This would be fine if it was actually done perfectly - ie. Devices get a signed ticket from the government identity provider, device can provides a cryptographically verifiable ticket to the site that its a valid identity and their age is within the $x age range but not tied to the user’s actual identity / document, and the device doesn’t ask the government identity provider to mint a new ticket each time it needs to attest (maybe 500 tickets are minted at a time and you auto renew 500 more each month)
However the likelihood of this actually being done correctly is slim to none.
But obviously this doesn't "solve the problem". It's another bandaid with an extensive list of failure modes and tradeoffs. It falls into the class of "the age limitations that exist currently are easily circumvented" type of solution.
In my opinion it is fine to leave it there and accept the tradeoffs. We could mandate better website marking, and mandate better device or app-level mechanisms, and improve monitoring and restricting tools, or we could do even less and keep it more or less heterogenous.
But I do not agree that it is "moatism" to talk about it on the website side. There is a real and genuine desire to actually have the kinds of age restrictions that are only possible with strong user identity broadly deployed. Refusing to engage because of imputation of malign motives on the other party's part is not going to persuade anyone, especially if they do not personally have those malign motives.
Collecting user biometric data and trying it to a nominally anonymous user identity is not required here.
This is 100% 'won't someone please think of the children' pearl clutching to hide what's actually going on - furthering control of the online exchange of ideas.