This does seem like a much better approach than censorship, however I still see two major issues:

1) Whatever entity determines the categorization of content wields great power over the reach of that content. In the FB case, this is likely sufficient to decide elections. Although one could argue FB has this anyway simply via control of the news feed.

2) There will likely be sufficient people who disable filtering that the amplification effect will still be present and large. Most people would agree that amplifying certain speech is dangerous. How should that be handled?