Nvidia, the kernel, Wayland, Bottles, some random shit I changed in a config file 2 years ago? Well thanks to ChatGPT I know.
They’re tools, and they can provide new insights that can be very useful. Just don’t expect them to always tell the truth, or to actually be human-like
Just don’t expect them to always tell the truth, or to actually be human-like
I think the point of the post is to call out exactly that: people preaching AI as replacing humans
it can, in the same way a loom did, just for more language-y tasks, a multimodal system might be better at answering that type of question by first detecting that this is a question of fact and that using a bucket sort algorithm on the word “strawberry” will answer the question better than it’s questionably obtained correlations.
There is an alternative reality out there where LLMs were never marketed as AI and were marketed as random generator.
In that world, tech savvy people would embrace this tech instead of having to constantly educate people that it is in fact not intelligence.
Because you’re using it wrong. It’s good for generative text and chains of thought, not symbolic calculations including math or linguistics
Because you’re using it wrong.
No, I think you mean to say it’s because you’re using it for the wrong use case.
Well this tool has been marketed as if it would handle such use cases.
I don’t think I’ve actually seen any AI marketing that was honest about what it can do.
I personally think image recognition is the best use case as it pretty much does what it promises.
So for something you can’t objectively evaluate? Looking at Apple’s garbage generator, LLMs aren’t even good at summarising.
For reference:
AI chatbots unable to accurately summarise news, BBC finds
the BBC asked ChatGPT, Copilot, Gemini and Perplexity to summarise 100 news stories and rated each answer. […] It found 51% of all AI answers to questions about the news were judged to have significant issues of some form. […] 19% of AI answers which cited BBC content introduced factual errors, such as incorrect factual statements, numbers and dates.
It makes me remember I basically stopped using LLMs for any summarization after this exact thing happened to me. I realized that without reading the text, I wouldn’t be able to know whether the output has all the relevant info or if it has some made-up info.
Give me an example of how you use it.
I have it write for me emails in German. I moved there not too long ago, works wonders to get doctors appointment, car service, etc. I also have it explain the text, so I’m learning the language.
I also use it as an alternative to internet search, which is now terrible. It’s not going to help you to find smg super location specific, but I can ask it to tell me without spoilers smg about a game/movie or list metacritic scores in a table, etc.
It also works great in summarizing long texts.
LLM is a tool, what matters is how you use it. It is stupid, it doesn’t think, it’s mostly hype to call it AI. But it definitely has it’s benefits.
Ask it for a second opinion on medical conditions.
Sounds insane but they are leaps and bounds better than blindly Googling and self prescribe every condition there is under the sun when the symptoms only vaguely match.
Once the LLM helps you narrow in on a couple of possible conditions based on the symptoms, then you can dig deeper into those specific ones, learn more about them, and have a slightly more informed conversation with your medical practitioner.
They’re not a replacement for your actual doctor, but they can help you learn and have better discussions with your actual doctor.
sounds like a perfectly sane idea https://freethoughtblogs.com/pharyngula/2025/02/05/ai-anatomy-is-weird/
So can web MD. We didn’t need AI for that. Googling symptoms is a great way to just be dehydrated and suddenly think you’re in kidney failure.
We didn’t stop trying to make faster, safer and more fuel efficient cars after Model T, even though it can get us from place A to place B just fine. We didn’t stop pushing for digital access to published content, even though we have physical libraries. Just because something satisfies a use case doesn’t mean we should stop advancing technology.
We also didn’t make the model T suggest replacing the engine when the oil light comes on. Cars, as it happens, aren’t that great at self diagnosis, despite that technology being far simpler and further along than generative models are. I don’t trust the model to tell me what temperature to bake a cake at, I’m sure at hell not going to trust it with medical information. Googling symptoms was risky at best before. It’s a horror show now.
AI is slower and less efficient than the older search algorithms and is less accurate.
We have one that indexes all the wikis and GDocs and such at my work and it’s incredibly useful for answering questions like “who’s in charge of project 123?” or “what’s the latest update from team XYZ?”
I even asked it to write my weekly update for MY team once and it did a fairly good job. The one thing I thought it had hallucinated turned out to be something I just hadn’t heard yet. So it was literally ahead of me at my own job.
I get really tired of all the automatic hate over stupid bullshit like this OP. These tools have their uses. It’s very popular to shit on them. So congratulations for whatever agreeable comments your post gets. Anyway.
Here’s a bit of code that’s supposed to do stuff. I got this error message. Any ideas what could cause this error and how to fix it? Also, add this new feature to the code.
