I've never in my 3 years of being on here seen anything violating the guidelines. After all, they have whitelists. Sorry if this is the wrong forum

Even if a Cloud Chat has a whitelist, there is still away to send inappropriate messages. For example, if there was a chat with these (and other) words:

• I
  
• me
  
• you
  
• do
  
• like
  
• not
  

You could send messages like “Do you like me”, “I do not like me”, and “I do not like you”.
The last one is, of course, inappropriate.

That doesn't seem that bad to shut down the whole chat system

GoldenBonnie09 wrote:

That doesn't seem that bad to shut down the whole chat system

But the people where still trying to get around the community guidelines (and the way I showed isn't the only one).

yeah I now see how people could do that. still sucks tho

But you can still make chats with preset messages, as long as they are appropriate.
Like “Hello”, “Goodbye.”, or “Cool!”.

Reason 1: Blacklist doesn't work
Like you are using a blacklist and you ban word “cat”. The user would send “c a t”, “ca t” and more. It would be very hard to find them all.
Reason 2: Whitelist

WindOctahedron wrote:

Even if a Cloud Chat has a whitelist, there is still away to send inappropriate messages.

More information is in this topic. (Click me!)

I think the main reason is actually just that they're basically impossible to moderate. I mean, if someone is insulting you on a cloud chat in a project, how do you report that? You can't report the user for it when there's no real way to show exactly what you're doing: the ST would just have to take your word that it happened, which opens up all sorts of possibilities for people reporting anyone they don't like for “insulting me in (insert chat project here)!” even when nothing of the sort happened. How would the ST know which reports are real, and which aren't?

The only way to “prove” this would be taking a screenshot, uploading it somewhere, and linking there in your report, but that could be faked with a little bit of effort as well.