First of all, they need to define what "harmful" is as quantifiable variable(s). Then the states in this case must prove that the contents from Meta are harmful in this context. Lastly, they need to prove there exists a statistical causal link from such "harmful" contents to the mental well-being of the children. Many parents kind of "sense" social media is harmful, but I don't see how they can "prove" that social media is really "harmful"
edit: grammar