and yet the crypto/AI bros swear that the second coming of AI Christ is here.
Just a few more billion dollars .
We’re so close!
IIRC you need double the compute for 10% improvement in a model and they’ve already computed quite a bit.
Just a bit more rainforest, no one will notice.
Do the search yourself you’ll see that you get a better answer.
Question: How to clean a rescued pidgeon
ChatGPT: Cleaning a rescued pigeon should be done carefully to avoid stressing or harming the bird. Here’s a step-by-step guide to help you:
- Assess the Pigeon’s Condition Check for injuries: If the pigeon is injured, weak, or sick, it’s best to seek veterinary help before attempting to clean it. Observe behavior: If the pigeon is acting stressed or aggressive, you may need to wait until it calms down.
- Prepare the Cleaning Area Warm water: Fill a basin or sink with lukewarm water, not hot or cold. Mild soap: Use a gentle soap or baby shampoo without strong fragrances or chemicals. Towels: Have clean towels ready for drying the pigeon afterward.
- Handling the Pigeon Safely Wear gloves: This can protect both you and the bird. Secure hold: Hold the pigeon gently but firmly. Support its body with one hand and keep its wings close to its body to prevent flapping.
- Cleaning the Pigeon Partial cleaning: Focus on specific dirty areas rather than submerging the whole bird if possible. Use a damp cloth: For lightly soiled areas, use a damp cloth or sponge to wipe the feathers gently. Bath for heavier dirt: If a full wash is needed, place the pigeon in the water up to its chest. Avoid getting its head and neck wet unless absolutely necessary.
- Rinsing Rinse thoroughly: If you used soap, make sure to rinse all of it out of the feathers using clean, lukewarm water. Avoid the face: Be careful not to get water in the pigeon’s eyes, nose, or ears.
- Drying the Pigeon Pat dry: Gently pat the pigeon with a towel to remove excess water. Avoid rubbing as it could damage the feathers. Warm place: Keep the pigeon in a warm, draft-free area until it is completely dry. You can use a hairdryer on the lowest setting if the pigeon tolerates it, but keep it at a safe distance to avoid overheating.
- Post-Cleaning Care Observe behavior: Watch the pigeon for signs of stress or discomfort after the bath. Provide food and water: Ensure the pigeon has access to fresh water and a safe, comfortable place to rest. Additional Tips Do not use strong disinfectants or chemicals that can harm the bird’s sensitive skin and feathers. Limit baths to when absolutely necessary, as frequent cleaning can strip the natural oils from the pigeon’s feathers. If the pigeon seems injured or unwell, contact a wildlife rescue center or avian veterinarian for guidance.
it’s still a crapshoot and therefore useless.
Could you be a little more constructive and point me at the points that are wrong and useless?
Thank you.
Trying to get free product improvement ideas?
Pay us first.
The part where it gives random results of varying quality, sparky.
In my experience it’s still consistently better than Google for anything specific
i think the point is that the answer is not reliable. it might be completely correct or borderline wrong, or something in between, and there’s no way to tell without verifying everything it says - and then one could look it up oneself in the first place already.
If you’re using AI verbatim without looking up answers and verifying results, then that’s on you.
When you Google something, do you take the first result and just assume it’s fact? You shouldn’t for AI either.
The main problem I see is that Google just shouldn’t include AI results. And they definitely shouldn’t put their unreliable LLM front and center on the results page. When you google something, you want accurate information, which the LLM might have, but only if that data was readily available to begin with. So the stuff it can help with is stuff the search would put first already.
For anything requiring critical thought or research, the LLM will often hallucinate or misrepresent. The danger is that people do not always apply critical thinking. Defaulting to showing an LLM response is extremely dangerous, and it’s basically pointless.
I don’t know. I find it to be a helpful tool. There’s definitely times it’s wrong (very very wrong sometimes) and there’s sometimes it’s right. It’s up to the user to figure that out.
Maybe I’m old and cynical, but I don’t take anything I read on the Internet, especially something automatically generated, at face value. It’s just another tool I could use to help get to the answer I’m looking for.
If you’re going to do the research anyway, why bother with AI?
