3 ms·
This. Even the current variant with generative AI in the mix isn't unexpected since the moment Musk tried to argue that basically video evidence as a category s
by xg15 3y ago
This. Even the current variant with generative AI in the mix isn't unexpected since the moment Musk tried to argue that basically video evidence as a category should be disallowed as, in theory, any video everywhere could be fake.
Thankfully he got laughed out of the courtroom for that. Yet.
https://www.npr.org/2023/05/08/1174132413/people-are-trying-to-claim-real-videos-are-deepfakes-the-courts-are-not-amused https://www.npr.org/2023/05/08/1174132413/people-are-trying-...
- AnthonyMouse 3y agoElon Musk exaggerates because he's Elon Musk, but it is generally a problem. If you have a clip of someone saying something in a news story from 2003, on the website of a news organization, with a publication date in 2003, to claim it's a deepfake would require you to claim the news organization forged the publication date. Every news organization that contemporaneously published the same video. But if we don't have reliable means to identify deepfakes (the high quality ones, not just the bad ones), well, sometimes the video in question wasn't widely published in a year before deepfakes were a thing. So what now? If you can't tell the difference in a large number of cases then your choices are admitting the forgeries into evidence or excluding the legitimate videos.