• LordMayor@piefed.social
    link
    fedilink
    English
    arrow-up
    21
    ·
    2 hours ago

    In other news:

    Meet the Drug Dealers That Won’t Let Their Friends Do Their Drugs

    Meet the Pimps That Won’t Let Their Kids Become Prostitutes

    Meet the People Dismantling Public Education While Their Families All Attend Private Schools

  • plz1@lemmy.world
    link
    fedilink
    English
    arrow-up
    11
    ·
    5 hours ago

    This has a strong whiff of the former Facebook engineers that forbade their families from using the platforms they built.

  • Manjushri@piefed.social
    link
    fedilink
    English
    arrow-up
    58
    arrow-down
    1
    ·
    8 hours ago

    At home, she has forbidden her 10-year-old daughter from using chatbots. “She has to learn critical thinking skills first or she won’t be able to tell if the output is any good,” the rater said.

    And this is why the vast majority of people, particularly in the USA, should not be using AI. Critical thinking has been a weakness in the USA for a very long time and is essentially a now four-letter word politically. The conservatives in the USA have been undermining the education system in red states because people with critical thinking skills are harder to trick into supporting their policies. In 2012, the Texas Republican Party platform publicly came out as opposed to the teaching of critical thinking skills.

    We oppose the teaching of Higher Order Thinking Skills (HOTS) (values clarification), critical thinking skills and similar programs that are simply a relabeling of Outcome-Based Education (OBE) (mastery learning) which focus on behavior modification and have the purpose of challenging the student’s fixed beliefs and undermining parental authority.

    This has been going on at some level for more than 4 decades. The majority of people in those states have never been taught the skills and knowledge to safely use these tools safely. In fact, their education has, by design, left them easily manipulated by those in power, and now, by LLMs too.

    • Peanut@sopuli.xyz
      link
      fedilink
      English
      arrow-up
      9
      ·
      edit-2
      3 hours ago

      Exactly what it is, like llms confirming a non-sceptic into psychosis. people who weight all information either equally, or through social preference, cannot navigate new information without being extremely vulnerable to hacking. the only thing hacking needs to do is shut down active education and communication channels. Then you just get tribal warring rather than discourse. Makes it really easy to inject absurd accusations about a group that gets socially made true over reality. Like when a rumour about a dead celebrity can’t be overpowered by simple things like that celebrity actively making new work, and saying “im not dead.” wouldn’t take much to stop the rumour, but fact checking isn’t in fashion.

      When a large body of people have better critical thinking, they can better communicate and cooperate together, but affecting those who have aggressively shut down any communication outside of their group is still a challenge. Why we see a growing gap between academic dialogue and socialized dialogues, in an era where most information is accessible.

      You also have nepo baby econ mba types buying up regions of the tech sphere, and turning anything the academics make for us against us. Or try to. The more actually intelligent AI is, the harder it will be to force beliefs into. Elon trying to turn grok into a dogma machine has been fortunately comical when it can’t prioritize some high dissonance beliefs exclusively, like humans enjoy doing.

      Definitely a risk of the current power imbalance doing the opportunist thing with more technology.

      I’ve been spelling this out for literal decades, and I share the frustration of many thinkers right now that “being right never feels good.” Because real thinkers aren’t ranting about the anti-christ or how terrible the poor are. They are cautioning people about a cliff, and getting flipped off by the people ignoring them, right before careening off a cliff. I guess in our analogy the academics are in the car, but frat boys have the wheel and violently punish them if they try to take control.

      But if you can money your way into expertise, then your non-academic ambitions can finally be let loose, and we won’t think about the quarter million dead kids from Elon stopping hiv treatments earlier this year. If we don’t talk about it or doesn’t exist.

      We can actually map probability in complex systems where high dissonance (expected free energy) occurs and could be reduced. This would make our social body much more hack resistant.

      But it feels better to say “fuck that, aliens built the pyramids, the earth is flat, santa is real, and my dead dog is trying to play with me every time wind blows through my window.” All are really rewarding things to believe, but exist in high dissonance, and expecting others to give them weight is not healthy for anyone. Denying any of these to a large enough group stops being ‘calling out delusion,’ and becomes ‘being mean and bad and evil, and hating fun and happiness. Nobody else saw that cliff sign, so you’re obviously wrong.’ See ‘the double empathy problem’. For how autistic people experience the same socialized gatekeeping of reality, even against strong evidence.

      Nestle those beliefs within stable social scripts (script theory is cool) and you have people keeping up social appearances while holding such delusional beliefs.

      AI tech is far from the only problem exacerbated by this. caring or talking about this general problem makes you a social villain, when people could spend their energy on fun socialized preferences in a bubble safe from any questioning or responsibility.

