I think you're being too literal, they mean they hate having to use it or they hate being constantly exposed to its shitty output. Obviously pretty much nobody hates, like, Markov chains.
I also hate the term AI.
And I'm not sure about the actual code either.
I had an idea for a SciFi story I wanted to write where a person's consciousness is uploaded into a computer. Now I can't even trito it without feeling gross because LLMs ruined everything for me, even AI scifi.
In fairness, uploading consciousness into a computer is a pretty old sci-fi trope so it would have been derivative even before the AI bubble. That being said, tropes are tropes for a reason, don't let shitty real life "AI" stop you from writing that story. Just don't call use the term artificial intelligence and you're good
Reminds me of one video game,
…which is a medium spoiler for the game itself
I reserve the hate (well, severe disdain and contempt, hate is personal in my book, haven't needed it for quite a while) for the C-Suites and owners, users get contempt if they're using it to think for them and a pass with some sympathy if they've found a way to use it as a tool while retaining executive function. LLMs and broader machine learning are fine, just a tool. You can use a wrench constructively or give someone a concussion, that's on you.
SamA is the exception, hate that market cornering fucker (and yes it's personal, I was going to go AM5 this year).
I hate AI. I don't hate LLMs, SLMs, generative models, etc... but the marketing campaign buzzword that's currently making hardware unattainable by average consumers and accelerating tracking, canvassing and profiling?
Absolutely hate it - with a vehement passion.
Man, I loved playing around with GPT2, I was hyped about GPT3, I paid for AI dunegon and NovelAI, I was so stoked about all of this. When ChatGPT came out, I was like "y'all are just learning about it ?" but I was so happy that this tech I loved was getting recognition.
I'm definitely in the "AI hate" camp now, this really is the worst timeline
I've worked with AI, off and on, for over 20 years now. The thing is, I would argue that it's always been the same. The point of AI has always been to take the place of humans. It has always been a scary line of research with profound implications for the future.
The biggest difference now, I would argue is the accessibility. It used to be that only academics and experts could use it, but now that it uses a natural language interface, any dumb assholes and any bad actors can easily use it. And they do. In droves.
The fuck are all these comments? AI is shit, fuck AI. It fuels billionaires, destroys the environment, kills critical thinking, confidently tells you to off yourself, praises Hitler, advocates for glue as a pizza topping. This tech is a war on artists and free thought and needs to be destroyed. Stop normalizing, stop using it.
Separate LLMs and AI.
LLMs are shit, fuck LLMs. They fuel billionaires, destroy the environment, kill critical thinking, confidently tell you to off yourself, praise Hitler, advocate for glue as a pizza topping. This tech is a war on artists and free thought and needs to be destroyed. Stop normalizing, stop using it.
And AI is a pipe dream no one is close to fulfilling, won’t be realized by feeding LLMs all of the data in existence, and billionaires are destroying our economy in their pursuit of it.
You are referring to AGI not AI.
The broad category of AI is most definitely real.
Doesn't work that way unfortunately. Ask a person on the street what AI is and theyll tell you whatever flavor slop generator they're familiar with. You're not going to see much pushback on ML around the Fediverse.
On the fediverse I think we can be more precise in our language.
My opinion. AI is a way to improve a computer models accuracy over time based on new data.
I could even argue that ChatGPT etc. are not AI because the LLMs are not directly learning from the inputs they are receiving.
yes they are!! do you know what the "T" means?? trained (edit: [loudly incorrect buzzer])!! over time!! from data!!
if you really want to be pedantic, chatgpt is ai!! there's rlhf, yaknow??
Change this out for any other technology that's been innovated throughout human history. The printing press semiconductors the internet.
The anti-ai rhetoric on this platform is becoming nonsensical.
At this point it's just bandwagon hate. These people don't even understand the difference between llms and AIs and the various applications that they have.
Blockchains in an age of Trump choosing a new Fed chair after trying to have Powell arrested.
Trust your government over software and cryptography, which has no basis in reality outside of the laws of physics and mathematics.
Yes. These all qualify. They're all massively successful technologies.
Well, aside from 3D TVs and smart glasses. But they're generally innocuous. Yes I also understand that smart glasses es have privacy issues but then again in this day and age what doesn't.
Bitcoin and Ethereum PoW used resources and raised (GPU and electricity) prices for everyone.
They literally killed and excommunicated people after the invention of the printing press for producing unauthorized copies of the Bible. Figures like William Tyndale paid with their lives for translating scripture into English, challenging the Church’s authority.
There is illicit material circulating freely on Tor, demonstrating that technology can distribute both knowledge and criminal content.
Semiconductors underpin some of humanity’s most powerful and destructive technologies, from advanced military systems to cyberweapons. They are a neutral tool, but their applications have reshaped warfare and global power dynamics.
You are fully entitled to dislike AI or technologies associated with it. But to dismiss it entirely is ignorant. Whether you want to believe it or not, we are on the precipice of a technological revolution, the shape of which remains uncertain, but its impact will be undeniable.
Ok. This is clearly rage bait.
You're an ignorant fool and I'm probably not the first person to tell you that.
The internet printing press firearms semiconductors nfts the blockchain.
A multitude of other technologies.
I find AI to be more reliable every day. Fail to see an issue of killing critical thinking. Also my experience, search engines are flooded with advertisements and garbage unrelated to my search. Can only hope the business world does not “Shittify AI in the same way.
