It's a lot more complicated than that due to edge cases like format conversion, resolution downscaling, chunking, etc but people are working on a solution in th

It's a lot more complicated than that due to edge cases like format conversion, resolution downscaling, chunking, etc but people are working on a solution in this vein: https://contentauthenticity.org/ https://contentauthenticity.org/ Full-disclosure: I work (on completely unrelated stuff) at one of the organizations involved