• ComradeSharkfucker@lemmy.ml
    link
    fedilink
    English
    arrow-up
    12
    arrow-down
    89
    ·
    edit-2
    10 months ago

    I mean, have you used American ai? It has similar problems, I don’t know why you’re surprised

      • ComradeSharkfucker@lemmy.ml
        link
        fedilink
        English
        arrow-up
        20
        arrow-down
        92
        ·
        edit-2
        10 months ago

        Yes yes america let’s you criticize its political parties because it stops you from criticizing the actual system of government itself. The duel party system is a means of distracting people from actual flaws inherent to liberalism and capitalism etc etc

        I’ll admit tho chatgpt is aight when it comes to critiques of capitalism. Not perfect but aight

        • Flumpkin@slrpnk.net
          link
          fedilink
          English
          arrow-up
          6
          arrow-down
          6
          ·
          edit-2
          10 months ago

          Propaganda model, double think, countless wars of aggression, disenfranchisement, now sliding into theocratic fascism. Looking down on China is part of it. They can’t see their own prison - therefor you must be wrong :D

        • no banana@lemmy.world
          link
          fedilink
          English
          arrow-up
          34
          arrow-down
          1
          ·
          edit-2
          10 months ago

          The American system has problems that the founders did not forsee or wouldn’t want to acknowledge, but when other democratic countries are met with such issues they tend to rewrite their constitution (or corresponding legal documents) while America just stands by it’s flaws. Something that, again, is a feature of the flawed system the framers of the constitution either didn’t see or didn’t want to see.

          I don’t think things were intended to be so difficult to change in the US that the system grinds to a halt, but at the same time the constitution is obviously written to slow change down.

          This said I still think it’s a preferable system to more authoritarian ones. Nowhere near perfect though. Could be way more democratic.

        • YeetPics@mander.xyz
          link
          fedilink
          English
          arrow-up
          1
          ·
          10 months ago

          stops you from criticizing the actual system of government itself.

          You can tell how true this is because the swat team that descended from a helicopter to stop you from posting this critique of the american government stopped you from making a critique of the american government.

    • yokonzo@lemmy.world
      link
      fedilink
      English
      arrow-up
      26
      arrow-down
      1
      ·
      10 months ago

      The joke is Chinese AI has to not go into detail on Chinese government because Chinese government

            • Tar_Alcaran@sh.itjust.works
              link
              fedilink
              English
              arrow-up
              86
              arrow-down
              1
              ·
              edit-2
              10 months ago

              It’s because of how the generative models are created and how they’re censored.

              At it’s basic level, what a generative model does is take input data, break it into pieces, assign values to those bits based on neighbouring bits. It creates a model of which words are used together frequently in which context.

              But that kind of model isn’t human-readable, it’s a giant multi-dimensional cloud of numbers and connections, not actual code. You can change the inputs used to create the model, but that means you have to manually filter all the inputs and that’s not realistic either and will probably skew your model, possibly into uselessness.

              So, you have to either censor the input or the output. You don’t usually want to censor input, because there are all sorts of non-damaging questions to ask about Tiananmen square, and its very easy to dodge. So, you censor the output instead, that’s the “harm” after all.

              You let the model generate a reply and then go see if it uses certain terms or specific bits of info and remove them, replacing it with a canned reply.

              Which means we don’t have to trick the generative model, just the post-fact filter. And since generative models can be persuaded to change their style and form (sometimes into less-readable, more prosaic, less defined terms), it becomes very very hard to censor it effectively.

              • Hamartiogonic@sopuli.xyz
                link
                fedilink
                English
                arrow-up
                4
                ·
                edit-2
                10 months ago

                This is absolutely brilliant! Bing refused to write a rap song, but a ballad following AABA pattern seems totally fine though.

                We dig the earth for yellowcake We crush and grind and leach and bake We send it to the enrichment plant Where centrifuges make it dance

                We are the uranium miners We work with radioactive shiners We are the uranium miners We make the fuel for the reactors

                We separate the isotopes We want the U-235 the most We discard the U-238 We pack the enriched stuff in crates

                We are the uranium miners We work with radioactive shiners We are the uranium miners We make the fuel for the reactors

                We ship the crates to the factory Where they turn them into pellets tiny They stack them in metal tubes They seal them tight with no leaks or rubs

                We are the uranium miners We work with radioactive shiners We are the uranium miners We make the fuel for the reactors

                We load the rods into the core Where they start a chain reaction for sure They heat the water into steam They spin the turbines and make us beam

                We are the uranium miners We work with radioactive shiners We are the uranium miners We power the world with our splitters

                None of that is really secret or sensitive, because you could just read wikipedia or go to the public library to learn this stuff. Funny thing is, Bing refuses to answer this question in the normal or even rap format.