70 years ago, it was predicted pay-television would replace advertisements. Instead television evolved to a fee based system and a higher ratio of ads. So you can bet a good thing will evolve in the same way.
Same with the internet. Fuels billionaires, destroys the environment with data centers and cables, kills libraries and textbook research, spreads nazi propaganda. We need to stop using technology in general.
Lets see a standard problem I'm randomly making up using a free AI, you tell me if this kind of thing can be useful to someone:
If I have a bucket that is 1 meter tall and 1 meter wide how much volume can it hold?
The volume V of a cylinder can be calculated using the formula:
V=πr2h
Where:
r is the radius, h is the height.
In this case, the bucket is 1 meter tall and 1 meter wide, which means the diameter is 1 meter. Therefore, the radius r is:
r=21 meter=0.5 meters
Now substituting the values into the volume formula:
V=π(0.5m)2(1m) V=π(0.25m2)(1m) V≈0.7854m3
Thus, the volume the bucket can hold is approximately 0.785 cubic meters.
Using llms for math questions is probably the worst usage for llms.
And all of this is easily calculated without ai. You can literally google it and let google do the math for you without ai.
Found the Mennonite.
The fuck are all these comments? The internet is shit, fuck the internet. It fuels billionaires, destroys the environment, kills critical thinking, confidently tells you to off yourself, praises Hitler, advocates for glue as a pizza topping. This tech is a war on artists and free thought and needs to be destroyed. Stop normalizing it. Stop using it.
Which ai and for which use? It’s a tool. It’s like getting mad cause a guy invented a hammer. It’s not the tool hurting you dude, it’s the people wielding it.
If that hammer also had massive environmental impacts and hammers were pushed into every aspect of your life while also stealing massive amounts of copyrighted data, sure. It's very useful for problems that can be easily verified, but the only reason it's good at those is from the massive amount of stolen data.
All those things you said are vague and nebulous and every day people are not gonna understand that message and will just think you’re hysterical or a conspiracy guy. The way the message is put forwards is super important
If an everyday person asked me why I don't like ai I would show them those reasons in more depth, but on lemmy most people have seen the articles about ai water use and the light/noise/water pollution of data centers.
Arguably, hammers also have a massive impact on the environment. They are also part of everyday life. Building you live in? Built using a hammer. New sidewalk? Old one came out with an automatic hammer. Car? Bet there was a type of hammer used during assembly. You can't escape the hammer. Stop running. Accept your inner hammer. Embrace the hammer, become the hammer. Hammer on.
Yup. you can use a chisel or even just a hammer, you just need the right person with it
they have somekind of plan, or maybe its all sunken cost scenario. Either way, they think they can get some benefit from it and they are so determined they are throwing insane amount of money in it even though there is no clear way to get any profit from it. So either they know something we dont or they are desperate to save their investments -> worse ai does, better its for all of us since once ai crashes the components stop being wasted on it, less electricity and materials are wasted on datacenters and best of all, all those fucking billionaires lose a lot of money they have invested or at least the investors who thought it good idea to support them lose and maybe dont do it again.
It baits investors into giving them money, mainly.
t’s the same as any other commercial tool. As long as it’s profitable the owner will continue to sell it, and users who are willing to pay will buy it. You use commercial tools every day that are harmful to the environment. Do you drive? Heat your home? Buy meat, dairy or animal products?
I honestly don’t know where this hatred for AI comes from; it feels like a trend that people jump onto because they want to be included in something.
AI is used in radiology to identify birth defects, cancer signs and heart problems. You’re acting like its only use-case is artwork, which isn’t true. You’re welcome to your opinion but you’re welcome to consider other perspectives as well. Ciao!
The use in radiology is not a good thing. Hospitals are cutting trained technicians and making the few they keep double check more images per day as a backup for AI. If they were just using it as an aide and the humans were still analyzing the same number of picturea that would be fine but capitalism sees a way to save a buck and people will die as a result.
This isn't a problem with AI though, it's a problem with the people cutting trained technicians. In places where such incompetent people don't decide that, you get the same number of trained technicians accepting (and being a part of) a change that gives them slightly more accurate findings, resulting in lives being saved overall. Which is typically what health workers want to begin with.
OK so why is AI so big right now because it isn't profitable. Even there most expensive tier is losing them money. Then you have the data centers getting breaks on electricity so the rest of us cost goes up to make the difference. Where is this magical profitability that is driving AI.
It's in part because people aren't open to contradictions in their world view. In part you can't blame them for that since everyone has their own valid perspective. But staying willingly ignorant of positives and gray areas is a valid criticism. And sadly there are plenty of influencers peddling a black-white mindset on AI, ignoring all other uses. Not saying intentionally or not, again perspective. I'm sure online content creation has to contend with a lot more AI content compared to the norm. But only on the internet do I encounter rabidly anti AI people, in real life basically nobody cares. Some use it, some don't, most do so responsibly as a tool. And I work in the creative industry...
"I've never seen it it must not exist"
I work in a creative industry too and it is the bane of not only my group but every other company I've spoken to. Every artist and musician I know hates it too.
I never said it doesn't exist. I'm sorry people in your area are being negatively affected if so. But the point still stands. My experience is just as valid.
I'm pretty anti AI as it is a tool of the billionaire class to enslave the masses. Look up TESCREAL, its the digital eugenics billionaires and fringe philosiphers believe in and it is the driving force in the AI push.
