There are two elements of this position that I urge you to consider:
First - if KF is guilty of crimes that would stop CloudFlare from providing them services - then it's both reasonable and normal for people to encourage CloudFlare to act before a verdict. It's extremely common for commercial actors to have policies based on their impressions of likely criminality (i.e. credit card transactions are marked fraudulent before a court of law weighs in). Also, even though it's impossible to perfectly provide services to all 'good' customers and deny services to all 'bad' customers - it seems desirable to minimize the services you provide to 'bad' customers. I.e. It's good to cut off likely fraudulent transactions before a court verdict and we should encourage companies to act ahead of verdicts.
So if you think a customer is probably doing something you'd cut off service for if it was determined in court, you should generally change your posture towards that customer somewhat. This is often a difficult determination and a difficult line to draw - but it's clearly a good thing to try and do.
Second, the ability to prove a case in court depends on a great number of elements aligning. There are a number of scenarios where an actor is doing things that, could they be proved in court, would get them denied service - but for reasons unrelated to the actions themselves a determination cannot be made. The moral weight of providing services to those actors is the same and it's still desirable for companies to seek to avoid providing services to bad actors. Your legal responsibility is different, but you would still like to avoid serving clients whose actions you abhor.
Otherwise, you get into the position where moral action is impossible outside of the opinion of an authority. Let's say you see a stranger's bag grabbed by someone else on the street - would you say the only moral course of action is to cooperate with an official investigation (if one happens)?
First - if KF is guilty of crimes that would stop CloudFlare from providing them services - then it's both reasonable and normal for people to encourage CloudFlare to act before a verdict. It's extremely common for commercial actors to have policies based on their impressions of likely criminality (i.e. credit card transactions are marked fraudulent before a court of law weighs in). Also, even though it's impossible to perfectly provide services to all 'good' customers and deny services to all 'bad' customers - it seems desirable to minimize the services you provide to 'bad' customers. I.e. It's good to cut off likely fraudulent transactions before a court verdict and we should encourage companies to act ahead of verdicts.
So if you think a customer is probably doing something you'd cut off service for if it was determined in court, you should generally change your posture towards that customer somewhat. This is often a difficult determination and a difficult line to draw - but it's clearly a good thing to try and do.
Second, the ability to prove a case in court depends on a great number of elements aligning. There are a number of scenarios where an actor is doing things that, could they be proved in court, would get them denied service - but for reasons unrelated to the actions themselves a determination cannot be made. The moral weight of providing services to those actors is the same and it's still desirable for companies to seek to avoid providing services to bad actors. Your legal responsibility is different, but you would still like to avoid serving clients whose actions you abhor.
Otherwise, you get into the position where moral action is impossible outside of the opinion of an authority. Let's say you see a stranger's bag grabbed by someone else on the street - would you say the only moral course of action is to cooperate with an official investigation (if one happens)?