image description (contains clarifications on background elements)
Lots of different seemingly random images in the background, including some fries, mr. crabs, a girl in overalls hugging a stuffed tiger, a mark zuckerberg ābig brother is watchingā poser, two images of fluttershy (a pony from my little pony) one of them reading āu only kno my swag, not my loreā, a picture of parkzer parkzer from the streamer ādougdougā and a slider gameplay element from the rhythm game āosuā. The background is made light so that the text can be easily read. The text reads:
i wanna know if we are on the same page about ai.
if u diagree with any of this or want to add something,
please leave a comment!
smol info:
- LM = Language Model (ChatGPT, Llama, Gemini, Mistral, ...)
- VLM = Vision Language Model (Qwen VL, GPT4o mini, Claude 3.5, ...)
- larger model = more expensivev to train and run
smol info end
- training processes on current AI systems is often
clearly unethical and very bad for the environment :(
- companies are really bad at selling AI to us and
giving them a good purpose for average-joe-usage
- medical ai (e.g. protein folding) is almost only positive
- ai for disabled people is also almost only postive
- the idea of some AI machine taking our jobs is scary
- "AI agents" are scary. large companies are training
them specifically to replace human workers
- LMs > image generation and music generation
- using small LMs for repetitive, boring tasks like
classification feels okay
- using the largest, most environmentally taxing models
for everything is bad. Using a mixture of smaller models
can often be enough
- people with bad intentions using AI systems results
in bad outcome
- ai companies train their models however they see fit.
if an LM "disagrees" with you, that's the trainings fault
- running LMs locally feels more okay, since they need
less energy and you can control their behaviour
I personally think more positively about LMs, but almost
only negatively about image and audio models.
Are we on the same page? Or am I an evil AI tech sis?
IMAGE DESCRIPTION END
i hope this doesnāt cause too much hate. i just wanna know what u people and creatures think <3
You must log in or register to comment.
I used to think image generation was cool back when it was still in the āgenerating 64x64 pictures of catsā stage. I still think itās really cool, but I do struggle to see it being a net positive for society. So far it has seemed to replace the use of royalty free stock images from google more than it has replaced actual artists, but this could definitely change in the future.
There are some nicer applications of image generation too, like dlss upscaling or frame generation, but I canāt think of all that much else honestly.
I agree š
I think we should avoid simplifying it to VLMs, LMs, Medical AI and AI for disabled people.
For instance, most automatic text capture ais (optical Character Recognition, or OCR) are powered by the same machine learning algorithms. Many of the finer-capability robot systems also utilize machine learning (Boston Dynamics utilizes machine learning for instance). Thereās also the ability to ID objects within footage, as well as spot faces and referencing it with a large database in order to find the person with said face.
All these are Machine Learning AI systems.
I think it would also be prudent to cease using the term āAIā when what we actually are discussing is machine learning, which is a much finer subset. Simply saying āAIā diminishes the termās actual broader meaning and removes the deeper nuance the conversation deserves.
Here are some terms to use instead
- Machine Learning = AI systems which increase their capability through automated iterative refinement.
- Evolutionary Learning = a type of machine learning where many instances of randomly changed AI models (called a āgenerationā) are run simultaneously, and the most effective is/are used as a baseline for the next āgenerationā
- Neural Network = a type of machine learning system which utilizes very simple nodes called āneuronsā for processing. These are often used for image processing, LMs, and OCR.
- Convolution Neural Network (CNN) = a Neural network which has an architecture of neuron āflitersā layered over each other for powerful data processing capabilities.
This is not exhaustive but hopefully will help in talking about this topic in a more definite and nuanced fashion. Here is also a document related the different types of neural networks
Mr crabs would use unethical llms, very accurate
true, he would totally replace his workers with robots, and then complain about hallucinated recipes.
There are so many different things that are called AI, the term AI doesnāt have any meaning whatsoever. Generally it seems to mean anything that includes machine learning somewhere in the process, but itās largely a marketing term.
Stealing art is wrong. Using ridiculous amounts of power to generate text is ridiculous. Building a text model that will very confidently produce misinformation is pretty dumb.
There are things that are called AI that are fine, but most arenāt.
Iāll just repeat what Iāve said before, since this seems like a good spot for this conversation.
