We need better technology embedded into video recorders to help prevent digital altering of any recorded video

Let’s face facts — we’re now in a situation where technology makes it all but impossible to take a video and then submit that video as proof that something happened without any type of CGI trickery or other BS post-processing that can alter what was recorded. The question is how do we use technology to help prevent other technology from being used to digitally alter what was recorded.

There’s a few things that I thought of that might make a difference. Probably the best solution (although uncontrollable) is to have an event take place where multiple people are recording the event from multiple angles and they get uploaded to Youtube or some other video sharing site.

This would require someone to somehow do post-processing on video taken by multiple people and then uploaded those videos to their accounts. Pretty much all but impossible to do — so this will always be something that would serve as decent proof.

But I wanted to reach out to this community to ask the community how we could use technology to help provide additional evidence that something is the real deal and I came up with one possible idea.

You have a video recorder that is also connected to the internet. As it is recording, it takes a frame every few seconds and digitally creates a “verification key” for that specific video frame. For instance, it takes a frame and performs a calculation for that frame and uses the public key of a remote server to “sign” that frame with the datetime also embedded in the signature. The person recording can then no longer alter that frame because the signature would not match. As long as the private key isn’t leaked, the digital signature could not be altered or changed. All of this happens in just a few seconds (thankfully we now have CPUs that can handle the heavy computation almost instantly).

What we need is someone to create a video signature service that can handle processing the signatures and an app that interfaces with the server to send frames every X seconds to be signed. This wouldn’t prevent people from dangling a hubcap off a fishing rod, but it would basically eliminate the possibility of CGI post-processing.

I’m sure there are some holes in this solution, but I’m pretty confident this community could run with this technique and make it more solid. One possible hole is someone hacking the app so that they can do their CGI post-processing and then send the video up to the verification server.

We’re probably at a point in time where we need to start thinking of ways to help prove authentic videos or at least help prevent CGI and other post-processing trickery.