Once facial recognition technology is deployed on a large scale, it can be leveraged in some scary ways.
First, when it's tied into a network of surveillance cameras, the cameras become useful for tracking a person rather than a location. They could easily search for any video caught of you, forever.
Second, when you have a network of cameras you can track everyone's location at a very fine resolution, orders of magnitude better than tracking via cell phone towers. (And the cell phone tower data can be used to help with facial recognition since you can narrow down candidates of who might be in the photo.)
Third, once you're tracking locations in a network, the accuracy becomes pretty much 100% since you just need to match an unknown person once and then you can propagate their identity as they move around.
First, when it's tied into a network of surveillance cameras, the cameras become useful for tracking a person rather than a location. They could easily search for any video caught of you, forever.
Second, when you have a network of cameras you can track everyone's location at a very fine resolution, orders of magnitude better than tracking via cell phone towers. (And the cell phone tower data can be used to help with facial recognition since you can narrow down candidates of who might be in the photo.)
Third, once you're tracking locations in a network, the accuracy becomes pretty much 100% since you just need to match an unknown person once and then you can propagate their identity as they move around.