Three raccoons in a trench coat. I talk politics and furries.

Other socials: https://ragdollx.carrd.co/

  • 2 Posts
  • 31 Comments
Joined 1 year ago
cake
Cake day: June 20th, 2023

help-circle


  • And social media websites that allow porn (Twitter, Reddit, Lemmy, Tumblr)

    Or just messaging apps where people send each other porn anyways (Whatsapp, Discord, Telegram)

    Horny teens have always found ways to get their hands on porn, even before the internet. Studies from the 80s and 90s found that the same percentage of boys had accessed porn as more recent studies where the internet has already become widespread:

    A study by Bryant [13] indicated that by the age of 15 years, 92% of boys had looked at or read Playboy, with average age of first exposure reported to be 11 years. Similarly, in regard to X-rated films, 92% of 13- to 15-years-olds reported that they had seen such a film. Most undergraduate men, when asked whether they have used sexual media in the past year, answer in the affirmative, although results vary between studies [14,15]. Studies in other cultures have found similar exposure patterns as in America. A recent study of 517 young men aged 18 to 27 years in Hong Kong indicated that 15.8 years was the average age of the first purchase of sexual media; just over half reported having used sexual media in the past 6 months [16]. The overall prevalence of having read, watched, or listened to sexual media in this sample was 91% (see Pan [17] for a lower prevalence estimate among Chinese young adults).










  • This did happen a while back, with researchers finding thousands of hashes of CSAM images in LAION-2B. Still, IIRC it was something like a fraction of a fraction of 1%, and they weren’t actually available in the dataset because they had already been removed from the internet.

    You could still make AI CSAM even if you were 100% sure that none of the training images included it since that’s what these models are made for - being able to combine concepts without needing to have seen them before. If you hold the AI’s hand enough with prompt engineering, textual inversion and img2img you can get it to generate pretty much anything. That’s the power and danger of these things.


  • IIRC it was something like a fraction of a fraction of 1% that was CSAM, with the researchers identifying the images through their hashes but they weren’t actually available in the dataset because they had already been removed from the internet.

    Still, you could make AI CSAM even if you were 100% sure that none of the training images included it since that’s what these models are made for - being able to combine concepts without needing to have seen them before. If you hold the AI’s hand enough with prompt engineering, textual inversion and img2img you can get it to generate pretty much anything. That’s the power and danger of these things.









  • In early 2019 bittorrent’s website views fluctuated between ~6M to ~9M. Now it’s around 3M to 4M.

    In early 2019 utorrent’s visits fluctuated between ~26M to ~75M. Now it sits around 25M to 21M.

    The fact that there were far more captures in early 2019 for both of them might be an indication that this was their peak, and while visits have reduced since then they’re far from dying.

    Streaming services may be part of the reason, though I also think it’s because many games and software have switched to freemium & microtransactions so spending money is optional, along with the fact that free and open source alternatives to mainstream software have become more robust and popular. When I was a kid I torrented Sony Vegas, but now that’s simply not necessary since we have DaVinci Resolve.