this post was submitted on 27 Jul 2025
561 points (99.1% liked)
Technology
73416 readers
4261 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
To avoid people from simply copying the "age proof" and having others reuse it, a nonce/private key combo is needed. To protect that key a DRM style locked down device is necessary. Conveniently removing your ability to know what your device is doing, just a "trust us".
Seeing the EU doesn't make any popular hardware, their plan will always rely on either Asian or US manufacturers implementing the black-box "safety" chip.
If it is about hiding some data handled by the app, that will be instantly extracted.
There are plenty of people with full integrity on rooted phones. It's really annoying to set up and keep going, and requiring that would fuck over most rooted phone/custom os users, but someone to fully inspect and leak everything about the app will always be popping up.
Look at the design of DRM chips. They bake the key into hardware. Some keys have been leaked, I think playstation 2 is an example, but typically by a source inside the company.
That applies to play integrity, and a lot of getting that working is juggling various signatures and keys.
The suggestion above which I replied to was instead about software-managed keys, something handed to the app which it then stores, where the google drm is polled to get that sacred piece of data. Since this is present in the software, it can be plainly read by the user on rooted devices, which hardware-based keys cannot.
Play integrity is hardware based, but the eu app is software based, merely polling googles hardware based stuff somewhere in the process.
I understand. In the context of digital sovereignty, even if the linked shitty implementation is discarded (as it should be), every correct implementation will require magic DRM-like chip. This chip will be made by a US or Asian manufacturer, as the EU has no manufacturing.
The key doesn't have to be on your phone. You can just send it to some service to sign it, identifying yourself to that service in whatever way.
It's that "whatever way" that is difficult. This proposal merely shifts the problem: now the login to that 3rd party can be shared, and age verification subverted.
A phone can also be shared. If it happens at scale, it will be flagged pretty quickly. It's not a real problem.
The only real problem is the very intention of such laws.
How? In a correct implementation, the 3rd parties only receive proof-of-age, no identity. How will re-use and sharing be detected?
There are 3 parties:
The site (2) sends the request to the user (1), who passes it on to the service (3) where it is signed and returned the same way. The request comes with a nonce and a time stamp, making reuse difficult. An unusual volume of requests from a single user will be detected by the service.
Neither 2 nor 3 should receive information about the identity of the user, making it difficult to count the volume of requests by user?
Strictly speaking, neither needs to know the actual identity. However, the point is that both are supposed to receive information about the user's age. I'm not really sure what your point is.
I must not be explaining myself well.
Yes, that's the point. They should be receiving information about age, and age only. Therefore they lack the information to detect reuse.
If they are able to detect reuse, they receive more (and personal identifying) information. Which shouldn't be the case.
The only known way to include a nonce, without releasing identifying information to the 3rd parties, is using a DRM like chip. This results in the sovereignty and trust issues I referred to earlier.
The site would only know that the user's age is being vouched for by some government-approved service. It would not be able to use this to track the user across different devices/IPs, and so on.
The service would only know that the user is requesting that their age be vouched for. It would not know for what. Of course, they would have to know your age somehow. EG they could be selling access in shops, like alcohol is sold in shops. The shop checks the ID. The service then only knows that you have login credentials bought in some shop. Presumably these credentials would not remain valid for long.
They could use any other scheme, as well. Maybe you do have to upload an ID, but they have to delete it immediately afterward. And because the service has to be in the EU, government-certified with regular inspections, that's safe enough.
In any case, the user would have to have access to some sort of account on the service. Activity related to that account would be tracked.
If that is not good enough, then your worries are not about data protection. My worries are not. I reject this for different reasons.
The reverse is also a necessity: the government approved service should not be allowed to know who and for what a proof of age is requested.
Of course not: both intentional and unintentional leaking of this information already happens, regularly. That information should simply not be captured, at all!
Additionally, what happens to, for example, the people in Hungary(*)? If the middle man government service knows when and who is requesting proof-of-age, it's easy to de-anonymise for example users of gay porn sites.
The 3rd party solution, as you present it, sounds terribly dangerous!
(*) Hungary as a contemporary example of a near despot leader, but more will pop up in EU over the coming years.
It would send the proof to you. It would not know what you do with it. I gave an example in the previous post how the identity of the user could be hidden from the service.
It would be a lot easier to get that information from the ISP.
In both your examples the government service has your full identity, then pinky promises to forget it.
Unless I'm misunderstanding something?
Not quite the same, as IP addresses are shared through NAT, VPNs exist, etc. With the proposed legislation it is illegal for website operators to deliver content to known VPN ips, as they cannot confirm that the end user isn't a EU subject.
It can be like buying alcohol in a store. They look at you and see your age. Or if it's unclear, the store clerk asks your idea and promptly forgets all about it. Except you're not buying alcohol but a login for some age verifier.