That being said I can see a use for a focused, local LLM/AI assistant. I have to search a lot of confidential technical manuals, schematics and trust cases in my job. We are thinking about testing out Ollama to upload all our documents too to make searching them easier.
You are the exact person I didn't mean 😄 the first is a very valid reason to dislike AI.
Even before our current time, "nobody cares" is not a thermostat reading of what "really matters". It almost sounds like you believe people know what's best for themselves, when the truth of the matter is that humanity has long proved otherwise.
You sound like a cartoon supervillain, Lex Luthor ranting to superman about the common animals not knowing what's best for themselves.
I don't believe that. What I'm saying is that these are all people I work with look very critically and skeptically at the world, as that's pretty much an inherent requirement for the creative industry. We all know what AI is and what it does, and most arguments against it hold no water to people with a realistic view of the industry to the point it simply cannot be black and white like some claim it to be.
There are a few good reasons to dislike AI, but those don't apply to all of AI. Some are value based, and other people have other values that are equally valid. And some can be avoided entirely. Like how you could ship packages with a coal rocket instead of a train on electricity, or just shipping less packages to begin with.
There is trust and experience between one another in the industry that we aren't using it unnecessarily, wastefully, and incorrectly, and AI is not anywhere near a requirement by consumers nor healthy minded businesses.
I fully agree. I still remember the time when using Photoshop was seen by some as not being "real artist", because "any idiot with a mouse can draw now". I'm not under any illusion this will last forever, the negative sentiment is boiling because of the bubble and it's negative externalities, not by the technology itself. So once that bursts, things will hopefully be a lot more peaceful.
k
Some people I work with when they do an internet search they go straight to the ai info. Literally read out the first bit of what's there and take that as the truth.
Great way to make yourself immediately untrustworthy
McKinsey isn't exactly stupid, its amorality run amok and a culture of cutthroats.
If you've ever worked with consultants or managers in general, like 50-75% of them are fucking stupid. Just because they can convince other idiots that they're not, doesn't mean they aren't. I've watched the blind lead the blind into financial ruin, while getting paid big bucks to do it.
I think it's a bit of both. Sometimes the people hiring them are truly clueless. The kinds of reports that management consultants make seem really well thought out and intelligent. Other times, upper management wants to make a big decision, and they think it's the right one, but they need something to show they considered all the alternatives and that an outside source agrees with them.
Also, management consultants are very stupid, but they're clever in a very narrow area. That's why they succeed with upper management, because like LLMs, upper managers think they're clever.
As a former consultant and manager, I wholeheartedly agree, but your % is too low. The culture of consulting is poison and makes monsters out of people.
Here's a great video about what they did to disneyland https://youtu.be/Q7pgDmR-pWg
And then everyone clapped.
Tears in their eyes
People around me use AI all the time to get answers to generalized topics. More and more they use it like a search engine / information augmentation system.
They are not technical people. They mostly know that the information needs to be double checked and might be wrong. But usually take it at face value if the importance is low.
Honestly this is about what they did before. They would search Google, click on the first blog, skim it, and repeat until getting some answer they believe.
I too use AI regularly for brainstorming, quickly summarizing massive text messages, and reformatting text from a jumbled mess into something more cohesive, etc.
I don't love it or hate it. In some cases it saves a lot of time and is useful tool. In other cases it outputs trash that we cannot use for any serious case.
Just like a hammer or a shovel, it's a tool. Can be used the right way and it can be used the wrong way.
I think of an LLM as extraordinarily lossy compression. All the training data is essentially encoded in the model. You can get an approximation of the data back out again with the right input.
I don't think it's any less reliable that random blogs on the web, and I don't have to wade through SEO tripe either.
The annoying thing though is that all the random blogs on the web are written with using these LLMs now. It makes it much harder to be critical of your sources, because they're all coming from a unnamed, proprietary LLM with no information about who owns it or the training data. At least before, I could look up the user or check out their other articles, now every article is randomly generated from some unknown prompt.
I would argue this isn't only a bad thing though. Even before AI, many bogus articles and information existed. Eg. that people swallow spiders in their sleep, which many outlets parroted.
I would guess most people never checked (m)any sources on most information they found so long as the 'vibe' felt trustworthy. There is no cure to make reality simple, and the more pressure we have to teach people to think critically, the better.
- AI is much better at creating internet spam.
- AI is a vector for even reputable places to "set and forget" any article they're in charge of. Any mistruths are simply 'glitches'.
- The pressure on people to think critically only matters if people actually start thinking critically. Kids use this technology to skip their homework.
That's what makes them shitty though.
When I have a hard technical problem I often search for and read through a dozen different sources. Many of them are wrong, or are right but not covering exactly the situation I'm looking at. Eventually I'll find one that's either right and answers my problem, or gives me the clue I need so I can figure out the solution for myself.
If I ask an LLM to solve the problem, it will make up an answer that would seamlessly blend in with all its training data. In other words, it's most likely to produce something that's wrong, or something that's right but not for my particular case, or something that's close but incomplete. That's effectively useless. At worst it blends in with its training data enough to convince me it's right, while not actually being right. At best it's something that is close enough to give me the clue I need. Most of the time it's going to be something that's wrong and I know it's wrong because if it were that simple I wouldn't have had to resort to the AI bullshit generator.
It can be helpful for quickly summarizing a vast body of knowledge or a highly complex topic, to get a general overview and see which strings to pull further, as long as you don't take everything at face value and understand that you still need to pull those strings yourself in order to acquire an understanding.