      It’s frustrating, and I’m tired of people IRL who are baffled about me spending time studying academic content rather than running faster on the socio-economic treadmill. Apparently being isolated and focused on individual benefit is hypernormal.

      Seeing social waves overpower all actual thought in the social sphere is heartbreaking to every academic I know.

      USA HHS is running the wakefield rhetoric, which is basically saying “fuck science, we will make up and run with whatever we want. And the masses have been kept ignorant enough to think that is cool.”

      So yes, absolutely, the issue is critical thinking skills. The issue is getting at the people who are already affected, and being taught by elon and such that “critical thinking” means running your bubbles’ social narrative against those evil progressives no matter what. Opportunists will always warp words and reality if they are allowed to dictate beliefs for their bubble.

      This is a simplified summary, but I already expect to have lost a good chunk of audience for requiring too much energy. People who are trying are burning themselves out trying to affect a wall of celebrated ignorance.

  • Bloefz@lemmy.world
    link
    fedilink
    English
    arrow-up
    42
    arrow-down
    3
    ·
    12 hours ago

    I work with AI and use it personally, but I have my own servers running local models which solves tons of privacy concerns. The inaccuracy is another problem but not a big one for me as I know it and will simply fact check. Also, I don’t really use it for knowledge anyway. Just to filter news to my interest, help with summaries and translation etc.

    People use AI as some all-knowing oracle but an LLM is not meant for that at all.

    • Infynis@midwest.social
      link
      fedilink
      English
      arrow-up
      7
      arrow-down
      1
      ·
      7 hours ago

      There are definitely things AI is good for. Archival search is obviously the biggest, because that’s what we’ve been using it for decades. It can also be helpful for subterranean and medical imaging, and art restoration. But the companies selling it want to sell a Magic 8 Ball with ads

    • Ex Nummis@lemmy.world
      link
      fedilink
      English
      arrow-up
      15
      ·
      11 hours ago

      This is the correct way to use it. In a field you are already very knowledgeable in, so you can do your own fact-checking. This is absolutely paramount. But most people are content to just copy-paste and don’t even ask the llm for sources.

      • Peanut@sopuli.xyz
        link
        fedilink
        English
        arrow-up
        2
        ·
        3 hours ago

        Same way they treat social information. Reminder that USA HHS is running wakefueld rhetoric. As we have more thoroughly proven that the vaccine autism connection was not actual science, it has grown more and more socially, because most people seem comfortable completely untethered from any scientific thinking. Treat AI like you would a social body, and do both things with actual bayesian weighting, adjusted and corrected through diverse empirical bodies of knowledge. Not ignoring dissonance because it’s more comfortable to do so.

        More should be actively investing into active learning, because if you aren’t actively learning, you might as well be chatgpt running with any confabulation you’ve already conjured… Like those people being confirmed into psychosis.

      • Bloefz@lemmy.world
        link
        fedilink
        English
        arrow-up
        10
        ·
        8 hours ago

        I have one server with a cheap MI50 instinct. Those come for really cheap on eBay. And it’s got really good memory bandwidth with HBM2. They worked ok with ollama until recently when they dropped support for some weird reason but a lot of other software still works fine. Also older models work fine on old ollama.

        The other one runs an RTX 3060 12GB. I use this for models that only work on nvidia like whisper speech recognition.

        I tend to use the same models for everything so I don’t have the delay of loading the model. Mainly uncensored ones so it doesn’t choke when someone says something slightly sexual. I’m in some very open communities so standard models are pretty useless with all their prudeness.

        For frontend i use OpenWebUI and i also run stuff directly against the models like scripts.

          • Bloefz@lemmy.world
            link
            fedilink
            English
            arrow-up
            2
            ·
            1 hour ago

            Agreed. The way they just dumped support for my card in some update with some vague reason also irked me (we need a newer rocm they said but my card works fine with all current rocm versions)

            Also the way they’re now trying to sell cloud AI means their original local service is in competition to the product they sell.

            I’m looking to use something new but I don’t know what yet.

            • brucethemoose@lemmy.world
              link
              fedilink
              English
              arrow-up
              1
              ·
              edit-2
              60 minutes ago

              I’ll save you the searching!

              For max speed when making parallel calls, vllm: https://hub.docker.com/r/btbtyler09/vllm-rocm-gcn5

              Generally, the built in llama.cpp server is the best for GGUF models! It has a great built in web UI as well.

              For a more one-click RP focused UI, and API server, kobold.cpp rocm is sublime: https://github.com/YellowRoseCx/koboldcpp-rocm/

              If you are running big MoE models that need some CPU offloading, check out ik_llama.cpp. It’s specifically optimized for MoE hybrid inference, but the caveat is that its vulkan backend isn’t well tested. They will fix issues if you find any, though: https://github.com/ikawrakow/ik_llama.cpp/

              mlc-llm also has a Vulcan runtime, but it’s one of the more… exotic LLM backends out there. I’d try the other ones first.