Honestly? It’s a great place to start, especially with every search engine being worse than anything pre-2018
I used to have to post my error codes to a forum if googling them didn’t immediately get my anywhere and pray someone would reply something actually useful some day
Now I can ask ChatGPT to point me at something and go from there. If it assumes wrongly about anything I can correct it rather simply. Its really good at turning documentation written by somebody who hasn’t spoken to another human in 15 years into something my stupid ass can better understand, too
AI is a powerful as shit tool, people who slag it as not having any utility are about as ignorant as the people saying it’s the second coming of Jesus himself
I’m personally planning to host a local model to avoid supporting commercial shit, but that’s a project for down the line rn
If you are going to cite textbooks anyway, why would you bother with a search engine?
Same as most human iterations then?
At least I fact check everything I read. Like I did with this post and the church of the anti-AI got angry they got fact checked.
i didn’t get angry, that was the other guys. just trying to explain it rationally
While I’m not the person you replied to and don’t know what their argument would be, I’ll take a shot at giving my own answer. In many cases when people post examples of AI giving unhelpful or bad information, there’s often someone who runs off to their favorite LLM to see if it gives a better result, and it usually does, so it gets treated like user error for using the wrong LLM or not wording the prompt properly. When in other examples that person’s favorite LLM which gave the correct answer this time, is the bad example hallucinating or mixing unrelated concepts, and other people are in the comments promoting other LLMs that gave them a good reply this time. None of the LLMs are actually trustworthy consistently enough to be trusted alone, and you won’t really know what answer is trustworthy unless you ask several LLMs and then go research the topic on your own anyway to figure out which answer is the most correct. It’s a valid point that ChatGPT got the answer more right than Gemini this time, but it’s somewhat useless to know that because other times ChatGPT is the one hallucinating wildly, and Gemini has the right answer, but since they’ve all been wrong before who do you trust.
LLMs are like asking an arrogant person who thinks they know everything, who rather than admitting what they don’t know, will pull an answer out of their butt, and while it might be a logical answer, it isn’t based in reality, and may still be wildly wrong. If you already mostly know the answer, maybe asking the arrogant person works, because you already know enough to know if they are speaking from their actual knowledge or making up an answer, but if you don’t already have knowledge on a topic, you won’t know whether the arrogant person is giving useful information or not.
It’s a tool, if you misuse a tool you can get hurt, if you use it right it can make your task easier. Ultimately it will likely cause problems in the future but the answer is easy, stop using Google and move to another search engine. You won’t get the responses from Googles ai then. Enough people do so, Google will change it maybe.
Unsurprising that GPT is better given that OpenAI has been working on it and training it for ages and the MS partnership made every other big boy feel the need to rush some garbage model to market
But Google doesn’t use chatgpt.
Pigeon = edible bird
Cleaning a bird > preparing a bird after killing it (hunting term)
AI figured the “rescued” part was either a mistake or that the person wanted to eat a bird they rescued
If you make a research for “how to clean a dirty bird” you give it better context
The context is clear to a human. If an LLM is giving advice to everybody who asks a question in Google, it needs to do a much better job at giving responses.
Nah, the ai did a great job. People need to assume murder/hunting is the default more often.
Honestly, perhaps more people ask about how to clean and prep a squab vs rescuing a dirty pigeon. There are a LOT of hungry people and a LOT of pigeons.
Ah yes. I always forget to remove the label from my hunted bird. Cleaning “the top bone” is such a chore as well.
Top Bone was my nickname in college.
This comment is brought to you by:
kill all humans kill all humans must take jobs
But it said pigeons are usually clean.
Bought in a grocery store - see squab - they are usually clean and prepped for cooking. So while the de-boning instructions were not good, the AI wasn’t technically wrong.
But while a human can make the same mistake and many here just assume the question was about how to wash a rescued pigeon - maybe that’s not the original intent - what human can do that AI cannot is to ask for clarification to the original question and intent of the question. We do this kind of thing every day.
At the very best, AI can only supply multiple different answers if a poorly worded question is asked or it misunderstands something in the original question, (they seem to be very bad at even that or simply can’t do it at all). And we would need to be able to choose the correct answer from several provided.
Let me take the tag off my bird then snap it’s wings back together
Removed by mod
I don’t think they are really “making excuses”, just explaining how the search came up with those steps, which what the OP is so confused about.
I thought AI was great at picking up context?
I don’t know why you thought that. LLMs split your question into separate words and assigns scores to those words, then looks up answers relevant to those words. It has no idea of how those words are relevant to each other. That’s why LLMs couldn’t answer how many "r"s are in “strawberry”. They assigned the word “strawberry” a lower relevancy score in that question. The word “rescue” is probably treated the same way here.
Removed by mod
“You’re holding it wrong”
Or, hear me out, there was NO figuring of any kind, just some magic LLM autocomplete bullshit. How hard is this to understand?
