This is what I call a “wild goose chase”. LLMs like to do that sort of thing, and it’s your job to notice that and pull the plug as soon as possible. If you’re not qualified to evaluate output, the risk of an aimless never ending chase increases dramatically. The good thing is, that you’ll learn to be more careful next time. Take those half hallucinated answers with a spoon of salt.
For me the question is “would I ask a complete dumbass to google this for me”? If I ever end up with a yes, I may consider asking AI. Though I still probably won’t, because I can’t stand the way they write.
Yeah, that’s a pretty safe approach. You never know which parts are crude averages of the training data and which parts got interpolated.
If you happen have subject matter expertise, you can tell where it all went wrong. On the other hand, at that point you already know what the right answer is so why bother asking an LLM in the first place. It’s a very paradoxical situation.
Makes me wonder if there are people out there right now on some crazy AI goose chase. Just taking them all over the place doing random nonsensical tasks.
The chances are very high. People always have technical problems, and they are also impatient enough to use an LLM. Why spend hours reading stackoverflow, and try solutions that aren’t exactly what you’re currently facing? That takes time, effort and improvisation skills.
Meanwhile, you could type a question to an LLM, and get an answer in a few seconds. In the best case scenario, you’ll get something useful out of it, but you could also start another wild goose chase. Humans are lazy, so they’ll fall into this trap very easily. It’s a gamble.
I’m sure it happens all the times because it doesn’t always have all the expectations/understanding of what the end result may look like when completed. Add that to the fact that it isn’t actually knowing to check for limiting factors you will run into later, because it is mostly just regurgitating ideas that worked in specific scenarios
LLMs can be good for troubleshooting, but if the first suggestion doesn’t work, the 10th won’t either. Never do any LLM “fixes” that you don’t know are reversible.
I have made a few agents in Mistral, and I think I also added a line about asking clarifying questions. Should probably make another one just for troubleshooting. There should be many lines about critical thinking, ruling out the obvious, avoiding assumptions, and asking questions.
Well, that was just a quick summary of an idea. You can’t just tell someone (or a an LLM) to think critically and expect it to work magically. However, you can follow certain guidelines to take a few steps in the right direction. There is a method to it you know. It starts with simple things like investigating whether or not a claim is supported by the evidence, are there unstated assumptions and so on. If you follow rules like this, you can avoid some of the simplest reasoning mistakes. Putting all of that into a prompt takes some time and effort. Doing it properly will probably result in a 500 line long manual of critical thinking.
I’ve done something similar with programming, and the LLM usually follows most of my style guide instructions reasonably well. It’s not perfect though, and it will deviate from some rules no matter how hard you try. Also, it requires constant hand holding, because it’s an LLM. Anyway, I don’t expect a critical thinking agent to be a fool-proof solution. As long as it can avoid some of the stupidest reasoning mistakes, it could be a bit more useful than the default version, and might actually be helpful in troubleshooting.
Googling the solution yourself can honestly be such a pain in the ass for some problems. Search engines have become so shite. You’ll find some stack overflow question from 8 years ago that’s no longer relevant, and in the end you might have to consult documentation directly which can take many minutes or hours. I reaaaally hate to say it but LLMs are actually useful for this usecase
Oh and the man pages of terminal commands. Usually it feels like those were written for the people who already know all about the command, but simply forgot which flag does what. Like, was that -n or -t again? Oh, let’s check the man pages. Got it, it’s -n, so let’s go with that.
What about those who have never used that command before? I feel like the man pages are the wrong place for people like that. You could spend 20 minutes reading and get very little value out of it. Instead, you could spend 2 minutes reading a tutorial blog, adapt the command to your use and get on with your life.
Alternatively, you could ask an LLM to generate a command that does what you want and… hope for the best. Hallucinations do exist. However, I’ve also discovered some awesome commands this way. In simple cases, it’s surprisingly fast. In complex cases, you’ll find yourself on a wild goose chase again.
Google’s quality has been nosediving for years. I don’t use it for searches any more because it’s practically useless. I hear Kagi is good, but it is a paid service.
If you have a list of commands to type in the terminal, but you need them only once, you aren’t going to make a bash script for that. If you think you’ll need to type that thing every week, making a script suddenly begins to make sense.
Same thing with agents. Do you think you’ll be asking these kinds of questions again in the future? If so, making an agent for it could make sense.
True, I wonder what the AI was chasing. I would imagine you could download the updated bios for that board to a USB using another machine and maybe get the bios working again, then they may be able to salvage things. Dells bios recovery can usually bypass any USB boot restrictions that were in the previous bios settings
It’s usually some random thought that is in the right neighborhood, but not quite spot on. A human troubleshooter would straight up say that it’s impossible to tell you what the problem is, so you would need to narrow it down by testing a few things.
An LLM just says that you need to update drivers or whatever. If the problem is caused by something obscure (like this one), an LLM will never be able to figure it out. This kind of stuff apparently just doesn’t exist in the training data, so there’s no way for the model to extrapolate and reach the right conclusion. Instead, it will continuously interpolate with the data it has, and you’ll end up with an infinite list of wrong answers.
Sycophancy really doesn’t help either. If you have any ideas what might cause the problem, the LLM will cling to those, no matter how wrong you might be. Troubleshooting requires critical thinking and LLMs don’t seem to be very good at that.
This is what I call a “wild goose chase”. LLMs like to do that sort of thing, and it’s your job to notice that and pull the plug as soon as possible. If you’re not qualified to evaluate output, the risk of an aimless never ending chase increases dramatically. The good thing is, that you’ll learn to be more careful next time. Take those half hallucinated answers with a spoon of salt.
Unfortunately, most clankers don’t think this way. It’s always either a “rare” fluke, or just because they didn’t use the most expensive model
LOL. Just throw more money at the problem until it goes away. Sure… that worked great last time.
