- cross-posted to:
- technology@beehaw.org
- cross-posted to:
- technology@beehaw.org
The UN Convention on the Rights of the Child clearly expresses that minors have rights to freedom of expression and access to information online, as well as the right to privacy.
These rights would be steamrolled by age verification requirements.
There is no way to remotely verify someone’s age across the Internet without violating their privacy. If there was, there would be no way to use it that doesn’t violate their other rights.
It security engineer here: zero knowledge proofs are exactly that. Proof your age isg higher than X, but not even how much higher. They can’t even profile you based on that information as they can’t recognize you across visits.
Some government identity cards already support that. For everybody else there are companies that offer that service.
BTW I’m against age verification. We had access to porn at a certain age, I want my children to be able to look when that gets interesting to them. But then again I’m pretty progressive and open with sexuality in general and I take time out of my day to actually talk to my kids about dangers on the internet.
If I search for zero-knowledge proofs relating to age verification the only thing I see is the hash chain method “based on a 2013 paper by Angel & Walfish” which is clever enough but does not in itself solve the problem of proving age while maintaining one’s privacy. It allows Alice to demonstrate to a verifier that she is over the age of 65 while revealing nothing else other than her name or some other identifying piece of information. Avoiding the reveal of any such information is what we would want to avoid.
Is there some better way to do it?
You only need to prove the number in your government id is greater that the required. Number. The number is signed by the government CA
Any reasonable government doesnt gove out ascending number-IDs, Right??!
Your government… you know… the people that already have all your data and issue your passport… cannot include a flag (properly cryptographically signed by them) that tells a service “Yes, the guy that just inserted this valid passport is an adult. You don’t need to get any other info. We already checked for you.”, no other connection or transfer of data neccessary?
If you know a way to do it without invading people’s privacy you’d better go tell the government of Spain about it, because they didn’t manage to find it when they designed their eIDAS scheme which they hoped would become the Europe-wide standard. Not sure if that’s still seen as likely but I haven’t heard about any other concrete proposals yet.
I’m talking about things you can do technically.
Governments don’t plan completely idiotic ideas because they don’t know better but because their actual reason for choosing the system they chose is NOT creating a workable system that protects your privacy.
That’s the whole point. Articles like this aren’t completely wrong. The systems planned are indeed a risk to privacy rights. But we need to stop pretending that it’s an accident and the government simply don’t know better or there is no better solution at all. Actual solutions exist and we need to talk about the fact that those are ignored intentionally because a working system that protects your privacy is simply not the goal here.
What do you mean by violating privacy?
If you have a passport, citizenship, or birth certificate your age is already documented.
yes, however the government should NEVER have access to what social messaging apps ANYONE uses without a warrant.
Isn’t that a matter of implementation whether they even receive this information or not during validation?
the fact that they received a validation request is informative that they probably shouldn’t be able to access, however regardless of implementation, assuming causality and the speed of light remain, this will be information the government will recieve. Some entity (probably the government) would* also need to know who to send the response to, technically they could just broadcast this over some low frequency transmission broadcast to everyone, but realistically the government would need some kind of address (IP, fax number, po box, etc.).
Technically this is an implementation detail, however the only ways to implement this type of thing that wouldn’t be comprimizing would involve citizen prompted government broadcasts and trust that the government won’t have records of who requested the broadcast and what number was sent (which would make it trivial for adults to just sell the age identifier) and would still worsen the average citizen’s security because it still takes effort to generate a unique identifier for every site.
If my gov creates a digital cert of age and signs it then I should be able to use that and the service provider can verify against gov public key, no? No information of visit exchanged.
As an alternative I also expect it to be possible via zero knowledge proof https://en.m.wikipedia.org/wiki/Zero-knowledge_proof
that still uniquely identifies you to the company and would make having multiple accounts that are fully separate harder. in addition there would be no way of knowing whether or not the government has or hasn’t hidden another layer of data (like your name) in the certificate.
This would also be trivial for children to bypass as it would need to be usable an unlimited number of times (or else individuals couldn’t have multiple accounts) therefore it would only take one adult sharing their cert and signature publicly for any child to have a valid certificate.