Unpopular Opinion
Welcome to the Unpopular Opinion community!
How voting works:
Vote the opposite of the norm.
If you agree that the opinion is unpopular give it an arrow up. If it's something that's widely accepted, give it an arrow down.
Guidelines:
Tag your post, if possible (not required)
- If your post is a "General" unpopular opinion, start the subject with [GENERAL].
- If it is a Lemmy-specific unpopular opinion, start it with [LEMMY].
Rules:
1. NO POLITICS
Politics is everywhere. Let's make this about [general] and [lemmy] - specific topics, and keep politics out of it.
2. Be civil.
Disagreements happen, but that doesn’t provide the right to personally attack others. No racism/sexism/bigotry. Please also refrain from gatekeeping others' opinions.
3. No bots, spam or self-promotion.
Only approved bots, which follow the guidelines for bots set by the instance, are allowed.
4. Shitposts and memes are allowed but...
Only until they prove to be a problem. They can and will be removed at moderator discretion.
5. No trolling.
This shouldn't need an explanation. If your post or comment is made just to get a rise with no real value, it will be removed. You do this too often, you will get a vacation to touch grass, away from this community for 1 or more days. Repeat offenses will result in a perma-ban.
6. Defend your opinion
This is a bit of a mix of rules 4 and 5 to help foster higher quality posts. You are expected to defend your unpopular opinion in the post body. We don't expect a whole manifesto (please, no manifestos), but you should at least provide some details as to why you hold the position you do.
Instance-wide rules always apply. https://legal.lemmy.world/tos/
view the rest of the comments
Your disclaimer is false. You said the post isn't about AI but it's entirely about AI. Because look, we all know that there's a backlash against AI. We all know that if you're planning an event, you probably shouldn't use AI if you aren't ready to piss off people who don't like it... That's the reality. Now, you could wish that reality were different, but it's not a big burden to say hey, don't have AI make the poster, just make it yourself. It's really not that hard.
Now, you also mentioned people who were accused of using AI when they didn't. Of course, that's not cool... But this is not a problem of reactions to AI. It's a problem of not fact checking. And if you came here telling us that people should fact-check better, or properly, than I think everyone would agree with you.
In summary, I think your post does the opposite of what you claim you want it to do. You don't want people to be divisive, but you've intentionally created a divisive set of guidelines. And then you've told people to shut up but you know they're not going to which means you're just looking for an argument... I guess this entire strategy bores me.
But you posted in the right place, so it's all good.
What the flog is wrong with having AI make a poster as long as it's reasonably clean, professional, and free of slop content? Indeed, much of what AI can do in those situations is little more than automation acting on PhotoShop-like sets of tools.
Make all the AI content you want, says I, but here's a tip-- don't make it obvious that AI was used in its production. Bam, problem solved.
This is whats wrong with it
Now if you could put that in to your own words...
EDIT: To the enlightened downvoters-- that person's trying to get me to read a 5,000pp science paper instead of answering themselves.
Nah you obviously dont care enough to read it so I'm good
GTFO. That's a ~5,000pp science paper. You want me to try to read and understand all that instead of you doing the simple of thing of using your own words to convey your point?
Fuck off, troll.
Nah, they're right and you're wrong here.
The thing about long papers is they generally have an abstract right upfront that'll bottom line all ~5000 pages in a few paragraphs.
Your responses make it obvious that you already know exactly what point they were making and you're being deliberately obtuse because you don't like the point being made.
Please.
That is an actual science paper in an area in which I have zero academic training and little familiarity with the relevant terminology, how that field quantifies things, and a bazillion other concerns. Sure, I might be able to spend a day's worth of my time struggling to understand the relevant point of the paper and guessing at WTF the other commenter was wanting me to mind read, but what's the point?
If you're too much of a tongue-tied freak to give a straight answer, and instead throw out some science paper and expect someone else to absorb it and guess at whatever your original point was, then.... you might just be an idiot.
And I've already answered concerns about AI energy consumption in this thread chain, so maybe instead of trying to ride a troll's lack of effort, you could actually read the thread?
Yeah, good luck with that. Later, tater.
Haha calm down man its not that deep, you can get the gist by even just skim reading it instead of writing paragraphs of text to alleged trolls
Not that I could find, including looking through your comment history in case I was missing it.
But unless your "answer" includes some peer reviewed science (like the paper you steadfastly refuse to read since you know it'll prove you wrong) it's not an answer so much as an excuse, or maybe even just a vibe.
Good on you for piping up with your alt. Got it tagged now, thanks.
Lol, I'm not anybody's alt, but it's good to know your ability to determine what's going on at the individual scale is as good as your ability to analyze current events.
Lmao the pro AI guy being too lazy to skim a paper for information is too on the nose. It's in 4.1 Median Model Energy Consumption.
The issue is that it'd take hundreds of times less energy to just put together something yourself, not to mention the massive energy cost of training the models in the first place. Plus you'd be developing a skill while doing creative work, which is great for your brain.
What "pro AI guy"?
I'm someone who's repeatedly voiced my deep distrust of an AI-centered future, and by no means trusts in data centers. My use-case is on the extreme lower end of the spectrum, so what appropriate rebuttal is there for that?
