Report Blog Entry
i love arguing with people in tiktok comments. its awful idk why i do it.
today i saw a post, op was making the point that, if the US would stop sending money to israel, they'd be able to stop homelessness and other things. i agree and think it's a good topic to post about. my only problem was that this post was a series of screenshots, a conversation with chatgpt. "how much would it cost to make education free?" and it replied, "about $x."
of course i found it alarming that seemingly the only research this person did was talk to chatgpt, so i commented something to the effect of, can we please not use chatgpt as research?
holy brainless. over 50 replies, from at least 20 unique users, and only TWO in support of me. here are the most alarming replies i have seen:
1. "people are using their brains more with chatgpt. makes it easier to think for the most mid among us." WHAT. ????? i'm completely speechless at this one.
2. 'google is pro govt / capitalism / censorship and chat is more unbiased ' okay, use a different search engine, a vpn... + who's to say chatgpt isnt controlled by them too..? THEN this person says theyve been using chat to research history, and because it explained bad things the cia did, they KNOW it's accurate and unbiased. HUH???
3. sarcastically, "the smartest ai on earth isnt a good news source," ..why are we making the assumption that ai is a good news source to begin with.. "oh so you trust google which is also an ai?" no, i trust certain articles and journalists and scientists and historians... "name your fav 5 journalists." GIRL WHO CARES WHAT MY FAVS ARE. USE YOUR BRAIN.
4. "i looked up [video's claims] on google and it said the same thing. how is google unreliable? laughing emoji"... i said CHAT is unreliable not ARTICLES on google SOBBING EMOJI. i can tell these people are brainrotted beyond saving bc their replies SOUND LIKE CHAT. "ai doesn't create answers, it searches for answers to your question while at the same time fact checking its credibility -" shut up please you clone.
5. "chatgpt is so much more reliable than using your brain." okay. sure. if it did not make up information, if it were not a LLM (cares more about sounding human than anything), if it were just an information gathering bot, fine, yes, it would be more reliable. ****but is this a society we want?? do we want everyone relying on a robot to find information for them, to think for them? do we not care for learning from our own mistakes anymore?? do we not care for hearing from others anymore? reading their books and articles and feeling the work they put into their writing and videos, do we not care to hear their voices????****
maybe it's just one tiktok comment section. but it scares me that anyone thinks like this. i just read fahrenheit 451 and i see those people living in that world and being fine with it.
Comments
Displaying 3 of 3 comments ( View all | Add Comment )
The Walrus
I very very much relate to this anger. It's so heart breaking to see so many people just not use their brains, and it seems like trying to reason with them is pointless. As someone who is seeing more and more use of LLMs in my peers (im in high school), i genuinely wonder if such dependence is good for our development. I'm sure always having the "right" answer at the tip of your fingers isn't good, but I haven't found many was to get my friends to stop. It's the path of least resistance and most people will take it, even if we do shout about all the horrible things A.I.'s cause.
Also, side note: I'm tired of teachers talking to our classes talking about how terrible A.I. is. I don't think repeating the same points will cause most people to change. If anything, they'll become desensitized and block out the noise. Trying to make people less dependent on A.I. is one of those things where you have to nudge people in the right direction and hope they decide to take the first step. You can't really force someone with depression to start caring for themselves, they'll just feel criticized and shut down. But, you can give them water, food, strength, and encourage them to step out of bed themselves.
you're right, that dependence is terrible for ur brain/development!! someone else that commented on this post sent me a study about how bad it is for critical thinking skills (https://www.mdpi.com/2075-4698/15/1/6)!!!
i just wish i knew the best way to get people to stop using it. i recently talked to someone who knew how bad ai is but didnt care to stop/be better/learn more/learn how to think for themselves and... what is there to say to that?
i think youre right about that too and it suuucckss
at least people like us will be ahead of them i guess? it's better for me in general when my competition has to have a stupid robot think for them...? that's so sad
by holly; ; Report
(thank you for the study, i will so read that later)
There isn't much to say to those people, which, yeah, it's sad. and, this may just be me and my left leaning bias, but i don't think we should view it as "at least we will be ahead". I assume people who use ai are people who never had access to a good education, so they use chatbots to make up for it. which will only make their cognitive skills worse and widen the class gap. I think the best we can do now is continue to speak up about ai, a lot of people (including me) don't have the resources to do anything else. But if this "anti-ai" thing becomes a big movement, we might influence people to step out of their little chat bot bubble and into actions that improve their ability to think.