Like, if I suddenly wanted to learn computer programming, I wouldn't know where to start. But querying an LLM can give me a general idea, define a few key terms and explain the difference between related concepts, without me having to browse through a hundred different tech blogs to answer all my questions in terms I can understand.
But I wouldn't suddenly think I'm a computer programmer after doing that. I would have a better idea of where to start learning. I would be able to decide whether to focus first on object-oriented programming or functional programming, static or dynamic typing, declarative or imperative syntax, etc., instead of getting overwhelmed from the start just trying to learn the differences between those concepts.
It can also suggest resources for further learning, books or websites written by humans, links to open-source software that does what I'm trying to do, etc.
I wouldn't expect it to write code for me, but it can be an efficient aid to self-learning and show me what programs and libraries to use for my intended purpose.
Or for astrophysics, for example. I wouldn't expect it to give me an accurate breakdown of the engineering specs required to build a pair of O'Reilly cylinders at a Lagrange point, but it can suggest software for rendering prototypes or for simulating the forces that need to be accounted for.
That wouldn't make me an astrophysicist, but it's kind of cool that you don't need to be one to learn about this stuff and tinker around in a field that's so vast and technical as to be otherwise prohibitive for non-experts.
It also depends on the LLM of course. I think Mistral and Lumo are generally pretty okay at doing what I described above. Their algorithms aren't corrupted by american venture capital, at least, so they have more incentive to give you an accurate response rather than being sycophantic and hugboxing.
I asked ChatGPT to review my resume and make changes tailored for the job description I was applying to (which I also gave it). Also told it that this was an internal position and not really an upgrade, but a sidestep that (I felt at the time) was more aligned with my long-term career goals.
I was really happy with the improvements it suggested.
Didn't get the job, but as I understand it, the hiring manager wanted to bring in a friend of his from the moment he posted it.
In retrospect, I'm kinda glad that's how it panned out. Myself and the new guy are operationally equal, he's incredibly competent. We compliment each other well, and get along great.
And he's friends with the big boss and thus has his ear.
The main reason I even applied to the job was because I wouldn't want to work with anyone but myself in that job. And he's close enough. .
I did get an interview for it, which ultimately just became a 1:1 with the boss and it gave us a chance to talk openly about where I see deficiencies that need fixing. All in all it went great. Six or so months later and I'm feeling a renaissance in the air at work. Like the things I talked about with him are now front-and-center and getting the attention they needed.
This company moves quite slowly, so six months (basically, new fiscal) seems incredibly quick.
I'm lazy as fuck. I want to solve problems in the easiest way humanly possible. With the least amount of effort output.
What about you? Do you take the hard way?
I'll be real with you, I typed lazy but wanted to type idiot. Read your fucking emails Jesus Christ. You still have to check all the shit generative AI writes because it lies constantly. It's very nature does not understand what it's generating.
Hard to tell if you're trolling or trying to add value to the conversation and just missing it.
A hammer doesn't know what it is building but it is still useful.
This is the nature of tools: for some they improve output, for some they don't.
Obviously, don't rely on them to read important emails for you. But so many things don't need additional checking. We've all done at least a decade of schooling. We all know basic math, science, and history. When we forget things, all it takes is a small reminder to get it back. Our brains are capable of recognizing whether we've seen something before or not. We're also capable of reasoning to determine whether something we read is consistent with everything else we know.
So many other things are also so unimportant that it doesn't matter at all if you're wrong. For example, some actor looks familiar, it lies to you about what film they were in, and you believe it. Is your life any worse off for it?
The fact is though the average person is starting to replace their search engine with chatgpt, gemini, grok or whatever other llm and I have seen more and more small association using generative ai to make their posters instead of working with artist or doing it themselves.
Is this because LLMs are getting better, or because search engines are getting worse?
Because they are definitely getting worse. I get redirected to a brand new slopsite daily.
Search engines have peaked in early 2010s and hav been deteriorating ever since, becoming virtually unusable since ~2020.
Seriously, google has become unusabe without adding "site:reddit.com" to almost every search. I would like to see something like Perplexity be compared to a proper search engine - if it existed.
everything peaked in the 2010s ;)
I agree with the concern for Reddit in responses. Many times include the shit posts.
It's because people thinks AI is like those in the movies (cause it's been advertised so, too), omniscient and infallible. A short while ago I overheard a "imagine, even the AI didn't know it!" which normally would be "Your search didn't return any results".
AI slop has just accelerated the downfall of search engines. Attention based economy, advertising and SEO are the reason you can't find anything useful anymore. The Internet itself is broken and even if there was a good search engine it would struggle to not suffocate in the seo crap out there.
At this point, I am willing to pay a subscription for a search engine if it's ad-free and shows me what I'm looking for.
Kagi, searxng, startpage, ddg if lazy..
Kagi is great but paid and some people have feeling about then using yandex.
There are so many solutions. People are lazy and keep using trash Google that doesnt even work. People ask me how in the world I find things. Because I am literate and I know how to use a computer. Its not hard guys.
I never said llm's or generative ai is good. I was talking about post just being wishfull thinking.
A good blogpost on this: The Enclosure feedback loop
When everybody uses AI to search, it becomes a closed system that holds all info. Doesn't need to be productive, but it gatekeeps the knowledge that was free on the internet. It's a self-reinforcing loop.
I work in infrastructure and what's concerning is that younger guys are skipping learning to script to automate processes and instead just getting slop from LLMs that they have no idea what it's doing.
