• 1 Post
  • 30 Comments
Joined 2 years ago
Cake day: July 11th, 2023
  • YungOnions@lemmy.worldtoTechnology@lemmy.zip•Director Admits "The Brutalist" Used AI to Alter Characters' AccentsEnglish
    72·
    7 hours ago

    The controversy surrounding the film kicked off last week after an interview with editor Dávid Jancsó was released by Red Shark News. Jancsó said that the production turned to software called Respeecher to clean up Brody and Jones’ dialog after the ADR stage — a process in which certain lines of dialogue are re-recorded in post — didn’t pan out, a decision that both leads were fully on board with, Jancsó said.

    Who cares? Seems like everyone involved was on board. Seems like a good use of the technology to me. I don’t see why anyone should be ‘admitting’ anything.

  • YungOnions@lemmy.worldtoNo Stupid Questions@lemmy.world•If AI spits out stuff it's been trained on
    7·
    3 days ago

    Sexton says criminals are using older versions of AI models and fine-tuning them to create illegal material of children. This involves feeding a model existing abuse images or photos of people’s faces, allowing the AI to create images of specific individuals. “We’re seeing fine-tuned models which create new imagery of existing victims,” Sexton says. Perpetrators are “exchanging hundreds of new images of existing victims” and making requests about individuals, he says. Some threads on dark web forums share sets of faces of victims, the research says, and one thread was called: “Photo Resources for AI and Deepfaking Specific Girls.”

    The model hasn’t necessarily been trained with CSAM, rather you can create things called LORAs which help influence the image output of a model so that it’s better at producing very specific content that it may have struggled with before. For example I downloaded some recently that help Stable Diffusion create better images of Battleships from Warhammer 40k. My guess is that criminals are creating their own versions for kiddy porn etc.