4 ms·
This makes me wonder if there's a business case for a privacy-preserving identity service which does age verification. Say you have a strong identity provider
by cloudfudge 9mo ago
This makes me wonder if there's a business case for a privacy-preserving identity service which does age verification. Say you have a strong identity provider that you have proven your age to. Just as the 3rd party site could use SSO login from your identity provider, perhaps the identity provider could provide signed evidence to the 3rd party site that asserts "I have verified that this person is age X" but not divulge their identity. Sidestep the privacy issue and just give the 3rd party site what they need to shield them from liability.
- dakiol 9mo agoThe question is: why would services like Google and others want to use such privacy-preserving identity solutions? They wouldn't gain anything from a non-invasive, user-friendly system, so I don't think they'd use it. They want more data, so they are going for it.
- cloudfudge 9mo agoI was thinking someone like Auth0 might want to offer it. They are not in the business of invasive user tracking but are in the business of trust.
- tzs 9mo agoConsidering that Google is releasing open source software they developed to facilitate such systems [1], apparently they are OK with the idea. It could simply be that they realize that online age verification becoming required for some online activities is inevitable for the same reasons age checks are required for some non-online activities, and when that comes to pass they want to be able to do in a way that doesn't expose them to too much risk. Yes, Google loves data but that doesn't mean they don't care about risk. The data they would from some of the age verification methods probably wouldn't improve their ability to advertise much but would cause a lot of problems if leaked. Another possibility might be that have no choice. My understanding is that in the EU member states that enact online age verification laws will have to require that verification can be done using the privacy-preserving system that the EU Digital Identity Wallet will support. Sites will be able to use other methods too (as long as the don't violate GDPR) so they could support something that gives them more information for advertising, but they will still have to support the privacy-preserving option. [1] https://news.ycombinator.com/item?id=44457390 https://news.ycombinator.com/item?id=44457390
- SJC_Hacker 9mo ago> The question is: why would services like Google and others want to use such privacy-preserving identity solutions? They wouldn't gain anything from a non-invasive, user-friendly system, so I don't think they'd use it. They want more data, so they are going for it. Consumer pressure and/or laws
- awkward 9mo agoThe article does go into this and gives lip service to the idea that a secure third party could expose age without exposing identity. Ultimately, there's still the problem that even if point of verification can be done in a zero trust way, you are still entrusting very sensitive information to a third party which is subject to data breach.
- tzs 9mo agoIf you do it right the only sensitive information exposed to the age gated site is that your age is above their threshold. The party that actually has to at some point verify who you really are of course has your sensitive information, and there is no obvious way to work around that. However, there is a way to make it so that it doesn't matter. That is by making them be a party that already has that information. Probably the simplest would be to make it be the same government agency that issues your physical identity documents like passports or drivers licenses. If we don't want it to be a government agency or we want to have competition banks would be a possibility.
- enahs-sf 9mo agoI’ve been noodling on this idea for a while but I think getting commercial acceptance would be hard. People have tried it with crypto albeit with lukewarm results. I think to have the network effects required to be successful in such an endeavor, it would have to come from a vendor like apple or google unfortunately. You kind of want an mTLS for the masses with a chain of trust that makes sense.
- cloudfudge 9mo agomTLS is no good because the target service could then uniquely identify you. I think you explicitly want a three-party scheme where the target service just accepts the idp's assertion about your age in a cryptographically secure way.
- sigwinch 9mo agoI feel like mTLS could still work in third-party signing.
- izacus 9mo agoThis is how Swiss e-ID was proposed to work: https://www.eid.admin.ch/en https://www.eid.admin.ch/en
- triceratops 9mo agoYes. In fact the 3rd party doesn't even need to know who you are. https://news.ycombinator.com/item?id=46447282 https://news.ycombinator.com/item?id=46447282
- cloudfudge 9mo agoThat's quite an elaborate system. It goes through a lot of gyrations (not the least of which is inventing a whole new type of crime and passing laws about it) and doesn't sound even as strong as the age verification "required" to buy cigarettes in the US. I'd think "welcome to pornhub. Either log in or do Privacy-enhanced Age Verification by Auth0 (TM)" would be a lot easier to get off the ground.