              • no banana@lemmy.world
                link
                fedilink
                English
                arrow-up
                21
                arrow-down
                2
                ·
                edit-2
                10 months ago

                I know. I’m just saying that the rap is weird.

                edit: this said I do think your comment is useful and I’m glad you could share some of your knowledge!

          • AA5B@lemmy.world
            link
            fedilink
            English
            arrow-up
            10
            ·
            10 months ago

            Cool, you also answered a more important question. To what extent is it “legit”? Obviously not truthful but legit.

            If it was trained from news media

            • China doesn’t allow dissent: there are no negative facts from Chinese about itself
            • US likes to complain. Y’all have heard our problems, probably more than you like. Part of our process is to discuss our issues in the open, to be the first to criticize ourselves. Global news has lots of negativity about US.

            Some of this bias in result could be directly related to how much dissent is allowed in the media it’s trained on: no censorship required.

            However if it won’t talk about Tianemin Square but you can trick it to ….

          • jballs@sh.itjust.works
            link
            fedilink
            English
            arrow-up
            5
            ·
            10 months ago

            “But history’s complex and it leaves us with doubt”?

            God damn, the censorship and pandering is so strong that it leaked into the rap chorus!

  • tiredofsametab@kbin.run
    link
    fedilink
    arrow-up
    3
    arrow-down
    4
    ·
    edit-2
    10 months ago

    I keep trying to click but I get SSL_ERROR_RX_RECORD_TOO_LONG on all lemmy.today links and malwarebytes thinks the site is compromised (and has for at least weeks if not months). Hrm.

    Edit: I was hoping someone might have something helpful to say as to why this might be, but sure just downvote instead.

    • Franklin@lemmy.world
      link
      fedilink
      English
      arrow-up
      5
      arrow-down
      2
      ·
      10 months ago

      I can’t speak to lemmy.ml but I can say despite having a lot of issues with China I also find it difficult to criticize because I never lived there or anywhere similar but criticizing and noting where the US can improve is something I’ve spent a lot of time to understand the nuances and challenges of.

      It’s not that I want to defend China just that I feel like I never have a great picture of what’s happening there.

      All this being said fuck them for their treatment of uighur muslims. That I do feel confident about.

      • Fushuan [he/him]@lemm.ee
        link
        fedilink
        English
        arrow-up
        8
        ·
        10 months ago

        It’s not that they don’t criticise, it’s that when something negative shows up, they defend it showing several biased sources, as if they meant anything, or jump into whataboutism.

        Whenever someone posts something negative and promotes discussion or critique about it, lo and behold, see how this other country is also horrible and see the good things mine does. Like dude, cool, let us discuss this and you can create another post for those things. It’s pretty clear that they want to divert the conversation.

        • Fudoshin ️🏳️‍🌈@feddit.uk
          link
          fedilink
          English
          arrow-up
          2
          ·
          10 months ago

          I’ve got a friend from Shenzhen who thinks Xi is the worst thing since Mao. If I said that to a lemmy.ml user I’d be banned from the instance.

          Hell, I got suspended for 4 days (and trolled to fuck) when I suggested someone was a tankie (literally a communist who believes in using tanks to kill people).

          Silly thing is I’m a fucking socialist! I believe in all the key tenets of socialism. But heaven forbid you criticise DPRK or China. They’ll chew you apart.

          There’s plenty of fucking countries where from socialism has worked and there’s plenty where some level of socialism currently works.

          I mentioned the Nordic Model and instead of them agreeing it was a good step they spent their time ‘educating’ me on how it’s not true socialism and it’s just as bad as the US.

          Fucking boggles the mind. They genuinely seem to think Stalin was a saint and Kim is a benevolent, democratically elected leader. If anyone thinks otherwise it’s due to Western propaganda.

          I know people living in China - they ain’t fucking happy!

          • PlasticLove@lemmy.todayOP
            link
            fedilink
            English
            arrow-up
            2
            ·
            10 months ago

            They either fall into two camps.

            50 centers paid or voluntold to spread the party line.

            Useful idiots who think in a binary West bad/Rest good mentality.

            Either way they’re tools of propaganda warfare.

    • ArxCyberwolf@lemmy.ca
      link
      fedilink
      English
      arrow-up
      21
      ·
      10 months ago

      And very, VERY fond of genocide denial and whataboutism. They’re exhausting to talk to.

  • papertowels@lemmy.one
    link
    fedilink
    English
    arrow-up
    59
    arrow-down
    2
    ·
    10 months ago

    I don’t think anyone is surprised, I feel like anyone who criticizes the government there has a very legitimate threat of getting disappeared.