Iām an idiot with no marketable skills. I want to write, I want to draw, I want to do a lot of things, but Iām bad at all of them. gpt like ai sounds like a good way for someone like me to get my vision out of my brain and into the real world.
My current project is a wiki of lore for a fictional setting, for a series of books that I will never actually write. My ideal workflow involves me explaining a subject as best I can to the ai (an alien technology or a kingdomās political landscape, or drama between gods, or whatever), telling the ai to ask me questions about the subject at hand to make me write more stuff, repeat a few times, then have the ai summarize the conversation back to me. I can then refer to that summary as I write an article on the subject. Or, me being lazy, I can just copy-pasta the summary and thatās the article.
As an aside, I really like chatgpt 4o for lore exploration, but Iād prefer to run an ai on my own hardware. Sadly, I do not understand github and my brain glazes over every time I look at that damn site.
It is way too easy for me to just let the ai do the work for me. Iāve noticed that when I try to write something without ai help, itās worse now than it was a few years ago. generative ai is a useful tool, but it should be part of a larger workflow, it should not be the entire workflow.
If I was wealthy, I could just hire or commission some artists and writers to do the things. From my point of view, itās the same as having the ai do the things, except itās slower and real humans benefit from it. Iām not wealthy though, hell, I struggle to pay rent.
The technology is great, the business surrounding it is horrible. Iām not sure what my point is.
I wish people stopped treating these fucking things as a knowledge source, let alone a reliable one. By definition they cannot distinguish facts, only spit out statistically correct-sounding text.
Are they of help to your particular task? Cool, hope the model youāre using hasnāt been trained on stolen art, or doesnāt rely on traumatizing workers on the global south (who are paid pennies btw) to function.
Also, yāknow, donāt throw gasoline to an already burning planet if possible. You might think you need to use a GPT for a particular task or funny meme, but chances are you actually donāt.
Thatās about it for me I think.
edit: when i say āyouā in this post i donāt mean actually you OP, i mean in general. sorry if this seems rambly im sleep deprived as fuckj woooooo
peeps who use these models for facts are obv not aware what the models are doing. they donāt know that these models are just guessing facts.
also yes, big sad about peeps in the south being paid very poorly.
can totally see your point, thank you for commenting! <3
There is an over arching issue with most of the extant models being highly unethical in where they got their data, effectively having made plagiarism machines.
It is not ok to steal the content of millions of small independent creators to create slop that drowns them out. Most of them were already offering their work for free. And I am talking about LMs here, writing is a skill.
Say what ever you want about big companies being bad for abusing IP laws, but this is not about the laws, not even paying people for their work, this is about crediting people when they do work, acknowledging that the work they did had value, and letting people know where they can find more.
Also, I donāt really buy the āitās good for disabled peopleā that feels like using disabled people as a shield against criticism, and Iāve yet to see it brought up in good faith.
A human can read examples of good articles to learn how to write a good article, but an AI canāt?
It seems kinda arbitrary, I donāt think you can say anything objective about whether AI is plagiarism or not besides the most literal definition of the law (which is impossible as it itself is made arbitrary through the idea of fair use)
I honestly am skeptical about the medical stuff. Machine learning canāt even do the stuff it should be good at reliably, specifically identifying mushrooms/mycology in general.
that is interesting. i know that there are plenty of plant recognition onces, and recently there have been some classifiers specifically trained on human skin to see if itās a tumor or not. that one is better than a good human doctor in his field, so i wonder what happened to that mushroom classifier. Maybe it is too small to generalize or has been train in a specific environment.
I havenāt looked closely enough to know, but I recall medical image analytics being ābetter than humanā well before the current AI/LLM rage. Like, those systems use machine learning, but in a more deterministic, more conventional algorithm sense. I think they are also less worried about false positives, because the algorithm is always assumed to be checked by a human physician, so my impression is that the real sense in which medical image analysis is ābetterā is that it identifies smaller or more obscure defects that a human quickly scanning the image might overlook.
If youāre using a public mushroom identification AI as the only source for life-and-death choice, then false positives are a much bigger problem.
yes, that is what i have heard too. there was a news thing some days ago that this ācancer scannerā thing will be available in two years to all doctors. so thatās great! but yes, we very much still need a human to watch over it, so its out-of-distribution-generations stay in check.