If it uses one of the online services, then it's a problem (environment, water etc).
You could somewhat make that argument for local models. But not many people are going to have any conception of that
Metrics are needed. Based on an article that Rimu shared a few days ago, I didn't get the impression that stuff like simple graphics work needed for a poster particularly added much strain.
There's also the simple fact that it's too late to be herding cats on this one. Most people who've gotten used to using AI for various tasks aren't going to give it up very easily. While I agree that it can be useful to talk up and spread the word about the destructiveness of AI costs, that energy and pressure ultimately needs to be funneled to lawmakers in order to draft and pass appropriate legislation.
It's also an issue that every person that complains about AI has absolutely no fucking idea what it is, how it works, how to differentiate different models/forms, or even why it's a special unique technology that is inherently bad. Unlike things like alcohol, X-rays, methamphetamines, gunpowder, oil, nuclear, mining, farming, etc, etc, that are only harmful when used/done wrong.
I don't understand how that could be at this point, but it would help explain the paranoia. To me, there's a huge difference between a simple, clean, low-cost use-case in which little or no trace of AI tools is detected, versus obvious AI slop media, versus having an LLM do extensive programming which winds up being filled with spaghetti code and whatever inefficiencies. The latter two involving a tonne more queries and tokens used.
It's all the same thing to a lot of people, I guess.
Because it has become a purity signal and tribal call rather than an actual coherent position. I've seen people get angry at researchers using AI to develop new drugs, and novel bacteriophages. I've seen people get angry at "AI" in word processing and image editing, claiming there's no difference between ChatGPT and the algorithms than have been used for decades.
There is such a idiotic mix-up between the horrible corruption of OpenAI, and the term "AI" itself, that morons with no understanding of anything are waving torches and grabbing pitchforks at the word itself, without any comprehension of what they're angry at.
Yeah, I guess that pretty-well covers it. Well said!
Not totally sure what you mean, there. I do use GPT a lot for complex searches, rarely anything else.
The large LLM companies really are doing some pretty shady shit. Loans on loans on loans (the bubble everyone is talking about). Shredding and destroying old books after scanning for training. Bowing down and supporting the current US administration that is rapidly descending to fascism. Funding the construction of these data super centers that are wrecking absolute havoc on the environment and people around them.
The tool is useful. Many of the companies promoting their models are fucking AWFUL. OpenAI is by far the worst offender, at the moment.
I hadn't heard that part. I thought they were mainly running on VC funding.
I've seen this discussed on Reddit, with the rebuttal being that a massive amount of books are regularly shredded & destroyed on a regular basis, and have been long before LLM's came along. Not trying to excuse anything on their part, but it's a relative drop in the bucket from what I understand. Also, it's not like we don't have projects like Guttenberg, Archive.Org, Z-Library, Annas and whatever else also doing their best to scan & preserve books.
Agree with everything else, altho I don't know enough to say which company might be worst. But point taken that they're all pretty toxic and worrisome.
100% agree
Maybe not, but it doesn't feel great when people think they can push you around like that. "Just comply with our demands, it's easy" is exactly the kind of attitude that makes it even more intolerable.
Yeah that's how a society works, so are you gonna cut yourself off and live off the land in a cabin in the woods where no one can tell you what you have to do or are you going to continue to comply with society's demands in order to reap it's benefits?
Going for a mix of strategic isolation and speaking out against smug mob rule bullies personally. The existence of rules is one thing, but when it's both about micromanaging the arguably innocuous use of information processing tools, and enforced in the most obnoxious way possible, that seems like the right hill to take a stand on.
Ok, but given the massive costs in energy and clean water are borne by all those "intolerable" people, I'm gonna go with they are the ones in the right here. Your arguments sound exactly like the kind of shit Exxon executives said about fossil fuels in the 90s.
When you have a 100% locally hosted AI powered by your private solar or wind array and cooled without pulling any water from a local water table,nyou can go back to smugly deriding the "intolerable" and "smug" people making demands of you, but as long as your slop-slinging is affecting all of us you have to hear about it as a bare minimum consequence.
I do use locally hosted AI and haven't used anything else in quite a while (in part because I agree with some of the criticisms), but yes, it uses resources. About the same as playing a videogame, I tested it with an outlet meter. As much as I do actually go out of my way to live in a way that limits my carbon footprint (relevant example: I throttle my GPU's maximum wattage to a more efficient setting), I'm not fooling myself to think that individual aggregate lifestyle choices is what it boils down to, that's misleading industry propaganda they've been pushing for decades.
That propaganda especially doesn't make sense with regards to AI, because its promise as a consumer product is not at all why corporations are funneling resources into its development, they are doing it because they believe it is something they can use for business and give them a more dominant position over society, so it would be done regardless of whether any individual chose to use it. Which isn't to say people using their services aren't contributing at all, doing so gives these companies training and surveillance data. The ethics of it is worth discussing and considering. General purpose "you breathe our air therefore your actions must all be justified by our moral standards regardless of the scale of impact" type arguments are petty and absurd and don't really make for a constructive discussion though.
It's not about AI, it's about people's reactions to AI slop.