Marc Sbrothers
Eugen RochkoI see generative AI on the web the same way I see forever chemicals in the real world. The effects may not be immediately apparent but they are poisoning mankind. Those who acquired knowledge and skills in the before era will be fine, but newer generations won't have the same reference points to tell reality from fiction. They will search for "baby peacock" on Google and have no reason to believe the actual birds look any different to what they're shown. I find that sad somehow.
HarleyI agree, AI can be an good thing though we can't forget about the human brain!
@harljo.uk Where have I said that AI can be a good thing? I don't think we agree.
That post was a little confusing, I'll reword
I agree that generative AI is harming the web. Though AI could be a good thing for humanity, (e.g curing diseases, x-rays, etc), aslong as we continue to use our brains. I see what you are saying about it will become harder to see fact from fiction
TheFwGuy 🇪🇺🇮🇹🇺🇸🖖@Gargron @harljo.uk IMHO talking about Generative AI I agree with you, but AI is also a good thing.
It is a powerful, useful and in some cases necessary tool IF used properly.
My opinion is that AI is great to deal with things, is a danger as you say when deal with people (the "generative" part).
I think the problem is the word "intelligent" after the artificial part.
AI is NOT intelligent, is a great tool to look for patterns and associations, but is not intelligent, it doesn't understand things.
It is a powerful, useful and in some cases necessary tool IF used properly.
My opinion is that AI is great to deal with things, is a danger as you say when deal with people (the "generative" part).
I think the problem is the word "intelligent" after the artificial part.
AI is NOT intelligent, is a great tool to look for patterns and associations, but is not intelligent, it doesn't understand things.
wonofone 🍉@wonofone @Gargron @harljo.uk AI is way bigger and interesting than chatGPT.
Big LLM are the ones who use an insane amount of energy for something IMHO useless. But AI is not only that, teach yourself about edge AI for example and how AI is used other than the stupid chatGPT and similar.
AI for the masses is the problem, where too many use AI for "content", I hate that, content is for humans and should be created by humans.
AI has many more useful things that can do.
Big LLM are the ones who use an insane amount of energy for something IMHO useless. But AI is not only that, teach yourself about edge AI for example and how AI is used other than the stupid chatGPT and similar.
AI for the masses is the problem, where too many use AI for "content", I hate that, content is for humans and should be created by humans.
AI has many more useful things that can do.
Krautdragon [ Art ]@thefwguy @wonofone @Gargron @harljo.uk
The problem is the term 'AI'. It is way too vague. Whenever I bash genAI, some idiot pops up mentioning how it could advance medicine. As if doctors need all our music, writings, landscape photographs or cringe fan art to cure cancer... This is a confusion and hate that parasitic 'AI' devs with their buzzword PR can be thanked for.
feliks✨@krautdragon @thefwguy @wonofone @Gargron @harljo.uk machine learning has been helping to advance medicine for tens of years. it would mischaracterize its contributions to claim otherwise
@feliks @thefwguy @wonofone @Gargron @harljo.uk OK too bad he didn't claim otherwise? I swear you smart caricatures must have some bad case of genAI defense tourette. We get it now - to advocate for human protection against bad AIs is lowest prio. It's much much more important to never ever insult the 'good AIs'. Good neither him nor I did that?
Or... good luck with adding big tiddy anime girls, cooking recipes, fantasy books, and furry art into your medicinical AIs?
Or... good luck with adding big tiddy anime girls, cooking recipes, fantasy books, and furry art into your medicinical AIs?
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk "genAI" is included in advancing medicine. you should research the role of LLMs in e.g. drug design, molecular design, and retrosynthetic analysis. your final mischaracterization shows that you don't understand what you are talking about
@feliks @thefwguy @wonofone @Gargron @harljo.uk And those generate fake peacock babies? No. I think you are just too up your own butt to simply accept what this whole post actually and obviously addresses. Please go write to Midjourney and co about how they ruined the reputation of good genAIs and stained the term forever? That would still be better activism than this pedantic and redundant weird lecture-wanking you lot do whenever someone dares speak up against recent genAI toys.
