Seems easy to strip off or not add, and then you'd have a video with unknown provenance which the echo chamber will eat hook, line, and sinker. Wouldn't you?
I'm also curious how this would stop someone from using AI to generate an image and then just using a digital camera to take a photo of their monitor. "The photo of <politician> executing someone in the street seems to be legit. There's provenance metadata showing that the image hasn't been tampered with since it was taken and cryptographically signed by Nikon's physical sensors"
edit: formatting
It would definitely stop pretty much any counterfeit if they added some rudimentary depth data into the image format as well, within the signed contents. That way simply taking a picture of a monitor would be obviously detectable, and not alterable without removing the signing. It wouldn’t have to be a high resolution depth map at all either.