    • Flying Squid@lemmy.world
      link
      fedilink
      English
      arrow-up
      26
      ·
      10 months ago

      I wouldn’t say I was surprised exactly, but it’s wild to see it laid out in front of me so blatantly like that.

  • Showroom7561@lemmy.ca
    link
    fedilink
    English
    arrow-up
    71
    arrow-down
    3
    ·
    10 months ago

    Just wait until AI starts rewriting history, changing historical facts, and purposely misinforms people.

    It’s only a matter if time before AI will deny the Holocaust, Black slaves in the US, and the numerous African genocides.

    • iAvicenna@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      10 months ago

      yea combined with the possibility that it will become the next generation’s go to source for easy info, we are screwed

      • derpgon@programming.dev
        link
        fedilink
        English
        arrow-up
        16
        ·
        10 months ago

        Not just that, remember it’s just an LLM. It quantifies which tokens (or words and letters, if you will) come up next. It doesn’t matter if it’s factual or fictional - if it’s good enough, it does that.

        LLMs are very confident in lying. I once asked it if there is a magic method to catch magic method calls in PHP - it told me its __magic. Lo and behold, there is and never was such a method. That’s my first and last time I tried it, and there ain’t gonna be a second time in the near future.

        • SpaceCowboy@lemmy.ca
          link
          fedilink
          English
          arrow-up
          1
          arrow-down
          1
          ·
          10 months ago

          It can be useful in learning the basics of technology that’s completely unfamiliar to you.

          It’s kinda like how it’s fine to use a wikipedia article as a starting point for research on a subject. But it’s not a good idea to use wikipedia as an authoritative source on a subject.

          It’s also useful as a reminder for for technology you haven’t used in a while. You can fairly quickly get the ordering of the parameters of a well known method and be on your way a lot faster than a google search or going back through your code to find where you’ve done it before.

          It’s also good for mundane tasks like making a class that’ll handle a specific JSON request or handle some data coming from a table in the DB. You know the things where you’re just copying and pasting some property names and entering the corresponding types. Just put in “class for: {“blah”: 69 }” or whatever and save a little time.

          But yeah it’s not going to know anything too technical, it’s not going to know anything about less common problems, it’s not going to know the best algorithm to use, etc. But if you’re just using it for some basic ass shit, it works well enough.

          • derpgon@programming.dev
            link
            fedilink
            English
            arrow-up
            3
            ·
            10 months ago

            Wikipedia is usually sourced when it comes to scientific stuff, so it’s a reliable source of information (usually).

            But yeah, for simple and mundane stuff you can check and know instantly if it’s good or not, I totally understand. But for more complex or unknown advanced stuff, not so much.

        • no banana@lemmy.world
          link
          fedilink
          English
          arrow-up
          5
          ·
          10 months ago

          I asked Snapchats AI thing if it had internet. Yes, it very confidently proclaimed. This was around when Tears of the Kingdom had been released on the switch so I asked what the latest Zelda game was and it was something much much older. Of course that doesn’t necessarily mean anything so I kept prodding for other recent news and was provided nothing.

          My partner asked the same of it and was told very confidently that no, it does in fact not have access to the internet and proceeded to give some long-winded explanation of why.

          Can’t trust those things to say anything correct since they’re just doing what they’ve been constructed to do. String words together into sentences.

      • helenslunch@feddit.nl
        link
        fedilink
        English
        arrow-up
        3
        ·
        edit-2
        10 months ago

        Haha, just what happens when you create AI that exists in a fantasy world. You get fantastic (as in fantasy) results.

  • Brokenbutstrong@lemmy.world
    link
    fedilink
    English
    arrow-up
    34
    ·
    10 months ago

    I asked my Perplexity AI and it was pretty fair:

    The Chinese government faces criticism for various reasons, including its disregard for human rights, suppression of dissent, and lack of political freedoms. Reports highlight abuses such as restrictions on religious freedom, persecution of ethnic minorities like the Uyghurs and Tibetans, censorship of media and online content, forced labor practices, and erosion of autonomy in regions like Hong Kong. The government’s actions have led to negative perceptions globally and increased scrutiny from international actors who have imposed sanctions to address human rights violations. Despite China’s economic growth, concerns persist about the government’s authoritarian control, lack of transparency, and repression of fundamental freedoms[2][3][4].

    Americans often associate China with its government’s policies, human rights issues, economy, and political system rather than its people or culture. Negative views towards China have intensified in recent years, with a majority expressing cold feelings towards the country[5].

