• NotMyOldRedditName@lemmy.world
    link
    fedilink
    arrow-up
    20
    ·
    2 days ago

    Seems fair to me. A search engine’s whole purpose is to surface real information, not made up stuff.

    Its very different than a chat bot, which also needs better rules around it, but I’d definitely hold search to a higher standard.

  • Gen_Euffe@sh.itjust.works
    link
    fedilink
    arrow-up
    72
    ·
    2 days ago

    Nice. Now, Google will probably just disable AI overview in germany cause they know they can’t make it not spew made up crap. The hope’s on more countries following suit on this.

    • SaharaMaleikuhm@feddit.org
      link
      fedilink
      arrow-up
      7
      ·
      2 days ago

      Don’t care about Google Search as I haven’t used that in 1-2 years, but I assume this will also apply to any other search engine.

      • Gen_Euffe@sh.itjust.works
        link
        fedilink
        arrow-up
        1
        ·
        2 days ago

        Similar. I’ve SearXNG set up on my main tower, and I preferably use ddg when I’m away. It’s nice that DDG’s not as intrusive with their AI, but I’d assume they’ll also have to abide

  • leadore@lemmy.world
    link
    fedilink
    arrow-up
    94
    ·
    2 days ago

    Correct and common sense ruling. The tech broligarchs can’t buy the judiciary in other countries like they do here in the US.

  • molave@reddthat.com
    link
    fedilink
    arrow-up
    10
    ·
    1 day ago

    Makes sense that if you use AI, it implies that what you put out using it, you’ve signed off on it.

  • Wilco@lemmy.zip
    link
    fedilink
    arrow-up
    5
    ·
    1 day ago

    This makes perfect sense. If you use AI to write a court document it has been set as precedence in multiple countries that the attorney signing the document is responsible for the content even if AI makes stuff up (which it does a lot).

    • merc@sh.itjust.works
      link
      fedilink
      arrow-up
      11
      ·
      2 days ago

      I hope for the one-two punch of:

      1. they’re legally responsible for anything the slop machine says
      2. the slop is machine generated, so it isn’t copyrightable

      Also, we got a third benefit if this ruling too. Google said, on the record, that it shouldn’t be responsible because you can’t trust anything that the slop-generator generates. Having that on the record could also be useful in a lot of other lawsuits.

      • Pommes_für_dein_Balg@feddit.org
        link
        fedilink
        arrow-up
        14
        ·
        2 days ago

        And according to German law, only if you had actual damages from it.
        And you don’t get a payout that’ll make you rich, you’ll just get awarded a financial amount that covers those damages.
        (After winning in court against Google’s lawyers, of course)

  • OwOarchist@pawb.social
    link
    fedilink
    English
    arrow-up
    30
    ·
    2 days ago

    I should do a bunch of google searches about myself and see if I can get its AI to say anything defamatory about me…

      • OwOarchist@pawb.social
        link
        fedilink
        English
        arrow-up
        3
        ·
        2 days ago

        …in Germany.

        In theory, supposing you could find a judge that wasn’t too corrupt, it should play out the same way in the US. What can Google possibly hide behind? It’s their in-house AI posting things on their own website. Who else can they point their finger at? Their only real defense would be to throw money at the problem.

        • Log in | Sign up@lemmy.world
          link
          fedilink
          arrow-up
          3
          ·
          1 day ago

          Hahaha! No! It’s the USA!

          Free speech (or at least the American version of “I shouldn’t face any consequences for saying nasty things”, rather than “the government can’t put me in prison for disagreeing with them”) is sacrosanct, or at least it is for wealthy and important people, which in the USA includes multinational megacorps like Google.

          So no, don’t expect a US court to hold a US tech giant to account for saying a thing. The courts have been writing exemptions for Google from liability for decades. Why would they stop now?

  • derAbsender@piefed.social
    link
    fedilink
    English
    arrow-up
    10
    arrow-down
    3
    ·
    2 days ago

    “because it alone has influence over the AI’s offering and the algorithms with which the AI operates.”

    Okay so anything someone generates with the AI is owned by the company?

    • ℍ𝕂-𝟞𝟝@sopuli.xyz
      link
      fedilink
      English
      arrow-up
      21
      ·
      2 days ago

      It’s like speeding with a stolen car.

      You are responsible for the speeding fine, but it does not mean the car is yours.

      • derAbsender@piefed.social
        link
        fedilink
        English
        arrow-up
        1
        ·
        2 days ago

        So the car is the thing the AI generated and the speeding fine is the mistake the AI generated content has?

  • DandomRude@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    arrow-down
    3
    ·
    2 days ago

    Well, I’m afraid that, unfortunately, this ruling will ultimately never be enforced, as it would make it impossible for the powerful multibillion-dollar corporations to operate this business.

    Another example of just how influential these corporations are is the fact that the most profitable piracy operation in history is even possible. Gigantic corporations of major influence have already filed multiple lawsuits over blatant intellectual property theft—yet even these players clearly stood no chance against the concentrated capital that is preventing the AI bubble from bursting immediately.

    Therefore: It is the right decision, but it will almost certainly remain without consequence.

    • quick_snail@feddit.nl
      link
      fedilink
      arrow-up
      5
      ·
      2 days ago

      Nah, it’s like how they fined them hundreds of millions of euros because it was 2 clicks on YouTube to block cookies (the law required 1 click).

      The result was that EU visitors to YouTube got 1 click “deny all” button for cookies. Else they’d be fined over-and-over.

      The same will happen here. Google will remove AI summaries in Germany. And the German people will be less stupid because of it.

      Hopefully this spreads to the whole EU, and then other progressive countries too

      • luciferofastora@feddit.org
        link
        fedilink
        arrow-up
        3
        ·
        2 days ago

        And the German people will be less stupid because of it.

        We’ll need a lot lore than that to save us, but it’s a start

      • DandomRude@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        2 days ago

        I don’t think this can be handled as easily as with cookies or third-party connections that require consent, because Google apparently plans to eventually replace the traditional search engine results pages (SERPs) entirely with an AI chat interface.

        In that case, they would have to offer a completely different product just for Germany, or leave this market altogether. Neither option is likely to be viable for Google.

        Setting up the AI chat so that it doesn’t provide dangerous answers is, in my opinion, also out of the question, since this can never be guaranteed due to the way LLMs work.

        • The_Decryptor@aussie.zone
          link
          fedilink
          English
          arrow-up
          2
          ·
          edit-2
          1 day ago

          In that case, they would have to offer a completely different product just for Germany, or leave this market altogether. Neither option is likely to be viable for Google.

          Might be only Germany for the moment, but I wouldn’t be surprised if you see more jurisdictions ruling this way, e.g. there’s a case in Canada at the moment where it claimed a musician was a sex offender.

          And locally, in Australia we don’t have the US style “safe harbor” protections, so companies are already liable for content shown on their sites regardless of who creates it. If courts already won’t accept “But somebody else wrote that” as a defence, they won’t accept it when it’s the companies own products producing it.

        • quick_snail@feddit.nl
          link
          fedilink
          arrow-up
          5
          ·
          2 days ago

          If they did that, then they’ll be fined constantly.

          The point is that they shouldn’t be doing that. We, as a international community, need to make that illegal.

          And this is the first stone for that.

          • DandomRude@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            2 days ago

            Yes, absolutely. I completely agree with you, but unfortunately, for the reasons mentioned, I still think it’s unlikely that this welcome ruling will have any real impact.

            Please don’t get me wrong. I’m just being realistic. This is by no means an argument against holding the tech giants accountable—we absolutely should!