by The Walrus; ; Report
i completely agree! :)
by holly; ; Report
LuciLucilia
Seen a lot of posts like this on TikTok and Instagram... it is so insanely worrying to me. LLMs aren't a perfect tool, hell, they're often not even a particularly good one. Here is an article on a study out of Columbia that found that AI weren't great at finding the desired results for their study's criteria and methodology, and were often either wrong, or only partially correct. Rarely were they completely correct. (Sourced https://www.cjr.org/tow_center/we-compared-eight-ai-search-engines-theyre-all-bad-at-citing-news.php since unlike AI-brained idiots, I can actually provide sources beyond what a digital-algorithm tells me).
Its also actually eroding people's intelligence so it seems, as seen by another study ( https://www.mdpi.com/2075-4698/15/1/6 ). Even more... it seems to be able to trigger psychosis in some people... not sure there are any studies on this, but it is pretty well documented online... Anyways, I am gonna reply to those how I would if I was actually in this comment section, for funsies (gotta work out those critical thinking capabilities, clearly not enough people do.)
1. Not only is this not the case as far as the research suggests, since all it really does is let the "most mid" fall into being the lowest low through half-assery, even if it was some great equalizer, the better models are still behind paywalls and fees, meaning class will still be a deciding factor for "the most mid"
2. Well, I guess this person must at least dislike Gemini, since they're so apparently opposed to Google. Which... they should be, but all of those other models have Google and its biases in their database too, not to mention that of all of the other tech monoliths that Google is acquaintances with. Plus, like you said, they could use lots of different search engines. I would go with Firefox for the browser and DuckDuckGo for the engine + a couple extensions like uBlock Origin, if they want something easy.
3. Google isn't even an AI... it has an AI feature (as does fucking everything now, unfortunately), as well as an AI model, but it is not *itself* an AI, it is a chromium-based browser and an engine. Also, I think most people don't really pay attention to the journalists themselves, more so the networks and channels and other organizations those journalists work under. Maybe they should pay more attention to individual journalists, but still. My favorite networks are Al Jazeera, TheGuardian maybe, and perhaps Democracy Now!. Really though, its better to get your news from a variety of places (I am not about to become a Ground News ad, do not worry).
4. It does not necessarily fact-check. They have criteria, some models a bit better than others, but they cannot fact-check since that would require them to have both sentient intelligence and the ability to investigate and compare. Queue all of the images of AIs being like "yes eating rocks is okay" because of some random satirical source they found or whatever.
5. I actually agree with this point even less than you do. The information those algorithms contain are not random, spontaneous, self-created, or novel; we, humans, made that information. We gave it to them, put it into their datasets... their information is our information. They can't even connect it any better than we do, in fact, they do it far worse, since they are currently incapable of understanding context, they do not have intuition (and likely will never have these things).
If we rely on them for all our information, then the information they have is all we would ever have. We would become stunted creatures incapable of providing more information to the world... unfortunately, that is what these people are campaigning for.
Anyways, hope you enjoyed 451, its a lot better than 1984 if you ask me.
thank u for sending the articles!! i skimmed them and found some interesting things. "generative search tools were often confidently wrong" juuust like the people that use it.. do they not stop to question anything? i heard once that the potential i see in others is my own potential, what i'd do in their situation.. and i feel like questioning "obvious" thoughts should be natural to everyone but i guess that's not true T_T
"increased reliance on AI tools is associated with reduced critical thinking abilities." im so glad someone actually studied this bc it's not just me speculating now :D ive been saying it makes sense that exercising your brain makes it stronger and not doing so makes it weaker. but i wonder if the loss of critical thinking skills is something that can be repaired. i know it's easy for young children to learn languages, and ppl mostly lose that ability as they grow.. is it the same with skills like critical thinking? once teenagers avoid learning it by using ai, can they easily learn it in the future? or are we cooked?