- triceratops 9mo ago> not the least of which is inventing a whole new type of crime and passing laws about it They're already writing laws about age verification. Tacking this onto a bill in progress would be the way to do it. As for the other "gyrations", selling age-restricted products IRL is already done. Gift cards and the infrastructure for them is well-established. Most importantly, this is something a regular person on the street can understand. This is vital for capturing the "save the children" vote.
- dredmorbius 9mo agoOr simply the connectivity provider, see: <https://news.ycombinator.com/item?id=46627433 https://news.ycombinator.com/item?id=46627433>.
- MiddleEndian 9mo agoI'm more interested in a business that reliably provides fraudulent IDs to services that unnecessarily want IDs that I cannot avoid for some reason.
- tzs 9mo agoYou've almost got it right. You just need to modify this part: > Just as the 3rd party site could use SSO login from your identity provider, perhaps the identity provider could provide signed evidence to the 3rd party site that asserts "I have verified that this person is age X" but not divulge their identity The way you compared it so SSO login makes it sounds like there would be interaction between the 3rd party site and the identity provider. That's bad because if someone got a hold of the records from both the site and the identity provider they might be able to match access time logs and figure out who you are. A fix is to make it so you get your signed document from the identity provider ahead of time, and that document is not tied to doing age verification with any particular site(s). You get it once and then use it with as many sites as you want. When you use it with a site to demonstrate age we need to do that in such a way that neither of you have to communicate with the identity provider. If the site needs to verify a signature of the identity provider on something you present they use the provider's previously published public key. We need to make it so that when you use the signed document from the identity provider to show your age to a site they don't see enough from the document to identify you, even if they have been compromised and are collaborating with the identity provider to try to identify you. Finally, the signed document should be bound to you in some way so that you can't just make copies and give them to others or sell them on the black market to people who want to evade age checks. BTW, since under this approach the identity provide isn't actively involved after their issue your signed document what probably makes the most sense is to have your government be the identity provider. In particular, the same agency that issues your driver's license or passport or nation ID (if your country has those). Such a system can in fact be built. The EU is including one in their EU Digital Identity Wallet project, which has been in development for several years and is not undergoing large scale field testing in several countries. It is supposed to be deployed to the public this year or next. The first version handles the binding of the document to you by tying it to your smart phone's hardware security element. They plan to later support other types of hardware security elements. 90+% of adults in the EU have smart phones (95-98% for adults under 54), and it is going up, so the first version will already cover most cases. Google has published some libraries for implementing a similar system. Both the Google libraries and the EU system are open source.
- cloudfudge 9mo ago
- dredmorbius 9mo agoWhat's ... boggled me about this issue since forever is that: 1. Most people access online content through either a personal or business broadband service (residential, mobile, or place-of-work). 2. Those services ... bill directly. Which means that it should be possible to specify an age preference for the service account as a whole, and/or subsets of it. The service can specify whether or not age-bounded online services are acceptable or not, as well as specific classes of age-bounded services. E.g., a workplace service would generally allow for >18 access, but might restrict usage of gaming, gambling, pr0n, or related sites. A household might request no age gating at all (all >18 or whatever minimum age is mandated) or several classes of service, say, if adults and children are present. 3. Where it's necessary to specify multiple preferences, multiple network segments could provide this logically (e.g., an IPv6 block with unrestricted and age-gated ranges), with distinct devices being allocated appropriate gateway addresses. 4. Effectively, the connectivity provider then attests for age, without requiring any finer-grained identity disclosure. Why ... A. Would this not work? B. Is it not being generally proposed?
- cloudfudge 9mo agoTo be blunt, because it sounds insane and simultaneously solving the problem at the wrong abstraction level, and based on criteria that have nothing to do with age. Age-based IP ranges? This sounds like a recipe for reinventing the entire internet in a non-backwards-compatible way. Networks are not people. Why would we treat the network as your identity?
- dredmorbius 9mo agoWhat strikes you as unreasonable? What I'm suggesting for different service levels is simply to map a specified access / age-verification level to a specific network access point. This is better suited to IPv6 which has a much larger address space than IPv4, and often allocates a range of addresses rather than a single IP, though NAT or IP shenanigans might be possible for the latter as well. The point is to put the heavy lifting of age validation at a level at which who is getting the service has already been vouched through service account provisioning.