Hacker News Reader

Comment by inanutshellus | Hacker News Reader

Well. We're governed by laws. At least... theoretically. So if we make it illegal to scan peoples' faces then, theoretically, an oppressive government can't use it against its populace.

On the flip side, you have the "good guy problem" that we normal "good" people will happily give up our rights in order to help catch bad guys.

But really... it's not a MAC address or an IP address. It's your face. What "defense" is there to facial recognition software? Even if you stay off of facebook, your friends and family will still post photos of you there and reference you and talk about you. "No man is an island" has never been more apparent.

View on HN

Theory is good for books, but for life we need practical solutions.

An automated system that scans friends galleries to find my face in their galleries maybe would be helpful. Then I would see who uses my face in what context, so I can react; maybe send a polite request to remove me from the picture, set it as private, or simply remove the picture from the gallery?

Or maybe an approach based on spreading disinformation would be enough, so I could produce fake photos of my face with other randomly chosen faces from all over the internet, so that face-gathering systems will store fake relations between me and some random people?

There are lots of methods we can use to defend ourselves.

The "ban something" approach works only for parties that are willing to cooperate. But it's not the cooperative parties that we should be afraid of, it's the uncooperative ones.

It ultimately boils down to who owns the signal.

You would love to have a system that scanned the world's photos and notified you automatically about people commenting on you. You don't necessarily want the government to have that same system.

I'd also add that this begs a serious question: who _do_ you trust to own the signal? A multinational corporation? A government? No one? Everyone?

Furthermore, how do you rectify disagreements over that trust? I may trust a multinational corporation (especially if I own stock in it and can vote on its directors), but you may trust your government, and our neighbor may not trust anyone.

Both of those questions need to be "solved" (if such a thing is even possible), but that's (obviously) a lot easier said than done.

What I meant was that the "bad guys" will have it and they'll exploit it. No regulation will help because they will do it illegally. I want to be able to defend against it, but I won't be able to do it if there will be regulations against using those algorithms.

So I would rather allow anyone to have access to face recognition software, than to allow using it only for illegal use. Because nobody can prevent it from being used illegally.

Reply on news.ycombinator.com