Do not trust AI to tell you if you can eat a mushroom. Ever. The same kinds of complexity goes into medicine. Sure, the machine learning process can flag something as cancerous (for example), but will always and forever need human review unless we somehow completely change the way machine learning works and speed it up by an order of magnitude.
yeah, we still very much need to have real humans go āyes, this is indeed cancerā, but this ai cancer detection feels like a reasonable āfirst passā to quickly get a somewhat good estimation, rather than no estimation with lacking doctors.
From what little I know if it, itās sorta twofold what it does:
-
It looks through documentation across a patient record to look for patterns a doctor might miss. For example, a patient comes in complaining of persistent headaches/fatigue. A doctor might look at that in isolation and just try to treat the symptoms, but an AI might see some potentially relevant lab results in their histories and recommend more testing to rule out a cancer diagnosis that the doctor might have thought unlikely without awareness of that earlier data.
-
Doctors have to do a lot of busywork in their record keeping that AIs can help streamline. A lot of routine documentation, attestations, statements, etc. Since so much of it is very template-heavy already, an AI might be able to streamline the process as well as tailor it better to the patient. E.g. the record indicates āassigned male at birthā and an ER doctor defaults to he/him pronouns looking only at the medical birth sex marker, but the patient is also being seen by a gender clinic at which she is receiving gender affirming treatment as a trans woman and brings up that earlier data to correct the documentation and make it more accurate and personalized for the patient.
In reality, I am sure that practices and hospital systems are just going to use this as an excuse to say āYou donāt need to spend as much time on documentation and chart review now so you can see more patients, right?ā Itās the cotton gin issue.
-
Having worked with ML in manufacturing, if your task is precise enough and your input normalized enough, it can detect very impressive things. Identifying mushrooms as a whole is already too grand a task, especially as it as to deal with different camera angles, lighting ā¦ But ask it to differentiate between a few species, and always offer pictures using similar angles, lighting and background, and the results will most likely be stellar.
Like I said, Iām just skeptical. I know it can do impressive things, but unless we get a giant leap forward, it will always need extensive human review when it comes to medicine (like my mycology example). In my opinion, it is a tool for quick and dirty analysis in the medical field which may speed things up for human review.
My biggest problem with AI is how it was pushed and marketed to us in ways that donāt make sense / are unethical. Even the environmental concerns would be ameliorated if AI werenāt being pushed into everything. (Using āAIā here to refer to things like LM, image, and art generators,etc.)
yes, i completely agree.
having some LM generate ācomment suggestionsā for content creators on youtube is such a genuine waste of compute and the environment. (yes this is a real thing)
it was marketed as this āsmart machineā which ends up being too dum for most people wanting to use it.
This not relevant to your comment, really, but I would like to applaud you for word usage.
This list is missing: AI generated images are not art.
i also think that way, but itās also true that generated images are being used all over the web already, so people generally donāt seem to care.
Genuinely, the only problem I see with the development of LLMs and AI in general is that said development has a massive tumor on its back called Corporate Interest. Thatās pretty much the one and only cause for absolutely every destructive, shady, or downright immoral aspect tied to these things nowadaysā¦
As tools in and of themselves, yes! LLMs have an immense potential not of replacing people, but of helping people get stuff done faster, which in turn would give us a lot of extra time to polish the everloving spit out of the stuff we make!
LLM/AI research should be 100% non-profit and democratised, with well-established guidelines and full transparency, as I see it. This is a huge step in our development as a species, and Altman-likes are not the people who should be in charge of it.
Edit: as for VLMs, I kindaā see them as a fad, to be honest. It still irks me when anyone adds āartā to anything artificially generated at the moment, but I get the feeling people will tire of the novelty once the need for genuine art will cease being satisfied by the above-mentioned.
oh, nonon, VLMs only accept text and images as input. they donāt produce images. they just have image inputs as an option.
what you are refering to are āimage generatorsā, or ādiffusion networksā. unfortunately, many news outlets already only use AI images for their stories. i find this pretty sad, cuz i liked that they made a human put together some panel for the news! but not anymoreā¦ now itās a mixture of stock footage and AI image crapā¦ big sad ;(
yes, i am negative to image gen models.
alsoalso yes, communism go, non-profits are cool, and i wish what you said became true
Oooh, thank you for the clarification and I apologise for the confusion!