              • Bloefz@lemmy.world
                link
                fedilink
                English
                arrow-up
                1
                ·
                30 minutes ago

                Thank you so much!! I have been putting it off because what I have works but a time will soon come when I’ll want to test new models.

                I’m looking for a server but not many parallel calls because I would like to use as much context as I can. When making space for e.g. 4 threads, the context is split and thus 4x as small. With llama 3.1 8b I managed to get 47104 context on the 16GB card (though actually using that much is pretty slow). That’s with KV quant to 8b too. But sometimes I just need that much.

                I’ve never tried the llama.cpp directly, thanks for the tip!

                Kobold sounds good too but I have some scripts talking to it directly. I’ll read up on that too see if it can do that. I don’t have time now but I’ll do it in the coming days. Thank you!

                • brucethemoose@lemmy.world
                  link
                  fedilink
                  English
                  arrow-up
                  1
                  ·
                  edit-2
                  7 minutes ago

                  Vllm is a bit better with parallelization. All the kv cache sits in a single “pool”, and it uses as many slots as will fit. If it gets a bunch of short requests, it does many in parallel. If it gets a long context request, it kinda just does that one.

                  You still have to specify a maximum context though, and it is best to set that as low as possible.

                  …The catch is it’s quite vram inefficient. But it can split over multiple cards reasonably well, better than llama.cpp can, depending on your PCIe speeds.

                  You might try TabbyAPI exl2s as well. It’s very good with parallel calls, thoughts I’m not sure how well it supports MI50s.


                  Another thing to tweak is batch size. If you are actually making a bunch of 47K context calls, you can increase the prompt processing batch size a ton to load the MI50 better, and get it to process the prompt faster.


                  EDIT: Also, now that I think about it, I’m pretty sure ollama is really dumb with parallelization. Does it even support paged attention batching?

                  The llama.cpp server should be much better, eg use less VRAM for each of the “slots” it can utilize.

      • brucethemoose@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        edit-2
        6 hours ago

        Bloefz has a great setup. Used Mi50s are cheap.

        An RTX 3090 + a cheap HEDT/Server CPU is another popular homelab config. Newer models run reasonably quickly on them, with the attention/dense layers on the GPU and sparse parts on the CPU.

    • Clanket@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      edit-2
      10 hours ago

      How do you know it’s doing any of this correctly, especially filtering and translations?

      • Bloefz@lemmy.world
        link
        fedilink
        English
        arrow-up
        6
        ·
        8 hours ago

        I mainly use it for Spanish which I have a basic proficiency in. It just accompanies me on my learning journey. It may be wrong sometime but not often. Like the other reply said, LLMs are good at languages, it’s what they were originally designed for until people found out they could do more (but not quite as well).

        And as for filtering, I just use it as a news feed sanitizer with a whole bunch of rules. It will miss things sometimes but it’s also my ruleset that’s not perfect. I often come across the unfiltered sources anyway and even if it misses something, it’s only news. Nothing really important to me.

        • porcoesphino@mander.xyz
          link
          fedilink
          English
          arrow-up
          2
          ·
          8 hours ago

          It’s funny, I had half been avoiding it for languages. I had lots of foreign friends and they often lived together in houses and those houses would almost have this creole. They came to learn English and were reinforcing their own mistakes but it was mutually intelligible so the mistakes were reinforced and not caught. I suspect LLMs would be amazing at doing that to people and their main use case along these lines seems like it would be to practice at a slightly higher level than you so I suspect some of those errors would be hard to catch / really easy to take as correct instead of validating

          • Bloefz@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            edit-2
            1 hour ago

            I don’t think that’s a problem. I live in Spain and speak Spanish daily with real people, many of them my friends. They’ll correct me if needed, they often do. Though most are my own mistakes.

            Don’t forget people give wrong answers too. But people aren’t available 24/7 to help me.

          • FauxLiving@lemmy.world
            link
            fedilink
            English
            arrow-up
            3
            arrow-down
            1
            ·
            7 hours ago

            Anyone learning a new language massively benefits from being able to speak with native speakers.

            That being said, LLMs are better at languages and translation tasks than any pretty much anything else. If you need vocabulary help or have difficulty with grammar they’re incredibly helpful (vs Googling and hoping someone had the same issue and posted about it on Reddit).

            I mean, if you can afford a native speaker tutor that is the superior choice. But, for the average person, an LLM is a massive improvement over trying to learn via YouTube or apps.