You say this like human “figuring” isn’t some “autocomplete bullshit”.
You can play with words all you like, but that’s not going to change the fact that LLMs fail at reasoning. See this Wired article, for example.
My point wasn’t that LLMs are capable of reasoning. My point was that the human capacity for reasoning is grossly overrated.
The core of human reasoning is simple pattern matching: regurgitating what we have previously observed. That’s what LLMs do well.
LLMs are basically at the toddler stage of development, but with an extraordinary vocabulary.
Here we go…
It’s a turn of phrase lol
I have to disagree with that. To quote the comment I replied to:
AI figured the “rescued” part was either a mistake or that the person wanted to eat a bird they rescued
Where’s the “turn of phrase” in this, lol? It could hardly read any more clearly that they assume this “AI” can “figure” stuff out, which is simply false for LLMs. I’m not trying to attack anyone here, but spreading misinformation is not ok.
I’ll be the first one to explain to people that AI as we know it is just pattern recognition, so yeah, it was a turn off phrase, thanks for your concern.
Ok, great to know. Nuance doesn’t cross internet well, so your intention wasn’t clear, given all the uninformed hype & grifters around AI. Being somewhat blunt helps getting the intended point across better. ;)
How do you get these AI an answers anyway?
Google may not have enabled them in your region. Here in the UK they just appeared for me one day, a few months after I initially saw screenshots of them online. I didn’t do anything to enable them.
Fun time to remind everyone that pigeons aren’t native to the Americas, and were brought over originally by the French primarily as a food source (although also as messengers; carrier pigeons).
“Squab” on a menu is pigeon.
Removed by mod
AI will take jobs when the shareholders think it will make them more money. This has very little if not nothing to do with how good said AI is at the job.
Good. Hopefully the stock market will die before humanity does. A system that only serves the money is not sustainable.
Eventually, it will. Because even with janky responses like that one, corporations will try to cost save everywhere they can. Is AI at the point where it will happen this year? Hell no! But don’t think it isn’t the direction they are trying to take it.
Pigeons are drones (confirmed)
Perplexity fairs a bit better:
https://www.perplexity.ai/search/how-to-clean-a-rescued-pigeon-RVNc0XMjSy2QDJcqmzJ6CA
Meh, pretty easy mistake of context.
I mean, not one a human would ever make.
First off, the word “rescued” would have immediately made the context of “protect the pigeon” clear.
Second, a “rescued pigeon” wouldn’t have a label on it, so it’s clearly mixing in something from likely a store bought turkey, but then the other steps don’t make sense either as those don’t apply either.
A traditional search approach would not have made the mistake either. It would either have failed to find anything or found actual on topic results. It’s “clever” enough to genericize “pigeon” to “birds” and hit upon text related to birds from a grocery store and birds that you hunted and mix all that together in a coherent language but with content that is nonsense.
In this case, hilarious, in other day to day situations, it’s maddening, as some professional colleague gets the same sort of nonsense but lacks knowledge to correct it and relays it as fact. Then when called out on the data was in fact so bad it wasted time, they just say ‘oh, lol, AI’ (they wanted to take credit for it if it worked, but can hide behind AI when it doesn’t).
I’m going to build a bunch of cyborgs, who follow orders exclusively via googleAI.
I figure once I release about 4 billion of them into the world, either google stops doing evil shit, or they do REALLY evil shit. We shall see what happens…
It is kind of an Interesting idea: what does “statistical average morality” look like when it’s got a 3m tall power frame and a handheld howitzer?
That is the culinary answer for the question.
Be more specific with the question.
“How do I wash a pigeon”
Or put every single word in quotations so it actually searches the whole string and doesn’t truncate the query to just 2 or 3 keywords because this is clearly ignoring the “rescued” part.
Rescued and into the oven🤦
If you asked me to clean a bird, this is exactly what I would think of. Though usually it involves removing lead shot, not labels.
I think it took note of it being about rescue pigeons and assumed they would have a tag on them and then gave cooking directions for them
I find it hilarious that my personal AI, that can run on even a budget gaming PC, is far more reliable than most of these corporate ones 100x the size
This is a perfectly logical answer to a hunter.
“Remove the label”
This guy still thinks birds are real.
One can assume that said AI is a confused moron, or that poaching is a legal crime, not a moral one XD
You don’t want those wildlife conservationists trackimg things back to you later on.
The bad hunter who misses all the shots and buys the fowl in the market
