Which makes sense, and also sounds like applying monetary pressure, say losing access to multiple states in the US, would incentivize prioritizing some engineers into solving this. The reason it doesn't exist yet, is because we've collectively decided through indifference and inaction that an "I swear bro" button is fine in the virtual world, but not the physical, and have now learned that it's not enough.

there is a HUGE difference between presenting ID in the physical world and digital world. In the digital world, once I provide an ID it is there forever. and it will be used for who knows what by who knows who. the only way to compare the two would be if in the physical world we provided an ID and whoever requested it gets to keep it and of course no place like that exists, you flash the ID (not that much unlike “I am 18 I swear” button in a lot of places) and you move on

Not that I think this is acceptable, but grocery stores around me scan your ID for alcohol purchases. In some states this is legally required.

At least with all the porn bills I've read, it's illegal to retain the information.

yes, without an option to opt-out and have your ID “manually” checked that is fucked up.

You didn't read the post you were replying to. That would be a "math nerd solution" that the states have not committed to accept, regardless of whether it actually works. Also, that sort of thing tends to require active cooperation from the entity issuing the IDs, which no state governments have committed to provide.

By the way, Pornhub actually has a preferred technical solution involving "device attestation". It's only marginally better than the snake oil the AV industry wants to use, and I don't like it, but they have said they'd stop objecting if something along those lines were standardized.

They just don't want to have to deal with a patchwork of mutually incompatible stuff (a) that's totally ineffective, (b) that leaves them either handling everybody's ID or dealing with questionable contractors to hold it, and/or (c) that leaves them having to convince the users to trust them or those questionable contractors with those ID images.

> require active cooperation from the entity issuing the IDs

It doesn't actually. Third party zero knowledge attestation works just fine. Think through the cryptosystem. It's possible for an authority to verify an ID for a given name and target site without revealing the former to the latter of the latter to the former

The commercial incentives around this are also terrible, given that the prevailing assumption is that a private third party will do the attestation as a for-profit service. It's practically guaranteed to add cost and liability for the porn vendor as well as risk of leaks, tracking, and data brokering for the consumer. It may as well have been designed to shut down free porn altogether.

And maybe it was, considering the history of porn regulation in the US. If you look at the requirements and contemporary rhetoric around 18 USC 2257 (yes, the thing that basically all legit porn sites with US operations have a disclaimer for), it was pretty blatantly intended to render the porn industry unable to operate. Porn producers were never actually intended to come into compliance, but rather presumed to be unable to (due to the ridiculous procedures and the supposed omnipresent use of underage, undocumented, and coerced performers).

> It may as well have been designed to shut down free porn altogether.

There was and is a fairly coordinated program of shopping this around to various legislative bodies (not only US states). It started maybe two years ago, I think. It didn't just catch on organically.

Some of it seems to be driven/funded by the AV companies, who presumably actually want the verification to happen. But a lot of the legwork and political contacts are provided by organizations that would definitely love to drive every porn site out of business. I doubt they expect to get all of them, or even all the free ones, but, sure, somebody like NCOSE knows that it's a major burden, and absolutely thinks that any damage to adults' access to porn is a positive feature.

It's not obvious that most of the legislators who vote for these things understand the implications at all.

So some of the people you could blame for this legislation definitely have such intentions, but probably not all of them.

> If you look at the requirements and contemporary rhetoric around 18 USC 2257 (yes, the thing that basically all legit porn sites with US operations have a disclaimer for), it was pretty blatantly intended to render the porn industry unable to operate.

Hmm. I'm ready to believe you, but 2257 doesn't seem totally infeasible to comply with in a VHS world. Dangerous to the performers, yes, because it requires tons of people to keep records of who they are and where they live, and those records are pretty much guaranteed to leak and be abused. An expensive nuisance, also yes. A chance to hound anybody who messes up out of business, and threaten them with prison, OK. Obnoxious overreach, sure. But totally impossible to comply with? I'm not sure about that. It actually seems easier than user AV.

The thing that always really got me about 2257 was that the claim was it was supposed to prevent another Traci Lords. They checked Traci Lords' ID. She showed ID. She had a real driver's license (based on a fake birth certificate, but it was the state that was supposed to check that, and anyway it was presumably a good fake). As far as I know, 2257 wouldn't even have slowed Traci Lords down.

> Hmm. I'm ready to believe you, but 2257 doesn't seem totally infeasible to comply with in a VHS world.

I'm not saying that it was supposed to be impossible to comply with per se, but rather that it was supposed to be impossible specifically for the porn industry as anti-porn crusaders imagined/alleged it to be at the time. There was a bunch of drug-war-like rhetoric about how porn had become more violent, exploitative, and outright criminal since 1970, when the previous government commission on the subject reported that porn wasn't an important social problem and should not be restricted for adults.

Do these cryptographic tools prevent the government entity providing them from knowing which websites verify your age? If not, then that's too much Uncle Sam watching over your shoulder as you watch your porn.

It depends on the tool.

https://zkdocs.com has some protocols described, if you're curious to learn more about this space.

Yes, they do precisely that, and in a way that depends only in math and cryptography. No possibility of leaking some naughty list after a hack.

Can a malicious gov site gather a list of which sites are asking for your verification?

No. The output of the government site is an attestation of age. It's a bundle of bytes. The site doesn't know or care what you do with the attestation.

How does the website prove compliance with the law under this scheme?

Same as a liquor store.

For a simple example, a JWT doesn't need to be done with shared secrets, and can instead use RSA private/public keys. The payload could contain just two fields like "issued" (to ensure the token isn't shared and reused) and "age" (what you want to share), you'd be able to decode the payload to see what's in it and know there's no identifying information, and the government site would provide the public key allowing anyone (like a porn site) to verify the JWT hasn't been tampered with without knowing who it was that got the JWT.

Can the gov site see who requests age verification? They already have to know who they're verifying because how else would they verify age?

Yes, but the idea is they wouldn't know why or for what site. In the JWT version I'm imagining, the porn site would say "go to your government website, generate this token, then copy/paste it here". There's no connection between the porn site and government site, and if the public key was retrieved and cached earlier there isn't even a timing-based way to connect them.

(JWT in general is a bad solution, but let's ignore that for the sake of discussion.)

I think you're severely underestimating the efficacy of traffic analysis and correlation efforts.

PrivacyPass is a better starting point, IMHO. The Cwtch developers (Open Privacy) even implemented PrivPass over Ristretto255, which is objectively awesome.

Didn't find many details, so please excuse the questions.

So the user would get a number of signed tokens from the gov to prove age, then a token could be shared with a site that verifies age without connecting to gov, assuming it's already aware of the gov's public key?

What prevents the same token from being used by more than one person?

How can the site prove compliance with the law? Will they need to store the tokens for each user? Can the gov tie a stored token back to a particular person that that verified?

Yeah, I was assuming the JWT might not be the best, but my point was that it's a simple example that doesn't require someone to read a math-heavy page.

> I think you're severely underestimating the efficacy of traffic analysis and correlation efforts.

The porn site would have the public key already stored/cached, so there isn't any traffic to correlate. There's still an issue with the first request, but like I said, this is a simple example that's easy to understand without getting into anything math-heavy.