That's what I was thinking, too. Given that each frame is about 42ms, that's not a trivial oversight. Much easier would be to use stenography to slightly manipulate the color of sets of pixels in the video. Far more information available to mess with, and extremely easy to hide (do it in various corners of various frames, keep the secret carefully guarded, or only let parts of the "secret" out to different groups, so that no one person knows all of the manipulation information, except, perhaps a head of security coordinating the entire thing)

came here to say the same thing.

the same ideas apply to frames, if you drop one from a scene you'd have to add it back elsewhere.

it just means comparing two copies would be easier to check for this technique than with a ms-based approach.

i think it also reduces your problem space as cutting a scene by over a few frames would be extremely obvious in most shows with competent editors.