Works reasonably well as long as you have some idea how to write the code yourself. GPT can do it in a few seconds, debugging it would take like 5-10 minutes, but that’s still faster than my best. Besides, GPT is also fairly fluent in many functions I have never used before. My approach would be clunky and convoluted, while the code generated by GPT is a lot shorter.
If you’re well familiar with the code you’ve working on, GPT code will be convoluted by comparison. If so, you can ask GPT for the rough alpha version, and you can do the debugging and refining in a few minutes.
That makes sense as long as you’re not writing code that needs to know how to do something as complex as …checks original post… count.
It can do that just fine, because it has seen enough examples of working code. It can’t directly count correctly, sure, but it can write “i++;”, incrementing a variable by one in a loop and returning the result. The computer running the generated program is going to be doing the counting.
Writing customer/company-wide emails is a good example. “Make this sound better: we’re aware of the outage at Site A, we are working as quick as possible to get things back online”
Dumbing down technical information “word this so a non-technical person can understand: our DHCP scope filled up and there were no more addresses available for Site A, which caused the temporary outage for some users”
Another is feeding it an article and asking for a summary, https://hackingne.ws/ does that for its Bsky posts.
Coding is another good example, “write me a Python script that moves all files in /mydir to /newdir”
Asking for it to summarize a theory or protocol, “explain to me why RIP was replaced with RIPv2, and what problems people have had since with RIPv2”
Make this sound better: we’re aware of the outage at Site A, we are working as quick as possible to get things back online
How does this work in practice? I suspect you’re just going to get an email that takes longer for everyone to read, and doesn’t give any more information (or worse, gives incorrect information). Your prompt seems like what you should be sending in the email.
If the model (or context?) was good enough to actually add useful, accurate information, then maybe that would be different.
I think we’ll get to the point really quickly where a nice concise message like in your prompt will be appreciated more than the bloated, normalised version, which people will find insulting.
Yes, people are using it as the least efficient communication protocol ever.
One side asks an LLM to expand a summary into a fluff filled email, and the other side asks an LLM to reduce the long email to a summary.
Yeah, normally my “Make this sound better” or “summarize this for me” is a longer wall of text that I want to simplify, I was trying to keep my examples short. Talking to non-technical people about a technical issue is not the easiest for me, AI has helped me dumb it down when sending an email, and helps correct my shitty grammar at times.
As for accuracy, you review what it gives you, you don’t just copy and send it without review. Also you will have to tweak some pieces that it gives out where it doesn’t make the most sense, such as if it uses wording you wouldn’t typically use. It is fairly accurate though in my use-cases.
Hallucinations are a thing, so validating what it spits out is definitely needed.
Another example: if you feel your email is too stern or gives the wrong tone, I’ve used it for that as well. “Make this sound more relaxed: well maybe if you didn’t turn off the fucking server we wouldn’t of had this outage!” (Just a silly example)
The dumbed down text is basically as long as the prompt. Plus you have to double check it to make sure it didn’t have outrage instead of outage just like if you wrote it yourself.
How do you know the answer on why RIP was replaced with RIPv2 is accurate and not just a load of bullshit like putting glue on pizza?
Are you really saving time?
Dumbed down doesn’t mean shorter.
If the amount of time it takes to create the prompt is the same as it would have taken to write the dumbed down text, then the only time you saved was not learning how to write dumbed down text. Plus you need to know what dumbed down text should look like to know if the output is dumbed down but still accurate.
Yes, I’m saving time. As I mentioned in my other comment:
Yeah, normally my “Make this sound better” or “summarize this for me” is a longer wall of text that I want to simplify, I was trying to keep my examples short.
And
and helps correct my shitty grammar at times.
And
Hallucinations are a thing, so validating what it spits out is definitely needed.
How do you validate the accuracy of what it spits out?
Why don’t you skip the AI and just use the thing you use to validate the AI output?
Most of what I’m asking it are things I have a general idea of, and AI has the capability of making short explanations of complex things. So typically it’s easy to spot a hallucination, but the pieces that I don’t already know are easy to Google to verify.
Basically I can get a shorter response to get the same outcome, and validate those small pieces which saves a lot of time (I no longer have to read a 100 page white paper, instead a few paragraphs and then verify small bits)
it’s not good for summaries. often gets important bits wrong, like embedded instructions that can’t be summarized.
My experience has been very different, I do have to sometimes add to what it summarized though. The Bsky account mentioned is a good example, most of the posts are very well summarized, but every now and then there will be one that isn’t as accurate.
“You’re holding it wrong”
