> merely the next iteration of Photoshop I think you're significantly understating the impact here. You're talking mostly about content-creation, and in this

> merely the next iteration of Photoshop I think you're significantly understating the impact here. You're talking mostly about content-creation, and in this context there are a lot of positives; I certainly agree with your statement: > They're going to breathe new life into media and make it significantly easier to make films, music, and all kinds of art. BUT. I don't think you can, with a straight face, say that there are going to be no negative impacts of this technology. You make a claim that this is just like Photoshopping in the 90s, but I think the key difference here is that in the '90s, there was a fallback medium; when photos became unreliable, you still had video, which couldn't easily be faked. Looking forward to the next 5-10 years, how does anyone figure out what's true when photos, video, and voice can all be trivially faked by anyone? What reliably medium can we fall back to? This could lead to a real epistemological crisis! > The public will come to understand AI-generated media as a part of daily life. People are not that stupid. My problem with this position is that it hand-waves away some pretty fundamental changes to the fabric of our reality. How exactly do you propose that people are going to understand the world, post-'fakes? If any video you see in your twitter feed could be fake, how can anyone make any inferences about the world outside of their own immediate direct experience?