A best effort implementation should be fine. Advertisers already have suspected age fields available (derived from behavioral patterns, linked information across devices, etc). When you develop a reasonable suspicion that it's a minor then stop fleshing out the profile and targeting them. That's what the big players already do (maybe not Meta, but certainly at least Google). That doesn't quite fall into your (1) point since it's not profiling that's being banned, but profiling-based ads.

What's the downside to the ban as described, when compared to the status quo?

If there's no profile already, then they can serve the ad, since it's not a profiling-based ad and they don't need to know if the viewer is a minor.

If there is a profile then they have the knowledge to avoid advertising based on that if it's a minor.

I don't see why ID verification needs to come into play?

I would love to see a ban on _all_ profiling-based ads, but this seems like a step in the right direction to me.

It seems like a roundabout way of banning profiling of non-ID’d users in general which seems… good? Like, if overlapping regulations have the practical effect of banning tracking, then that’s a win for society, right?

If you ban profiling-based ads to minors, and it's not possible to reliably detect when a user is a minor, then you have de facto banned all profiling-based ads. Since you deem that solution acceptable (as do I!) I don't see what the problem is.

> There's no way to implement this without (1) usage-based profiling (what's being banned in the first place) or (2) ID verification.

(3) zero knowledge age verification hosted by the gov which already has all the data anyways