I have a better suggestion: assume literally everything AI says is made up nonsense (because it is) and never ask it anything or read its summaries.
For me the question is “would I ask a complete dumbass to google this for me”? If I ever end up with a yes, I may consider asking AI. Though I still probably won’t, because I can’t stand the way they write.
Oh and the high-impact language they use feels really jarring and out of place. Everything is just so intense with these LLMs.
Yeah, that’s a pretty safe approach. You never know which parts are crude averages of the training data and which parts got interpolated.
If you happen have subject matter expertise, you can tell where it all went wrong. On the other hand, at that point you already know what the right answer is so why bother asking an LLM in the first place. It’s a very paradoxical situation.
Makes me wonder if there are people out there right now on some crazy AI goose chase. Just taking them all over the place doing random nonsensical tasks.
The chances are very high. People always have technical problems, and they are also impatient enough to use an LLM. Why spend hours reading stackoverflow, and try solutions that aren’t exactly what you’re currently facing? That takes time, effort and improvisation skills.
Meanwhile, you could type a question to an LLM, and get an answer in a few seconds. In the best case scenario, you’ll get something useful out of it, but you could also start another wild goose chase. Humans are lazy, so they’ll fall into this trap very easily. It’s a gamble.
I’m sure it happens all the times because it doesn’t always have all the expectations/understanding of what the end result may look like when completed. Add that to the fact that it isn’t actually knowing to check for limiting factors you will run into later, because it is mostly just regurgitating ideas that worked in specific scenarios
LLMs can be good for troubleshooting, but if the first suggestion doesn’t work, the 10th won’t either. Never do any LLM “fixes” that you don’t know are reversible.
It’s been a mixed bag for me. The case has to be a relatively simple one for it to work. LLMs need a lot of hand holding to be useful.
I find I need to either/both:
I have made a few agents in Mistral, and I think I also added a line about asking clarifying questions. Should probably make another one just for troubleshooting. There should be many lines about critical thinking, ruling out the obvious, avoiding assumptions, and asking questions.
You people do realize that this isn’t possible, right? LLMs do not “think”, let alone think critically.
Well, that was just a quick summary of an idea. You can’t just tell someone (or a an LLM) to think critically and expect it to work magically. However, you can follow certain guidelines to take a few steps in the right direction. There is a method to it you know. It starts with simple things like investigating whether or not a claim is supported by the evidence, are there unstated assumptions and so on. If you follow rules like this, you can avoid some of the simplest reasoning mistakes. Putting all of that into a prompt takes some time and effort. Doing it properly will probably result in a 500 line long manual of critical thinking.
I’ve done something similar with programming, and the LLM usually follows most of my style guide instructions reasonably well. It’s not perfect though, and it will deviate from some rules no matter how hard you try. Also, it requires constant hand holding, because it’s an LLM. Anyway, I don’t expect a critical thinking agent to be a fool-proof solution. As long as it can avoid some of the stupidest reasoning mistakes, it could be a bit more useful than the default version, and might actually be helpful in troubleshooting.
By the time you have added all of these you have just googled the solution yourself?
Googling the solution yourself can honestly be such a pain in the ass for some problems. Search engines have become so shite. You’ll find some stack overflow question from 8 years ago that’s no longer relevant, and in the end you might have to consult documentation directly which can take many minutes or hours. I reaaaally hate to say it but LLMs are actually useful for this usecase
Oh and the man pages of terminal commands. Usually it feels like those were written for the people who already know all about the command, but simply forgot which flag does what. Like, was that -n or -t again? Oh, let’s check the man pages. Got it, it’s -n, so let’s go with that.
What about those who have never used that command before? I feel like the man pages are the wrong place for people like that. You could spend 20 minutes reading and get very little value out of it. Instead, you could spend 2 minutes reading a tutorial blog, adapt the command to your use and get on with your life.
Alternatively, you could ask an LLM to generate a command that does what you want and… hope for the best. Hallucinations do exist. However, I’ve also discovered some awesome commands this way. In simple cases, it’s surprisingly fast. In complex cases, you’ll find yourself on a wild goose chase again.
omg this. The convention should be to have at least the first page be a quick introduction with examples of the most common commands.
Google’s quality has been nosediving for years. I don’t use it for searches any more because it’s practically useless. I hear Kagi is good, but it is a paid service.
If you have a list of commands to type in the terminal, but you need them only once, you aren’t going to make a bash script for that. If you think you’ll need to type that thing every week, making a script suddenly begins to make sense.
Same thing with agents. Do you think you’ll be asking these kinds of questions again in the future? If so, making an agent for it could make sense.
True, I wonder what the AI was chasing. I would imagine you could download the updated bios for that board to a USB using another machine and maybe get the bios working again, then they may be able to salvage things. Dells bios recovery can usually bypass any USB boot restrictions that were in the previous bios settings
The AI probably forgot what the initial prompt was halfway through the conversation due to context length lol
It’s usually some random thought that is in the right neighborhood, but not quite spot on. A human troubleshooter would straight up say that it’s impossible to tell you what the problem is, so you would need to narrow it down by testing a few things.
An LLM just says that you need to update drivers or whatever. If the problem is caused by something obscure (like this one), an LLM will never be able to figure it out. This kind of stuff apparently just doesn’t exist in the training data, so there’s no way for the model to extrapolate and reach the right conclusion. Instead, it will continuously interpolate with the data it has, and you’ll end up with an infinite list of wrong answers.
Sycophancy really doesn’t help either. If you have any ideas what might cause the problem, the LLM will cling to those, no matter how wrong you might be. Troubleshooting requires critical thinking and LLMs don’t seem to be very good at that.