Some have also relegated learning problem solving to it as well so when things go wrong, they're clueless without it.
Google started making their search engine worse and always pushing things that they thought would make them money. It's not surprising people are trying something else.
“Enshittified”; a new descriptor for my vocabulary. 😀
I think the research can be pretty cool. Every implementation has been kinda horrible.
The research/tinkerer community overwhelmingly agrees. They were making fun of Tech Bros before chatbots blew up.
I have made the conscious decision to try and not refer to it as AI, but predictive LLM or generative mimic models, to better reflect what they are. If we all manage to change our vernacular, perhaps we can make them silgtly less attractive to use for everything. Some might even feel less inclined to brag about using them for all their work.
Other options might be unethical guessing machines, deceptive echo models, or the classic from Wh40k Abominable Intelligence.
I mostly agree. Machine Learning is AI, and LLMs are trained with a specific form of Machine Learning. It would be more accurate to say LLMs are created with AI, but themselves are just a static predictive model.
And people also need to realize that "AI" doesn't mean sentient or conscious. It's just a really complex computer algorithm. Even AGI won't be sentient, it would only mimic sentiency.
And LLMs will never evolve into AGI, any more than the Broca's and Wernicke's areas can be adapted to replace the prefrontal cortex, the cingulate gyrus, or the vagus nerve.
Tangent on the nature of consciousness:
The nature of consciousness is philosophically contentious, but science doesn't really have any answers there either. The "Best Guess™" is that consciousness is an emergent property of neural activity, but unfortunately that leads to the delusion that "If we can just program enough bits into an algorithm, it will become conscious." And venture capitalists are milking that assumption for all it's worth.
The human brain isn't merely electrical though, it's electrochemical. It's pretty foolish to write off the entire chemical aspect of the brain's physiology and just assume that the electrical impulses are all that matter. The fact is, we don't know what's responsible for the property of consciousness. We don't even know why humans are conscious rather than just being mindless automatons encased in meat.
Yes, the brain can detect light and color, temperature and pressure, pleasure and pain, proprioception, sound vibrations, aromatic volatile gasses and particles, chemical signals perceived as tastes, other chemical signals perceived as emotions, etc.. But why do we perceive what the brain detects? Why is there even an us to perceive it? That's unanswerable.
Furthermore, where are "we" even located? In the brainstem? The frontal cortex? The corpus callosum? The amygdala or hippocampus? The pineal or pituitary gland? The occipital, parietal, or temporal lobe? Are "we" distributed throughout the whole system? If so, does that include the spinal cord and peripheral nervous system?
Where is the center of the "self" responsible for the perception of "selfhood" and "self-awareness"?
Until science can answer that, there is no path to artificial sentiency, and the closest approximation we have to an explanation for our own sentiency is simply Cogito Ergo Sum: I only know that I am sentient, because if I wasn't then I wouldn't be able to question my own sentiency and be aware of the fact that I am questioning it.
Why digital circuits will never be conscious:
The human brain has about 14 billion neurons. The average commercial API-based LLM already has about 150 billion parameters, and with FP32 architecture that's already 4 bytes per parameter. If all it takes is a complex enough system of digits, it would have already worked.
It's just as likely that consciousness doesn't emerge from electrochemical interactions, but is an inherent property of them. If every electron was conscious of its whirring around, we wouldn't know the difference. Perhaps when enough of them are concerted together in a common effort, their simple form of consciousness "pools together" to form a more complex, unitary consciousness just like drops of water in a bucket form one pool of water. But that's just pure speculation. And so is emergent consciousness theory. The difference is that consciousness as a property rather than an effect would explain why it seems to emerge from complex enough systems.
It's just a really complex computer algorithm
Not particularly complex. An LLM is:
$P_\theta(x) = \prod_t \text{softmax}(f_\theta(x_{<t}))$
where $f$ is a deep Transformer trained by maximum likelihood.
That "deep Transformer trained by maximum likelihood" is the complex part.
Billions of parameters in a tensor field distributed over a dozen or more layers, each layer divided by hidden sizes, and multiple attention heads per hidden size. Every parameter's weight is algorithmically adjusted during training. For every query a matrix multiplication is done on multiple vectors to approximate the relevancy between each token. Possibly tens of thousands of tokens being stored in cached memory at a time, each one being analyzed relative to each other.
And for standard architecture, each parameter requires four bytes of memory. Even 8-bit quantization requires one byte per parameter. That's 12-24 GB RAM for a model considered small, in the most efficient format that's still even remotely practical.
Deep transformers are not simple systems, if they were then it wouldn't take such an enormous amount of resources to fully train them.
The technical implementation, computational effort and sheer volume of training data is astounding.
But that doesn't change the fact that the algorithm is pretty simple. Deepseek is about 1,400 lines of code across 5 .py files
You're really breaking the shitting on AI vibe when you make it sound like the height of human capacity and ingenuity. Can I just call it slop and go back to eating glue?
"Asking one's chat bot" sounds so much less impressive than "leveraging AI". Using the right language throws some cold water on the corporate narrative.
The men of iron are so freaking cool! They're still around in modern 40k hiding biding their time.
Maybe one day we'll have a whole new army of AIs in 40k!
And then everyone clapped... Blah, blah, blah.
And, I do not like AI.
That was December 2024.