We really are losing a lot of our personality as a species by using generated imagery, yesā¦ Itās, unfortunately, been a general trend over the last couple of decades in pretty much all things, architecture especially imho (referring to āaverageā buildings, not the ones specifically designed to be crazy, which are cool, but far and few betweenā¦)
yesā¦ older cities look so much more interesting! where u can see the wooden beams and such! for some reason building big blocks is cool now thoā¦ I guess itās good for storage, but surely people donāt this super boring.
LMs give the appearance of understanding, but as soon as you try to use them for anything that you actually are knowledgable in, the facade crumbles.
Even for repetitive tasks, you have to do a lot of manual checking to ensure they did not start hallucinating half way through.
I havenāt really used AIs myself, however one of my brothers loves AI for boilerplate code which he of course looks over afterwards. If it saves time and you only have to do some minor editing then that seems like a win to me. Probably shouldnāt be used like this in any non-hobby project by people who arenāt adept at coding however
Iām a programmer as well. When ChatGPT & Co initially came out, I was pretty excited tbh and attempted to integrate it into my workflow, which kinda worked-ish? But was also a lot of me being amazed by the novelty, and forgiving of the shortcomings.
Did not take me long to phase them out again though. (And no, itās not the models I used; I have tried again now and then with the new, supposedly perfect-for-programming models, same results). The only edgecase where they are generally useful (to me at least) are simple tasks that I have some general knowledge of (to double theck the LMās work) but not have any interest in learning anything further than I already know. Which does occur here and there, but rarely.
For everything else programming-related, itās flat out shit.I do not beleive they are a time saver for even moderately difficult programs. Bu the time youāve run around in enough circles, explaining ānow, this does not do what you say it doesā, āthatās the same wring answer you gave me two responses agoā, āyou have hallucinated that functionā, and found out the framework in use dropped that general structure in version 5, you may as well do it yourself, and actually learn how to do it at the same time.
For work, I eventually found that it took me longer to describe the business logic (and do the above dance) than to justā¦ do the work. I also have more confidence in the code, and understand it completely.
In terms of programming aids, a linter, formatter and LSP are, IMHO, a million times more useful than any LM.
this matches my experience too. good IDEs or editors with LSP support allll the way.
also wanna add that itās weird to me that we turn to LLMs to generate mountains of boilerplate instead ofā¦ yāknow, fixing our damn tools in the first place (or using them correctly, or to their fullest) so that said boilerplate is unnecessary. abstractions have always been a thing. it seems so inefficient.
Makes me feel warm around the heart to hear that itās not just me š«
ikr, it makes the horrors just a little more bearable āØ
I also 100% agree with you. My work has a developer productivity team that tries to make sure we have access to good tools, and those folks have been all over AI like flies on shit lately. Iāve started to feel a bit like a crazy Luddite because I do not feel like Copilot increases my productivity. Iām spending like 90% of my time reading docs, debugging and exploring fucked up edge cases, or staring off into space while contemplating if Iām about to introduce some godawful race condition between two disparate systems running in kubernetes or something. Senior developers usually do shit that would take hours to properly summarize for a language model.
And yeah, if I have to write a shitload boilerplate then Iām writing bad code and probably need to add or fix abstraction. Worst case, thereās always vim macros or a quick shell oneliner to generate that shit. The barrier to progress is useful because it warns me that Iām being a dummy. I donāt want to get rid of that when the only benefit is that I get to context switch between code review mode and system synthesis mode.
Yeah, with seniors itās even more clear how little LMs can help.
I feel you on the AI tools being pushed thing. My company is too small to have a dedicated team for something like that, buuuutā¦ As of last week, weāre wasting resources on an internal server hosting Deepseek on absurd hardware. Like, far more capable than our prod server.
Oh, an we pride ourselves on being soooo environmentally friendly šš
for even moderately difficult programs.
My brother uses it to generate templates and basic structs and functions, not to generate novel code. Thatās probably the difference here. I believe itās integrated into his text editor as well? Itās the one github offers
Edit: Probably wouldnāt be useful if it wasnāt integrated into the editor and therefore the generation being just a click away or some sort of autofill. Actually writing a prompt does sound tedious
Iāve heard this argument so many fucking times and i hate genai but thereās no practical difference between understanding and having the appearance of such, that is just a human construct that we use to try to feel artificially superior ffs
No. I am not saying that to put man and machine in two boxes. I am saying that because it is a huge difference, and yes, a practical one.