the psychosis thing TOTALLY makes sense to me. if u say the right things to chatgpt u can get it to tell you whatever you want, it's not hard at all, and that can easily feed into a bipolar episode/schizophrenic delusion :( people can find meaning in any little thing, but when an artificial *intelligence* is spelling it out right in front of you? scary. i found a short article about it, it's not a study though (https://www.papsychotherapy.org/blog/when-the-chatbot-becomes-the-crisis-understanding-ai-induced-psychosis)
*class will still be a deciding factor for "the most mid".* i hadn't thought about that and thats an important addition to the conversation!! sure, ais that cost money are not much better now, but who knows how accurate theyll be in a year? the poorer people will already be limited by their current ai use + lack of critical thinking skills, and on top of that, they wont have (easy) access to the most accurate information. that's for the people with money to spare
unfortunately (fortunately for my sanity) op deleted my comment so i cant tell that person these good suggestions (im writing this on firefox + ublock origin rn :3)
lmaoo ground news ad, i can always feel their aura from 30 seconds away in every philosophy video i watch T_T
HAVE THEY NOT SEEN THE IMAGES OF GOOGLE AI TELLING PEOPLE TO EAT ROCKS OMG. "one reddit user says kill yourself" LMAO how can anyone fully trust ai after thattt
*We would become stunted creatures incapable of providing more information to the world.* THISSS IS ABSOLUTELY RIGHT
i did like 451! i expected a depressing ending but surprisingly it was happyish? ik its a literal nuke but like.. at least that little part of the world gets a reset yk? maybe i just think its a happy ending bc i read it soon after finishing 1984 T_T i still dont know how to feel about that ending, at least what it meant for winston. he died loving big brother (WAA)... but love is hate, right?
i think i enjoyed reading 1984 more bc i spent a lot more time with it. i couldnt find an audiobook for MONTHS so id only read a few pages at a time (february - june... school got in the way too oops). but with 451 i found some english teacher reading it on youtube and finished it in 3 hours.
overall i think i liked both equally? 8/10. but i just started reading books again this year (after like 10 yrs) so i think i need to read a lot more, see what i actually like, then reread them both and rate them again. same with handmaids tale which i finished last night (i read 2 books in 3 days!! which is insane, for me!!)
by holly; ; Report
Yes, one of the biggest findings of that article is just simply that AI are often biased to present information confidently, even when they should present it with more uncertainty or speculatively.
. I don't really care for 1984 since its really clear to me that Orwell sees the Soviet Union as somehow worse and more authoritarian than the Nazis, which they just simply were not, not even close. Isaac Azimov has a really good review of 1984... I think its a bit too harsh at some points, but
I can't speak for the critical thinking thing, though I'd also be super interested to know, but actually, its kind of a misconception that babies/children learn language better than adults. Adults can learn grammar, verb conjugation, and vocabulary with relative ease in a relatively short amount of time if they put the effort and time into it. Babies could not so quickly learn a language. The thing that babies are just... way significantly better at is learning pronunciation and gaining the ability to speak like a native in the language. It is wayy harder for adults to learn to speak like a language's native.
Its neat that it is being documented by some bigger organizations now. A study would be cool, not exactly sure what they could / would do to achieve that, but I am sure something could be worked out.
Yes exactly, and already AI is causing a lot of trouble in academia. Its both taking entry level internship opportunities, as well as just straight up ruining a lot of people's critical thinking if they use it. This is something that makes me, in a terrible roundabout way, kind of hopeful, since it means that this can't really last, its unsustainable.
Awe thank youu, I am also on Firefox with uBlock, though I also have other browsers, engines, etc for if I want to search even more unbiasedly than that.
I feel like I really conceptually like what Ground News does, but I am immediately skeptical of any product that is in a Youtube ad.
They probably just see it as "early kinks", which in some ways it was, its not so easy to get AI to tell you to eat rocks now, but that also just means that all the lies it tells you are much subtler and harder to detect, especially if you don't have the critical thinking skills to detect them because AI stole even that from you.
Yess, thank youu.
Winston didn't ever not love Big Brother, hate is love
"He wasn't much affected, apparently, by the Nazi brand of
totalitarianism, for there was no room within him except for his private war
with Stalinist communism. Consequently, when Great Britain was fighting for
its life against Nazism, and the Soviet Union fought as an ally in the
struggle and contributed rather more than its share in lives lost and in
resolute courage, Orwell wrote Animal Farm which was a satire of the Russian
Revolution and what followed, picturing it in terms of a revolt of barnyard
animals against human masters." -Isaac Asimov, Review of 1984.
Owhh I am proud of youu! I haven't actually read HM's T.
by LuciLucilia; ; Report
chi【罪】(AUTOPLAY)
i'm fully convinced that chatgpt users are lobotomy victims
genuinely. the lack of the most basic logic made me feel like *I* was losing my mind. "why would companies replace everything with it if it wasnt reliable smirk emoji" DO YOU THINK. AT ALL.
by holly; ; Report