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk i was replying to your post addressing your mischaracterizations
the peacocks and all that follows is a clear problem already discussed by competent people. does the original poster's idea of children of the future being confused by the inability to discern between reality and generated content seem frightening? yes! does it seem plausible? not at all
@feliks @thefwguy @wonofone @Gargron @harljo.uk I legit simply don't give a rats ass if I'm mischaracterising the poor precious medical genAIs and I will keep using the term genAI in criticism. And I can assure you that the idiots telling me I must stop opposing Midjourney and co because it all leads to better medicine have less knowledge about AIs than I do. Now please wank your intelligence elsewhere.
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk to be clear here it's not the term of genAI that was eliciting a response but rather this explicit passage:
> As if doctors need all our music, writings, landscape photographs or cringe fan art to cure cancer... This is a confusion and hate that parasitic 'AI' devs with their buzzword PR can be thanked for.
@feliks @thefwguy @wonofone @Gargron @harljo.uk What. That makes your response even worse, sorry.
Yes doctors do not need those silly things for their AI tools. And yes, it is the recent buzzword genAI toy market's fault that AI is now a dirty umbrella term.
Nothing of that was a mischaracterisation. Wtf 😭
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk i was being imprecise. it includes also the sentence before that. because "genAI" is clearly advancing our understanding. ofc not by those models that most people interact with
sounddrill 
@krautdragon @feliks @thefwguy @wonofone @Gargron @harljo.uk I've been on this platform for longer than a year atleast but this is honestly the first time I've seen this on fediverse
Devolving threads are common on xitter(the X is pronounced sh), and reddit, even coming across them in literal minutes after opening the respective platforms for the first time but I'm honestly proud of the Fediverse especially mastodon for throwing only self-respecting people my way so far.
@krautdragon @feliks @thefwguy @wonofone @Gargron @harljo.uk Guys, this isn't reddit. Calm down.
AI - means nothing by itself. If I market it right, even an if else block of code can be considered AI.
Generative AI is extremely bad, especially for creativity and the environment. Good, we're all on the same page. What's the issue here?
@sounddrill @feliks @thefwguy @wonofone @Gargron @harljo.uk I must politely disagree. He assumed I'm stupid, actively ignored context and chose this ridiculous discussion. He wanted me to never mention genAI in bad context as not to dishonor medical genAIs + blocked me because he just could not force me to apologise to the good medical genAIs. His weird priorities don't align with mine and this was simply my natural bullshit intolerance.
@feliks @thefwguy @wonofone @Gargron @harljo.uk And yet again another IT guy insisting an artist simply must be too stupid to "understand". I am long aware that there's AIs built from exploiting creatives and then there's various AI built from various other data used for science. But even if we say "abusive genAIs", dorks come along telling us Midjourney is not abusive and should be seen like a fellow person. :') I'm just tired of these games man.
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk the exploitation of artists by not rewarding them accordingly is horrible and similar to the damage done to the clickworkers of foreign nations. it's always the same game: if you can be exploited you will. so reddit wasn't including any content creators in their deals. i see your pain and would prefer you and others to live without it. however i would also prefer you to not play down the advances that machine learning brings
@feliks @thefwguy @wonofone @Gargron @harljo.uk Yeah, if you already insist so much for the honor and feelings of dead inanimate tools already, then I don't want to imagine how many online fights you start in defense of the honor and feelings of creative human beings! 😳
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk it is less about the tools and more about the value they provide for society
@feliks @thefwguy @wonofone @Gargron @harljo.uk omg, you mean medicinical AIs can bring immense value for humanity? I never considered!
@feliks @thefwguy @wonofone @Gargron @harljo.uk Also you cannot reward artists properly for genAI anyway. It's like letting a few naive individuals sell the land of millions of people, forever. For paychecks that must be worse than Spotify's.
@krautdragon @thefwguy @wonofone @Gargron @harljo.uk what do you believe would be a better implementation?
@feliks @thefwguy @wonofone @Gargron @harljo.uk In an ideal world, where the impossible can be real? Outlaw data sell-out "deals" that are only there to greenwash inherently predatory products as "ethical". And just keep creatives data out of that shit. I dunno why loosers are sooo desperate for funky colorful deepfake generators. But well. It would be also fine to simply guarantee that such generators are set back, easily poisened and stay extremely limited and primitive.
