I was curious about the article's inclusion of Australia, so I looked into the reference. The similarities are the "intermediate liability" and method of calcul

I was curious about the article's inclusion of Australia, so I looked into the reference. The similarities are the "intermediate liability" and method of calculating fines, which Australia introduces in the "Criminal Code Amendment (Sharing of Abhorrent Violent Material) Act 2019" Except, the Australian law has limits on the applicability of the amendment for research purposes, journalism, good faith, and matters of public interest. But it does fall into similar traps. Its not possible to know whether the violent content was published by someone other than the perpetrator, i.e. a victim, public bystander, or journalist. The social media company doesn't even need to know the content exists for them to be liable as intermediary (which is an interesting contrast to the safe-harbour provisions of the DMCA). It is a particularly broad amendment which will probably just result in the wholesale banning of anything approaching violent media. I can see why such a thing might hinder free expression and the press within Australia. (At the same time, if it were to prevent footage the Christchurch mass shooting from being shared, I can see the upside.)