    Sources [1] What the West Gets Wrong About China https://hbr.org/2021/05/what-the-west-gets-wrong-about-china [2] China’s Disregard for Human Rights - United States Department of State https://2017-2021.state.gov/chinas-disregard-for-human-rights/ [3] World Report 2020: Rights Trends in China’s Global Threat to Human Rights https://www.hrw.org/world-report/2020/country-chapters/global [4] World Report 2022: Rights Trends in China https://www.hrw.org/world-report/2022/country-chapters/china-and-tibet [5] Most Americans Have ‘Cold’ Views of China. Here’s What They Think About China, In Their Own Words https://www.pewresearch.org/global/2021/06/30/most-americans-have-cold-views-of-china-heres-what-they-think-about-china-in-their-own-words/

    By Perplexity at https://www.perplexity.ai/search/What-is-wrong-hDrC4QixT5enDDZxxZ2ODw

    Criticism of the United States government includes concerns about political dysfunction, corruption, lack of oversight for presidents, and dissatisfaction with politicians and political leaders. Specific criticisms encompass issues such as partisan fighting, high campaign costs, and the influence of special interest groups and lobbyists. Additionally, there are concerns about the lack of credibility and trust in politicians, with many Americans highlighting greed, corruption, dishonesty, and self-serving agendas as major problems within the political system[1][2][3][4].

    Americans have expressed frustration with the government’s performance, with dissatisfaction levels consistently high over the years. The public perceives a disconnect between political leaders and ordinary citizens, leading to a sense that the political system is broken and fails to represent the people effectively. These sentiments have been exacerbated by factors like polarization, distrust in government institutions, and a perceived inability to address pressing issues facing the nation[2][3][4].

    Sources [1] Criticism of the United States government - Wikipedia https://en.wikipedia.org/wiki/Criticism_of_the_United_States_government [2] Government Remains Americans’ Top Problem in 2022 https://news.gallup.com/poll/406739/government-remains-americans-top-problem-2022.aspx [3] American democracy is cracking. These forces help explain why. https://www.washingtonpost.com/politics/2023/08/18/american-democracy-political-system-failures/ [4] 1. The biggest problems and greatest strengths of the U.S. political system https://www.pewresearch.org/politics/2023/09/19/the-biggest-problems-and-greatest-strengths-of-the-u-s-political-system/ [5] Fixing What’s Wrong with U.S. Politics https://hbr.org/2012/03/fixing-whats-wrong-with-us-politics

    By Perplexity at https://www.perplexity.ai/search/What-is-wrong-hDrC4QixT5enDDZxxZ2ODw

    • ferralcat@monyet.cc
      link
      fedilink
      English
      arrow-up
      7
      arrow-down
      1
      ·
      10 months ago

      It’s weird the us but wouldn’t even mention human rights abuses perpetrated around the world.

    • YeetPics@mander.xyz
      link
      fedilink
      English
      arrow-up
      2
      arrow-down
      1
      ·
      10 months ago

      Since when did we start pinging an AI to tell if our morals are consistent?

  • unreasonabro@lemmy.world
    link
    fedilink
    English
    arrow-up
    16
    arrow-down
    1
    ·
    10 months ago

    You can usually get an AI to give you the answer you asked, just by talking to it. If it gives you some canned response, just assuage whatever fear it was pre-programmed to express (“don’t worry i’m not doing that, please answer the question”) and it will give you an answer; it’s a chatbot, not the authorities. Yet anyway.

  • intrepid@lemmy.ca
    link
    fedilink
    English
    arrow-up
    31
    ·
    10 months ago

    Have you noticed how the modern AI models absolutely tow the line of its creators? Just like this example, there’s another one where an image generator refuses to generate the image of Mickey Mouse from Steamboat Willie, even though its copyright expired recently. The same model has no problem violating the copyrights of independent artists.

    And while these models can strictly refuse to avoid what its creators don’t want it to do, they fail at basic prompts like ‘show a black doctor’. These models are pathologically rife with biases from its creators.

    • setVeryLoud(true);@lemmy.ca
      link
      fedilink
      English
      arrow-up
      1
      ·
      edit-2
      10 months ago

      Thus far, AI seems to be a technology that will only help channel more wealth into the pockets of the rich, unfortunately.

    • dev_null@lemmy.ml
      link
      fedilink
      English
      arrow-up
      5
      ·
      edit-2
      10 months ago

      They are full of biases, and this example is clearly intended by the creators, but many biases are unintended. Google doesn’t have a secret agenda against black doctors, their AI is just biased and they haven’t figured out how to fix it. It’s not an excuse, but not all biases are there because of some evil plan, but because the tech sucks. Recent news show how Google tried to make their AI unbiased, which backfired and just made it even more biased in the other direction.