4 ms·
Tell HN: I had an idea to mitigate the risk of being framed with a fake video
If you're a public figure, there's a substantial chance that someone will publish a video of you doing something incriminating. If it looks plausible it's your word against a third party's. What can you do?
One starting point would be to wear a device that takes pictures 24/7, geotags them and uploads a signature to a blockchain. Then, if and when a video surfaces of you doing some unspeakable act, you can issue the following challenge:
- When and where do you say this was?
Then the attacker would have to guess a plausible time and place. It must be one in which there is no other footage of you to contradict the account. Let's say they say: this happened on July 4th at 3pm, inside a room at the Grand Hotel. They know you where there because they saw you check in. If you were wearing this device, you could produce a trail of pictures spaced one minute apart, geotagged and verifiable. Of course it's not perfect, but it's a start.
It's pretty clear that videos and pictures are no longer proof of anything, so I believe these types of preventive steps will be a matter of necessity in the near future (at least for public figures).
- totaldude87 7y ago>>One starting point would be to wear a device that takes pictures 24/7, and someone hacks into that and leaks much more damaging "actual" videos/images than deepfakes. Remember celebrity iCloud fiasco?
- caryd 7y agoYour solution implies they don't do anything wrong 24/7. That's not the case with nearly any public figure.
- diego 7y agoNo, it doesn't matter if they do wrong things lots of the time. What matters is that if a deep fake surfaces then they can use the pictures of that specific period of time to prove they were not doing that. If they happened to be doing something right at that specific time, that's enough.