McKinsey & Company consulting firm has agreed to pay $650 million to settle a federal investigation into its work to help opioids manufacturer Purdue Pharma boost the sales of the highly addictive drug OxyContin, according to court papers filed in Virginia on Friday.
Drug dealer must sell drugs.
It's a meme, but at least in my childhood it did occasionally happen, the "whole class clapping" thing. One time I did really good in a second year Spanish class end of term activity. A girl proposed on the spot, and the whole class clapped. Intensely embarrassing but, as far as I can recall, it was not even sarcastic.
And everyone clapped
Obama was there, he awarded the medal of honor, my parents were proud of me, the AI chump was instantly killed
This absolutely did not happen.
It's the same as the crypto-blockchain-NFT bullshit. A bunch of idiots with too much money put down on it, then when it doesn't become the hit they expect they start with the propaganda about how it's the greatest thing, and then when THAT fails they just take away other choices or try to cram it into everything anyhow
The problem is that the propaganda is working. Despite what this meme implies, many many people do use and like AI chat bots and in my line of work, I am asked nearly daily which AI is the best to use and how users can have their own AI that answers emails or mocks up ideas or how it can make their daily job easier. I'm the wrong person to ask that to but I understand why they're asking me. I'm their IT guy. I don't particularly care if you use AI in your job because my job is just to make sure your computer keeps working.
Personally I don't hate it but I just think there's really no urgent need for it. If they're using it to take jobs away from people well what is everyone going to do for work? To the billionaires think there's going to be a gigantic human die off and they're going to be elite class of 100,000 people and they'll be served by intelligent robots? If that's their angle good luck.
They all grew up on classic sci fi. And, being dumbasses, they took Solaria as an aspiration rather than a warning.
I don't hate AI I like it for a lot to things. It is especially awesome at writing 1 off scripts/small projects or allowing me to test multiple different versions of a programming solution to see what one I like most. I use it to search the web and search my own knowledge database. Its an awesome tool. I get why people would hate it having it shoved down their throats all the time but that's not a problem I ever have to deal with cos almost everything I run is a full foss stack.
It’s so shitty it ruins everything
I actually like AI very well. Get better detailed answers and still can zero in on specifics with options. If I ask three people, I get three totally different answers.
The issue I (and many) have with the technology isn't the thing itself, but the all-consuming drive to wedge it in everywhere.
It's a tool. Use it where it may be appropriate, and don't demand others use it before they deem it time.
I do not hate AI, I am learning how to use it. AI is really great technology and like a calculator, you can be more productive when you know how to use it.
It’s a shiny new tool, humans wanna use it on everything. Hopefully we find a balance. There’s a lot Ai can do that a human literally cannot, background work and such, I’ve seen it implemented in great ways. The problem isnt the tool, it’s the corporate dirtbags who want to maximize profits out of it
They haven't even figured out how out make profit off of it, let alone maximize it. I really don't understand the current approach. "Let's slam this unfathomably expensive technology into everything and not turn a profit. Oh, also, it's destroying the environment and making electronics and power really expensive"
People need to explain in clear detail to others the environmental impact of ai and how it causes that impact. You’re telling people that making a funny photo is fucking up the environment, you have to be more clear. Im not a dummy but the idea that ai is damaging the actual environment is vague and nebulous. It’s never been explained or shown to people in large way how it’s actually happening, so it’s not real to people because it’s not widely explained
The thing is that image generators aren't even such big environmental hazards. They are often comparable to other activities like gaming, since you can often run them locally. At scale that's also a bad thing to have, but we don't get people up in arms about many equivalent things.
It's the biggest LLMs that are extraordinarily inefficient, since you need entire compute clusters to get those running. There's just not really a way to put every AI tech into the same box except by not caring about the truth, so it makes the people that don't draw that nuance look very scummy to people aware of that nuance.
The problem is that even the open source models (and they're not fully open source anyway) aren't exactly trained ethically. They're still trained on stolen data, they are still consuming gargantuan quantities of water and using the same amount of power as a small city.
They are in no way harmless. The only thing that's better about them is that they don't steal your private data. The environmental issues are still there though.
The pictures you got for your story are a result of plagiarism.
I'm not actually making a value judgement. If they make you happy, that's great and im happy for you.
Kinda like how diamond engagement rings make people happy. You gotta accept how that sausage was made
Just being contrarian but everything creative is plagiarism.
It’s not, though. That’s just something hacks plagiarists say.
Edit: damn, this got ‘em coming out of the wood work. If you feel the need to argue against this, you’re not an artist, you’re just a plagiarist and a capitalist boot licker.
But it is though. Chappell Roan is just a derivative of Kate Bush. AI is not copying it's using what it knows as a template. No different than you or I. To think otherwise is to have your head up your ass.
Now we can argue if we should be allowing AI to do this and I believe no we shouldn't as it detracts from people being creative.
But that's not the argument you were responding to.
The use cases don't cover the costs.
What do those pictures add to the story? Do you believe that because you're privileged enough to run it locally, that makes it ethical to use for you specifically, but not others that rely on datacenters?
Have you ever been to a movie theater? What did the pictures add to the story?
An appreciation for the artwork, craftsmanship, writing, and storytelling that means something because it’s a representation of the time, effort, dedication, art, and the human endeavor and spirit that went into it.
The same is true of all artwork. Some is good quality and takes months or years to complete, some is a scribble that took a few seconds. They're not different, friendo.