An LLM can talk about a topic for however long you wish, but it does not know what it is talking about, it has no understanding or concept of the topic. And that shines through the instance you hit a spot where training data was lacking and it starts hallucinating. LLMs have āreadā an unimaginable amount of texts on computer science, and yet as soon as I ask something that is niche, it spouts bullshit. Not itās fault, itās not lying; itās just doing what it always does, putting statistically likely token after statistically liken token, only in this case, the training data was insufficient.
But it does not understand or know that either; it just keeps talking. I go āthat is absolutely not right, remember that <ā¦> is <ā¦,>ā and whether or not what I said was true, it will go "Yes, you are right! I see now, <continues to hallucinate> ".
Thereās no ghost in the machine. Just fancy text prediction.
youāre right, it doesnāt do classification perfectly every time. but it drills down on the amount of human labour required to classify a large set of data.
about the knowledge: it really comes down to which model you are talking to. āgeneralistā models like GPT4o or claude 3.5 sonnet have been trained to know many things somewhat, but no single thing perfectly.
currently companies seem to train largely on IT-related things. these models are great at helping me program, but they are terrible at specifically writing GDScript (a niche game-programming language) since they forget all the methods and components the language has.
Even with LMs supposedly specialising in the areas that I am knowledgable (but by no means an expert) in, itās the same. Drill down even slightly beyond surface-level, and itās either plain wrong, or halucinated when not immediately disprovable.
And why wouldnāt it be? These things do not possess knowledge, they possess the ability to generate texts about things weād like them to be knowledgable in, and that is a crucial difference.
A lot of those points boil down to the same thing: āwhat if the AI is wrong?ā
If itās something that youāll need to check manually anyway, or where a mistake is not a big deal, thatās probably fine. But if itās something where a mistake can affect someoneās well-being, that is bad.
Reusing an example from the pic:
- Predicting 3D structures of proteins, as in the example? OK! Worst hypothesis the researchers will notice that the predicted structure does not match the real one.
- Predicting if you have some medical problem? Not OK. A false negative can cost a life.
Thatās of course for the usage. The creation of those systems is another can of worms, and it involves other ethical concerns.
of course using ai stuffs for medical usage is going to have to be monitored by a human with some knowledge. we canāt just let it make all the decisionsā¦ quite yet.
in many cases, ai models are already better than expert humans in the field. recognizing cancer being the obvious example, where the pattern recognition works perfectly. or with protein folding, where humans are at about 60% accuracy, while googles alphafold is at 94% or so.
clearly humans need to oversee AIs output, but we are getting to a point where maybe humans make the wrong decision, and deny an AIs correct generation. so: no additional lives are lost, but many more could be saved
I mostly agree with you, I think that weāre disagreeing on details. And youāre being far, far more level-headed than most people who discuss this topic, who pretend that AI is either e-God or Satanic bytes. (So no, you arenāt an evil AI tech sis. Nor a Luddite.)
That said:
For clinical usage, just monitoring it isnāt enough - because when people know that thereās some automated system to catch their mistakes, or that theyāre just catching the mistakes of that system, they get sloppier. You need really, really good accuracy.
Like, 95% accuracy might look like a lot, right? If it involves death or life, it means a death for each 20 cases, itās rather high. In the meantime, if AlphaFold got it wrong 60% of the time instead of just 6%, it wouldnāt be a big deal.
Also, note that weāre both talking about āAIā as if it was a single thing. Under the hood itās a bunch of completely different things; pattern recognition AI, predictive AI, generative AI, they work so differently from each other that weād need huge walls of text to decide how good or bad each of them is.
Ultimately, the issue is our current societies being fucked. If AI were refined, sensibly monitored and generally used by people who can recognize mistakes (where it matters), and keep their fossil fuel usage in check, AI could be a big step towards gay space communism. Like, who wants to do menial labor? Let AI do it where sensible and pay the former workers the money thatās saved by doing that. But as it is, itās mostly going to be used to further the agendas of authoritarians and capitalists.
yesyey, this very much. in the hands of people who know the capabilities of the models, they tend to use them well and speed up their work. gay space communism would be totally cool if shiddy jobs could slowly be automated away <3
but yea, big sad cuz evil capitalists go āyesyes we make ai for ur businessā even tho world would be better without business ~ ~
