Close Menu
    Technik

    Wenn Big Tech sich darum kümmern würde, KI-Schwachsinn zu bekämpfen, würde uns das nicht darin ertränken / Es ist schwieriger, ein Chaos zu beseitigen, das man immer noch aktiv anrichtet.


    Wenn Big Tech sich darum kümmern würde, KI-Schwachsinn zu bekämpfen, würde uns das nicht darin ertränken / Es ist schwieriger, ein Chaos zu beseitigen, das man immer noch aktiv anrichtet.

    https://www.theverge.com/ai-artificial-intelligence/882956/ai-deepfake-detection-labels-c2pa-instagram-youtube

    View 13 Comments

    13 Kommentare

    2. Secure-Address4385 on

      You can’t sell shovels all day and then complain about the dirt.

    3. jason_mo on

      AI companies are fighting AI slop the way oil companies are fighting climate change.

    4. rnilf on

      There’s no way the brain-dead people who actually use these swiping social media apps will actually do anything, even if AI metadata is somehow accurately presented to them.

      They’re being spoonfed content like babies, and you’re expecting the baby to ask if the food has gluten in it or if it’s vegan before their parent shoves it into their mouths.

      The only way AI generated content doesn’t hit their feed is if the social media companies themselves actually filter out the AI content on their end, which they never will because that’ll be a big hit to engagement and their own investments into AI.

    5. Unable-Pomelo4040 on

      The business model is engagement. Slop just happens to scale better than quality.

    6. crazycatlady331 on

      Big Tech is run by some truly evil people. As in movie/comic book villain level evil.

    8. CopiousCool on

      I think they’re pushing the image manipulation side of AI to keep people engaged so that they can keep the ball rolling until it’s competent for other fields.

      Part of the problem is that it’s far easier to copy and merge images than it is to weigh the value of information relevant to a question or it’s answer so that may never actually come, hence the seemingly endless amount of slop years after release yet code production still struggles (OOP especially for example).

    9. UniverseShot on

      The problem is the middle management in big tech. They care about their nepo babies that might understand less computer science than even they do.

      The best way to subvert trust and control in the industry is to create a situation where workers can be blamed for intentional sabotage against trust in AI.

      The risk is that we will end up a generation behind, because the people hired to „fix“ this will have no idea how to architect AI solutions.

    11. Wormser on

      If in fact, slop wins and the majority of content — written, photographic, video — is AI, how long can this command our attention? One of the central premises of the web was the ability to connect with/consume things produced by HUMANS.

      Current state: we are arguing about what is real and what is slop. This means we assume at least some or a lot of what we see is produced by meat and not machine. Why bother trying to figure it out unless you care?

      Future state: AI wins and most content is machine made and importantly, people know this. Will the audience consume this content at the same rate as it does now? What happens to the attention economy when what we know what are consuming is AI generated vs posted by humans?

    12. am_reddit on

      Remember when all the big internet companies were united in their opposition to spam?

      Now they’re all in the spam-creation business.

    13. tekprodfx16 on

      Everyone on Reddit vastly overestimates people’s love for AI slop. It’s not going away. This is only the beginning of AI slop so learn to love it or just needlessly complain and suffer like you already are 

    Leave A Reply