🅱🅻🆄🅴🅱️These replies are weird. I hear what you are saying.
You need an eye for fake images and they will get better. It was a problem before with fake gurus, fake news etc, but now it will be a bigger problem.
I don't think every last bit is bad, but yeah, it's a real weird world right now when it comes to verifying anything.
CharlieActual@harljo.uk Please stop calling it "AI" .. it's not. It's just not.
Lance R. Vick@CharlieActual @harljo.uk But it is -artificial- intelligence, which is not intelligence, in the same way an artificial plant is not a plant.
Dźwiedziu@lrvick
Mobs from the OG DOOM have more “intelligence” than an LLM “AI” /s
The term “AI” is much overhyped for wankeeteering reasons.
@dzwiedziu @CharlieActual @harljo.uk
Sure it is overhyped. So was the term "the internet" back in the 90s that was going to solve every problem the world ever had.
It did solve a few problems, and others were created.
I tried to fight hobby multirotors or uavs being called "drones" for years but at some point I had to say "drone" for anyone to know what I am talking about.
This is an inescapable truth of how language evolves, so sure, its "drones" or "AI" or maybe later "AGI" .
@lrvick @harljo.uk I understand what you are saying, "it's just a name, don't take it so seriously" but I argue it is important we don't let salesmen define our language. An artificial plant is *decorative* - people understand that, now imagine if artificial plants were being marketed by people saying "oh yeah, next year - you can just fill your fields with those and you won't need real plants, these do everything real plants do but better!!" Now imagine people are taking those claims seriously and people are losing their jobs over it.
Do you see the problem? I understand you feel these are overreactions, but please, reflect on that and reconsider. It's important we do not let salespeople define our language and it's more important we call shit when we see it - there is no up-side to failing to do so.
@CharlieActual @harljo.uk Oh I totally understand the problem.
Same reason I don't like hobby multi-rotors being called "drones", but that ship has sailed.
But also there was a thing marketers overhyped and made all sorts of crazy claims about: "The Internet".
Eventually as the public spends enough time with the thing, they fantasy definitions of a term by marketers get replaced in most minds with their own experiences with said thing.
No matter the term, marketers will hype.
Bob LeFridge 
There was a period where those of us who knew how to use Google properly were considered by some to be minor Gods. I preferred that time.
It's sad that young people will never experience the potential of the weirdly wonderful web.
kaffandoRemember when it used to give you cached results, so you could go directly to the source. Which was always real, not fake.
Them good ol' days.
mivox is exhausted 
@BobLefridge @kaffando @Gargron I was sad, but not surprised, when they took the IPO path and disappeared “Don’t be Evil” from their mission statement.
Nothing good ever comes from an IPO. 
KanaMauna@mivox @BobLefridge @kaffando @Gargron
The fact that they couldn't keep the slogan, if only just for PR purposes, speaks volumes to the depths of their depravity.
Fedo ¶@BobLefridge @Gargron to be fair, most of the people were never able to effectively search something online. I was seen as some sort of genius with computer and graphic softwares at school, but I was just able to look for the information I needed to do something, which is nothing hard to do, rather is a way to cope with laziness imho.
Maybe now the web is truly something unreliable while before it just “didn’t looked” that way?
Maybe now the web is truly something unreliable while before it just “didn’t looked” that way?
Kwink@Gargron I recently asked someone to teach me how to write code. He taught me some basic terms and then told me to ask Claude AI to write me whatever code I need.
Jared White (ResistanceNet ✊)He is not exactly wrong...
You learn by doing and you build recipes as you go that you can reuse later. While AI won't get you everywhere, it is one use of generative language models that is actually... Pretty good usually.
And when it breaks down you hit the books. Then years later you are 'good' at coding.
HowardNot sure I agree with this. There is buillding evidence that reliance on AI coding tools inhibits critical thinking and long term knowledge.
AI is Creating a Generation of Illiterate Programmers
A couple of days ago, Cursor went down during the ChatGPT outage. I stared at my terminal facing those red error messages that I hate to see. An AWS error glared back at me. I didn’t want to figure it out without AI’s help. After 12 years of coding, I’d somehow become worse at my own craft. And this isn’t hyperbole—this is the new reality for software developers.
