The simplest solution is to require all online devices to have a "child mode" that can be activated during setup, and require all parents to enable this for minors under 16. In this mode, the device takes screenshots every few seconds of active use, and makes this viewable on the parents' devices as a timeline. This must be private with full end-to-end encryption and limited data retention in the clients (7 days or so).

It's much simpler than blocking, and much more effective. Most parents don't know what to block proactively, blocklists are imperfect, and the biggest threats are hiding in the most innocent looking apps (Discord, Roblox, Reddit, even just messaging with friends from school).