It's the same as any other commercial tool. As long as it's profitable the owner will continue to sell it, and users who are willing to pay will buy it. You use commercial tools every day that are harmful to the environment. Do you drive? Heat your home? Buy meat, dairy or animal products?
I honestly don't know where this hatred for AI comes from; it feels like a trend that people jump onto because they want to be included in something.
AI is used in radiology to identify birth defects, cancer signs and heart problems. You're acting like its only use-case is artwork, which isn't true. You're welcome to your opinion but you're welcome to consider other perspectives as well. Ciao!
I mean, machine learning is great. I was replying to someone talking about using generative ai to make photos, not a doctor or a data scientist, to my knowledge.
I kinda love that response though, I'm going to save it. You can tell it's not even from a chatbox, it's just kinda too unhinged. Like 'oh you are curious how I view the ethics of my actions? You must hate cancer survivors.' I think ai bros do like to think of themselves as pushing the bounds of science, they don't see themselves as consumers.
I only used Venice.ai because it stores everything in my browser (which I can erase immediately). It is just fun to get things off my chest and use a rubber ducky method for any issue I might be having, but it has been useless otherwise.
Nope.
Almost as if we don't have as much choice as we think
This needs an NSFW tag my man
That's... not how NSFW tags (or rather spoiler tags) work... You got to actually hide it so people that don't want to see NSFW don't have to see it. I can confidently say I saw the content of your image before I could even read the NSFW in your text, maybe because it's a huge 512x512 image.
NSFW
NSFW content goes here
Nope. I've been using it for preliminary writing editing. It's not creating anything, just giving advice on how to make it clearer.
Huh?
If your analysis of whether AI is good or bad is simply "I use it and I like it", then you are a child, or an adult with the mental capabilities of a child.
I'm happy life is so simple for you though!
I am using a tool that's available to me. I'm not going to not use it just because some people use it wrong. What the fuck?
And if it wasn't, I never would have discovered a new skill I had no clue I could do before, so I'm going to insist that you lick my nutsack. How's that for maturity?
And so, you have surrendered your voice to the machine. You are now something less than human.
It's not writing anything for me. It's suggesting where I'm not making things clear, or where a quicker pace or punchier phrasing might help, awkward prose, inconsistent character voice, etc. And something to bounce ideas off of, like "does this make sense?"
And I can respond, and it adjusts, and helps me get to a baseline standard before a human looks at it--who I sent a draft to a few days ago. It's 100% my own ideas, words, scene staging, and story. I still rewrote entire sections even though it said they were solid.
I don't know a lot of writers, and getting friends and family to read 5 pages, let alone 2,000 words, or 13 chapters is near impossible. Critique meetups and such are only so helpful.
And I've never done this before, yet I've actually created something, which ChatGPT made a little easier by acting as a mostly competent editor with mostly mediocre creative instincts, anyway. Sometimes it's nonsense, or forgets character traits, but it's more helpful than zero support.
6 months ago I had no clue I was capable of anything like this, but now I'm doing it, and frankly it's a pretty original story. Would you not consider that a positive use?
But by doing so, you've surrendered your voice. You say it makes things clearer, but sometimes ambiguity is good. You say it makes quicker and punchier phrasing, but some writing is best done low and slow. LLMs are by their nature the least common denominator, all the color of the creative world melted down and blended into a uniform grey. By relying on the LLM to alter your writing style, you're making your writing more bland, generic, and indistinguishable from everyone else. You're giving up what makes your writing you're writing. You're just another hand for the machine.
I do not consider it a positive use, as I would rather read imperfect human writing than grammatically perfect machine drivel. Imperfections are not a reason to enjoy human-created works. They're the only thing that makes the works worth appreciating.
I'm writing fiction. Sometimes parts benefit from slower or quick pacing. Sometimes you write something that makes no sense.
Not OP, but I'm glad you're getting empowered. That's imo the best use for it. It's crazy to me how many people write this off, not understanding many aspiring creatives need these kinds of stepping stones to stay motivated. Because logically, if what you make takes off and becomes popular, at some point human employees are probably the better option than AI. And as you said, without AI you might have never taken the first leap. So it would end up creating more livelihood for creatives than taking away.
I appreciate that, thank you. Just wanted to suggest one way it can be not totally evil, and I am a admittedly a bit grateful for it as a handy assistant--like a Jarvis with minor brain damage.
ALL the cool kids hate AI. You want to be cool, don't you? Come on, everyone, let's hate the latest marvel of technological innovation together!
It uses electricity, which nothing else does. It uses water, which is then destroyed forever and can never be used again after it's been used once for AI. It's occasionally wrong, which nothing else ever is. It makes billionaires richer, which nothing else ever does. It reduces your creativity, which everyone has in spades. It discourages you from thinking for yourself, unlike the mob mentality telling you to hate it.
Stop acknowledging the many positives and just accept instead that it's only ever going to be horrible. Forever. Always in all ways. It's time to turn off your brains and just hate AI with us. Woo! 🥳
Did you write this or just prompt it?
and then everyone clapped
ChatGPT alone has nearly a billion daily active users. Even accounting for corporate types who are pressured to use it, saying that nobody likes it or wants it is delusional.
The closest figure I could find said that chat gpt sees around 800 million users per week, not per day. The per month statistics at the height of January 2026 was 5.72 billion visits. Which means about 185 millionish (rounding up) per day. The statistics often go from visits to users. I imagine users means unique users and visits can mean the amount of times anyone has accessed the website. Regardless all of this would mean a majority does not use chat gpt
And what's going to happen when the VC funding dries up and they have to charge what it actually costs to run?
