• imadabouzu@awful.systems
    link
    fedilink
    English
    arrow-up
    36
    Ā·
    29 days ago

    Despite what the tech companies say, there are absolutely techniques for identifying the sources of their data, and there are absolutely techniques for good faith data removal upon request. I know this, because Iā€™ve worked on such projects before on some of the less major tech companies that make some effort to abide by European laws.

    The trick is, it costs money, and the economics shift such that one must eventually begin to do things like audit and curate. The shape and size of your business, plus how you address your markets, gains nuance that doesnā€™t work when your entire business model is smooth, mindless amotirizing of other peopleā€™s data.

    But I donā€™t envy these tech companies, or the increasing absurd stories they must tell to hide the truth. A handsome sword hangs above their heads.

    • OhNoMoreLemmy@lemmy.ml
      link
      fedilink
      English
      arrow-up
      20
      Ā·
      29 days ago

      The other reason they donā€™t do it is because many models are trained on a large corpus of pirated texts, and documenting this would be a confession.

      Not just in an ā€˜I scraped the new york times without permissionā€™ kind of way, but in a ā€˜I illegally downloaded a torrent containing bestsellers from the last 30 yearsā€™ kind of way.

      • Tar_Alcaran@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        11
        Ā·
        29 days ago

        Exactly. Itā€™s not that they canā€™t, or that itā€™s too expensive, itā€™s that doing so will reveal their crimes.

        • imadabouzu@awful.systems
          link
          fedilink
          English
          arrow-up
          9
          Ā·
          29 days ago

          In a sense, to me, it is the same thing. If your business is built upon repurposing everyone elseā€™s inputs indiscriminately to your benefit and their detriment, it is, too expensive, to reveal that simple truth.

      • Soyweiser@awful.systems
        link
        fedilink
        English
        arrow-up
        3
        Ā·
        27 days ago

        Bestsellers? There used to be torrents of basically all releases. My provider blocks torrent sites and I dont use a vpn so im not sure if people still do this, but downloading basically all books (in english) at once released in a certain period was possible

    • Rhaedas@fedia.io
      link
      fedilink
      arrow-up
      5
      Ā·
      29 days ago

      Wouldnā€™t removal of the data effect on the model require a basic retraining? A bit too late for all the open source ones out there.

      • imadabouzu@awful.systems
        link
        fedilink
        English
        arrow-up
        3
        arrow-down
        1
        Ā·
        edit-2
        28 days ago

        Thatā€™s a good question, because there is nuance here! Itā€™s interesting because while working on similar projects I also ran into this issue. First off, itā€™s important to understand what your obligation is and the way that you can understand data deletion. No one believes it is necessary to permanently remove all copies of anything, anymore than it is necessary to prevent all forms of plagairism. No one is complaining that is possible at all to plaigarise, weā€™re complaining that major institutions are continuing to do so with ongoing disregard of the law.

        Only maximalists fall into the trap that thinking of the world in binary sense: either all in or do nothing at all.

        For most of us, itā€™s about economics and risk profiles. Open source models get trained continuously over time, there wonā€™t be one version. Saying that open source operators do have some obligations to in good faith to curate future training to comply has a long tail impact on how that model evolves. Previous PII or plaigarized data might still exist, but its value and novelty and relevance to economic life goes down sharply over time. No artist or writer argues that copyright protections need to exist forever. They literally, just need to have survival working conditions, and the respect for attribution. The same thing with PII: no one claims that they must be completely anonymous. They just desire cyber crime to be taken seriously rather than abandoned in favor of one party taking the spoils of their personhood.

        Also, yes, there are algorithms that can control how further learning promotes or demotes growth and connections relative to various policies. Rather than saying that any one policy is perfect, a mere willingness to adopt policies in good faith (most such LLM filters are intentionally weak so that those with $$ and paying for API access can outright ignore them, while they can turn around and claim it canā€™t be solved too bad so sad).

        Yes. It is possible to perturb and influence the evolution of a continuously trained neural network based on external policy, and theyā€™re carefully lying through omision when they say they canā€™t 100% control it or 100% remove things. Fine. Thatā€™s, not necessary, neither in copyright nor privacy law. Never been.

        • David Gerard@awful.systemsOPM
          link
          fedilink
          English
          arrow-up
          8
          Ā·
          edit-2
          28 days ago

          Are you sure that meets the letter of the law? GDPR would say ā€œfuck that version of nuance, fix it.ā€ Microsoft now tries filtering on Bing Copilot in Germany, to variable results. What does the relevant California law say and mean?

          • imadabouzu@awful.systems
            link
            fedilink
            English
            arrow-up
            1
            arrow-down
            1
            Ā·
            28 days ago

            I am not a lawyer. But you wouldnā€™t be surprised to hear that

            1. I donā€™t have inside story of Bing in Germany. It could be that Microsoft either doesnā€™t want to do it well, or hasnā€™t yet done it well enough. Iā€™m not promising either in particular, but it can be done.
            2. Generally as an engineer you have a pile of options with trade offs. You absolutely can build nuanced solutions, as often the law and the lawyers live in nuanced realities. That is the reality of even the best sorts of tech companies who are trying.

            My commitment is that maximalism or strict binary assumptions wonā€™t work on either end and donā€™t satisfy what anyone truly wants or needs. If weā€™re not careful about what it takes to move the needle, we agree with them by saying ā€˜it canā€™t be done, so it wont be done.ā€™

            • V0ldek@awful.systems
              link
              fedilink
              English
              arrow-up
              8
              Ā·
              27 days ago

              My commitment is that maximalism or strict binary assumptions wonā€™t work on either end and donā€™t satisfy what anyone truly wants or needs.

              Whatā€™s truly lovely about GDPR is that it is maximalist, strict, and binary. For any ā€œbutā€¦ā€ of a corporation the GDPR answer is ā€œfucks given: 0, this is YOUR problem, comply or perish.ā€

              Which makes it so baffling every time a techbro fails to understand it or claims ā€œGDPR doesnā€™t apply to me.ā€ Just donā€™t fuck around with PII and donā€™t collect any without explicit permission from the user! How is this difficult?!

            • David Gerard@awful.systemsOPM
              link
              fedilink
              English
              arrow-up
              4
              Ā·
              28 days ago

              Iā€™m referring specifically to this where they could only put in a shaky bodge.

              When you donā€™t know an example, consider looking it up and not just waffling anyway.