cross-posted from: https://lemmy.world/post/37665245

https://www.cnet.com/tech/services-and-software/i-didnt-snap-this-photo-at-a-coffee-shop-microsofts-new-ai-tool-created-it/

You can also flag the ai images in the result.

Perfect solution? Not even close. Still...

-

you are viewing a single comment's thread
view the rest of the comments
[โ€“] 7 points 1 year ago (3 children)

Too bad it's probably not going to take a whole lot longer to get to the point where it's virtually impossible to tell. It's already difficult for many of us.

  • source
  • hideshow 3 child comments
  • [โ€“] 4 points 1 year ago (2 children)

    I'm hoping there'll be more indicators within the metadata of an image or video. If not the AI putting in AI copyright data or something, at least some tells from human made content like phone model or GPS data-- stuff that tends to get removed or hidden can at least verify to the service hosting the image that it's legit.

    Doesn't stop someone who really wants to fake it from falsifying that data but it certainly would help trim the ones who are just trying to profit off cheap, mass produced content.

  • source
  • parent
  • hideshow 2 child comments