First step will be that they'll slap ads into it. Afaik stuff like that is already planned.
Regardless, it also means hundreds of millions of people use it and other AI tools on a daily basis. They're not all being forced to at gunpoint. Trying to pretend that the opinions of one's own social circle are "everyone" and that people who disagree do not exist is not a persuasive argument for anything.
I don't think it's "regardless" at all.
My point was that saying that "everyone hates this" and "nobody wants this" (what the shirt in the post says) is flatly untrue. Whether ChatGPT has a billion daily active users or 500 million weekly active users or whatever is just trivia that doesn't really change the gist of what I'm saying. Lots and lots and lots of people really like AI tools and use them every single day. Ignoring that fact and pretending like everyone agrees with you is dumb.
They’re not all being forced to at gunpoint.
No, but a lot of them are being ordered to do so by their employer.
If it's something like Google or Windows that's suddenly started generating AI "answers" or whatever when you just use your computer as normal, does that get counted in those statistics?
I bet it does, and I bet it accounts for a huge percentage of it.
[...] has nearly a billion daily active users.
So has Facebook, or opioids.
Your point is invalid.
Are you trying to claim that nobody likes Facebook? That nobody like opioids? You can say that they're harmful, but that's not the argument OP was making. His point still stands.
Using it specifically is, at least for me, something different than having it included in everything else. It's everywhere, no matter if it actually offers real benefits to the user and that's the issue.
For about 3 weeks I wasnt able to share documents with coworkers without sharing them with Copilot first..
Using it by choice when you specifically ask to is one thing.
When every action you take on a system is visibly being fed to the bots by default it gets intrusive and creepy.
The OP is about AI getting forced into things, which rightfully many people are pissed off about. But you're right, ChatGPT and many other LLM tools are very popular.
Yep. Fediverse is in a bubble. People in general have no feelings about it. They don't love it or hate it, they just use it. They have joked about how it gets stuff wrong until like a year ago and that's an old joke now.
A lot of people here have passionate hatred about it and they project that it someone doesn't hate it as much, they must love it. But the largest majority have no feelings towards it. It's just a tool, useful for some things, not as useful for others, that's it.
You are a prime example for a simple truth: circlejerking ("opinion bubbling") doesn't depend on platform.
There are a shitload of users. Yet, "everybody"...
Perhaps in their social circle, AI is the new porn. Who ever would watch that, after all?
But, more likely, everyone clapped and that was it....
ChatGPT alone has nearly a billion daily active users.
According to whom?
The people who run it? Who have every financial incentive in the world to inflate their numbers?
Even they aren't that brazen.
It's a "trust me bro" number that the commenter can later dismiss with "well it wasn't precise, but you get my point, and if you don't it's you who's boneheaded, not me"
@lemmy.world
A place to share screenshots of Microblog posts, whether from Mastodon, tumblr, Twitter X, KBin, Threads or elsewhere.
Created as an evolution of White People Twitter and other tweet-capture subreddits.
RULES:
- Your post must be a screen capture of a microblog-type post that includes the UI of the site it came from, preferably also including the avatar and username of the original poster. Including relevant comments made to the original post is encouraged.
- Your post, included comments, or your title/comment should include some kind of commentary or remark on the subject of the screen capture. Your title must include at least one word relevant to your post.
- You are encouraged to provide a link back to the source of your screen capture in the body of your post.
- Current politics and news are allowed, but discouraged. There MUST be some kind of human commentary/reaction included (either by the original poster or you). Just news articles or headlines will be deleted.
- Doctored posts/images and AI are allowed, but discouraged. You MUST indicate this in your post (even if you didn't originally know). If an image is found to be fabricated or edited in any way and it is not properly labeled, it will be deleted.
- Absolutely no NSFL content.
- Be nice. Don't take anything personally. Take political debates to the appropriate communities. Take personal disagreements & arguments to private messages.
- No advertising, brand promotion, or guerrilla marketing.
RELATED COMMUNITIES:
go to feed...
@lemmy.world
A place to share screenshots of Microblog posts, whether from Mastodon, tumblr, Twitter X, KBin, Threads or elsewhere.
Created as an evolution of White People Twitter and other tweet-capture subreddits.
RULES:
- Your post must be a screen capture of a microblog-type post that includes the UI of the site it came from, preferably also including the avatar and username of the original poster. Including relevant comments made to the original post is encouraged.
- Your post, included comments, or your title/comment should include some kind of commentary or remark on the subject of the screen capture. Your title must include at least one word relevant to your post.
- You are encouraged to provide a link back to the source of your screen capture in the body of your post.
- Current politics and news are allowed, but discouraged. There MUST be some kind of human commentary/reaction included (either by the original poster or you). Just news articles or headlines will be deleted.
- Doctored posts/images and AI are allowed, but discouraged. You MUST indicate this in your post (even if you didn't originally know). If an image is found to be fabricated or edited in any way and it is not properly labeled, it will be deleted.
- Absolutely no NSFL content.
- Be nice. Don't take anything personally. Take political debates to the appropriate communities. Take personal disagreements & arguments to private messages.
- No advertising, brand promotion, or guerrilla marketing.
RELATED COMMUNITIES:
go to feed...
I don't hate AI. That's pointless. I hate the people who use AI to ruin everything, which is the majority of AI users today.
save