Albert CardonaThe movie Gattaca comes to mind. In an analogy: there will be individuals who are "original" and do their own thinking, living among a sea of those who do not.
The novella "Profession" by Asimov also comes to mind. Again, as an analogy: there are consumers of generative AI tools and there are makers of generative AI tools, and the non-consumers among the latter may retain an upper hand. Perhaps in more interesting ways than e.g., a supercar being custom made rather than mass produced. It's not the exclusivity that is at stake but the very originality and meaning of each life.
@albertcardona @Gargron
I view "makers of AI" no longer as original thinkers since many of them just live as aggressive data parasites abusing other original thinkers/doers.
Likely I'm misunderstanding, but I'd say AI makers are AI consumers.
Artists for example are "custom car makers". But they are not AI makers - at least not... voluntarily. And definitely not having an upper hand in the situation, since 'human-made art loving clients' will be a shrunken down subsection of our former market
Tudor Munteanu@Gargron De acord.... acum mai nou deja autorii se folosesc de chatgpt pentru a scrie cărți...mi se pare groaznic...
Franchesca@Gargron I see similar issues for software engineering skills. As up and coming engineers rely on AI, the AI code will feed back into the AI models until they make less and less sense (or they keep a snapshot of the model fed from the pre-ai pollution era, but then it can’t suggest new code). Meanwhile the critical thinking skills earned by just screwing around with code will not be formed (I expect to do some expensive consultancy post retirement in about 20 years time to help with the cleanup)
Jeff Atwood@Gargron you vastly underestimate the intelligence of children and the unruly teenagers who change the world
Allan Chow@codinghorror @Gargron generative ai will give rise to the dankest of memes
gudenau@Gargron Stuff like this makes me wish I had more memories from web "1.0".
SpaceLifeFormWhich is why the fascists want to dumb down the education system.
They want kids to believe the bullshit and never learn critical thinking skills.
They want the kids to not have elders to learn from.
franebleu@Gargron
Every symbiote started his integration as a parasite. It's just that not every host survives long enough.
Every symbiote started his integration as a parasite. It's just that not every host survives long enough.
Michael Kohlman@Gargron For myself, the differentiator between "Gen AI" and what preceded it, is what preceded it were systems where one was looking up a fact
So, if I did not already possess the knowledge, I could find out that 2+2=4, or that a Hydrogen Atom consisted of 1 Electron and 1 Proton. Now we are moving into a realm where we are attempting to replace the creative process with the illusion of creativity
This is not a "kids with calculators" argument. What is happening will be far more impactful.
"This is not a "kids with calculators" argument. What is happening will be far more impactful."
Truth! Kids with calculators could solve a new problem. LLMs are literally plagiarism engines, they do not create new things, they amalgamate existing things.
LLMs are not AI, they are not intelligent, they do nothing new.
I wish I could ignore this phenomenon but amazingly people will not shut up about it. I need to write some fucking scripts.
@CharlieActual @Gargron LLM is most certainly not an Intelligence, they are still glorified ELIZA programs. Perhaps vastly improved versions of such, but still, that is all they are
But LLMs are being "sold" as a replacement for creativity and people will buy into it because the lazy solution is the easy out. That is why I worry, because it will potentially atrophy creativity in upcoming generations; and as Eugen stated, the atrophy won't be immediately apparent. It makes me sad as well
The Animal and the Machine@Gargron
I agree with the sentiment but not the projection. There will be a grey gap but ultimately a verified source will arise.
Those who stray from reality will suffer an eventual reality correction. The longer they maintain it the bigger the correction, all the way up to the fall of empires.
At the moment the lack of European military cover led to US ‘soft power’ on open markets. The American firms won as their unethical practices can out compete those who don’t burn out people or work to not polite the commons.
The rise of European independence will provide the opportunity to do AI right and ethical. We are the hope and so we are the ones who can grasp the destiny. After all, I don’t need to worry about the garbage of X because a better way was invented by good people.
Thank you for that by the way.
Everything is connected and we demand better AI and we don’t mean bigger garbage. I’m confident we will again be able to reliably find a picture of a baby peacock.
Clm