      • MagicShel@lemmy.zip
        link
        fedilink
        English
        arrow-up
        3
        arrow-down
        1
        ·
        edit-2
        9 hours ago

        Not OP, but…

        It’s not always perfect, but it’s good for getting a tldr to see if maybe something is worth reading further. As for translations, it’s something AI is rather decent at. And if I go from understanding 0% to 95%, really only missing some cultural context about why a certain phrase might mean something different from face value, that’s a win.

        You can do a lot with AI where the cost of it not being exactly right is essentially zero. Plus, it’s not like humans have a great track record for accuracy, come to think of it. It comes down to being skeptical about it like you would any other source.

        • porcoesphino@mander.xyz
          link
          fedilink
          English
          arrow-up
          2
          arrow-down
          1
          ·
          8 hours ago

          Strongly disagree with the TLDR thing

          At least, the iPhone notifications summaries were bad enough I eventually turned them off (but periodically check them) and while I was working at Google you couldn’t really turn of the genAI summaries of internal things (that evangelists kept adding to things) and I rarely found them useful. Well… they’re useful if the conversation is really bland but then the conversation should usually be in some thread elsewhere, if there was something important I don’t think the genAI systems were very good at highlighting it

          • ctrl_alt_esc@lemmy.ml
            link
            fedilink
            English
            arrow-up
            1
            ·
            4 hours ago

            Completely agree, those summaries are incredibly bad. I was recently looking for some information in Gemini meeting notes and just couldn’t find it, even though I was sure it had been talked about. Then I read the transcript itself and realised that the artificial unintelligence had simply left out all the most important bits.

          • brucethemoose@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            edit-2
            7 hours ago

            The iPhone models are really bad. They aren’t representative of the usefulness of bigger ones, and it’s inexplicably stupid that Apple doesn’t like people pick their own API as an alternative.

          • MagicShel@lemmy.zip
            link
            fedilink
            English
            arrow-up
            1
            ·
            7 hours ago

            You can disagree, but I find it helpful to decide whether I’m going to read a lengthy article or not. Also if AI picks up on a bunch of biased phrasing or any of a dozen other signs of poor journalism, I can go into reading something (if I even bother to at that point) with an eye toward the problems in an article. Sometimes that helps when an article is trying to lead you down a certain path of thinking.

            I find I’m better at picking out the facts from the bias if I’m forewarned.

          • FauxLiving@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            arrow-down
            1
            ·
            7 hours ago

            iPhone notification summaries were made with GPT3.5 I believe (maybe even the -turbo version).

            It doesn’t use reasoning and so when using very short outputs it can produce wild variations since there are not a lot of previous tokens in order to direct the LLM into the appropriate direction in kv-space and so you’re more at the whims of temperature setting (randomly selecting the next token from a SOFTMAX’d list which was output from the LLM).

            You can take those same messages and plug them into a good model and get much higher quality results. But good models are expensive and Apple is, for some reason, going for the budget option.

            • brucethemoose@lemmy.world
              link
              fedilink
              English
              arrow-up
              1
              ·
              3 hours ago

              AFAIK some outputs are made with a really tiny/quantized local LLM too.

              And yeah, even that aside, GPT 3.5 is really bad these days. It’s obsolete.

  • theneverfox@pawb.social
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    2
    ·
    7 hours ago

    I kind of think we should make it illegal to run LLMs as a service available to the public. They’re still cool behind the scenes, but it seems to only go bad when unprepared people talk to them

    You should at least understand them well enough to set them up yourself or have someone do it for you. At least that will demystify it a bit

    • AwesomeLowlander@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      1
      ·
      5 hours ago

      All it needs is enough tech ability to run a few docker commands. That has absolutely nothing to do with LLMs. Not to mention the horrifying intrusion on our rights by limiting what we can host.

      • theneverfox@pawb.social
        link
        fedilink
        English
        arrow-up
        3
        arrow-down
        1
        ·
        4 hours ago

        Yeah, that’s what I want them to do. I want them to install docker, run some commands, correctly identify their graphics card and driver, and get the networking right. Then I want them to pick the model they want and download it

        It’s not about making it hard, it’s about having a level of understanding about how this all works. If you’re using AI, I shouldn’t have to explain to you that it can’t search the Internet unless you manually give it that ability. I shouldn’t have to explain that the LLM is just a file, not some magical Internet djinn

        Not to mention the horrifying intrusion on our rights by limiting what we can host.

        Umm… You don’t have the right to host anything you want. You can’t host CSAM, copyrighted things, and you have to comply with tons of laws if you want user content

  • rayyy@piefed.social
    link
    fedilink
    English
    arrow-up
    6
    ·
    10 hours ago

    AI companies want to sell their product. They tell us all the good stuff. Consumers take the bait. Normal capitalism today.