Requiring that content be sourced from actual humans in good faith, and identifying people violating terms of service by spamming with puppet bots would be a go

Requiring that content be sourced from actual humans in good faith, and identifying people violating terms of service by spamming with puppet bots would be a good start. If the service is being operated by a national government, then you could require posters to prove residence or citizenship. But would anyone really want to use such a service? Could it possibly be even produced or operated efficiently? Once you put strict legal requirements on the operating entity it will get very slow, expensive, and user unfriendly.