Good question. Should we apply this axiom (and resulting benefit of the doubt) to power structures / organizations? One way to look at it is: which approach is more likely to lead to a better understanding of the problem and thus the solutions?
If we are curious, we can ask questions like: what are the root cause incentive structures that led to anti-social actions? and: What systematic adjustments can we make that disincentivize such actions in the future? I think that these questions lead to a more broad perspective on how to address problems. Where a judgemental approach narrows the possible solutions and opens us up to the cure being more deadly than the disease.
It's a tough issue, for sure, because the playing field is just so unequal. To even be able to approach these issues in a more equal way you'd probably have to start a lawsuit and go through a process of discovery because the corporation is bound by all these bizarre legal, financial, and political motives that don't apply when people are interacting with each other. This leads to corporations responding to such requests in a way that's inherently against the interests of the public answering these questions. Without a system in place that further aligns and incentivizes companies interests with their users the only way you can really approach this in a more equal manner is something like a lawsuit. Many times the problem is incredibly complex and while some people, even C level people may be all for providing this transparency it breaks down at scale and the company is at the end of the day beholden to its own interests and perpetuating itself.
The official answers they are giving is basically meaningless legalese and that's becoming more and more common. As I said before since Roblox isn't entering into this conversation in good faith the only way to force that conversation would be legal actions.
Any kinds of laws further aligning the needs of clients/the general populace against the needs of big corporations/political elite have a really hard time passing because the system is self sustaining. We already understand this dynamic and we have been asking these questions for a long time but they keep getting pushed down and torn apart as "unrealistic".
This is a good video about how hierarchical systems self select for people who play the game rather than people who stick to their morals: https://www.youtube.com/watch?v=rStL7niR7gs
The issue here is a bigger one because it's going to self sustain and to make progress at this point you have to kind of just fight fire with fire (be part of the elite, own a big corp, or throw massive amounts of money at it) and getting involved with those systems can lead you on the path towards getting in a situation where you're betraying your own goals for entering them...
If we are curious, we can ask questions like: what are the root cause incentive structures that led to anti-social actions? and: What systematic adjustments can we make that disincentivize such actions in the future? I think that these questions lead to a more broad perspective on how to address problems. Where a judgemental approach narrows the possible solutions and opens us up to the cure being more deadly than the disease.