I think instead of blocking, users could flag posts (by sending a flag event) and clients could decide to hide, block or show a disclaimer.
It would also be transparent who flagged it. The validity of the flags could also be determined this way (is the flagginguser verfified...).