I think you may have misinterpreted my comment (you know I don't work for Square right?).
> Hard problems should be challenges, not excuses.
Which is precisely why I mentioned in my original post that my preference (and Eventbrite's) goes towards doing the opposite and trying hard to give the user the benefit of the doubt. By doing this we expose ourselves to more risk because we want to transact with everyone -- without risking to disclose too much about our policies, you'll notice our system is very tolerant for mistakes and missing inputs, and in case something does go wrong we have a dedicated support team to deal with these issues. This makes building and training our algorithms that much more complicated if we want to keep our risk profile in check but it is a trade-off I'm happy to make, so I'd like to think I'm not guilty of what you're accusing me of.
Overall, my point was just to provide some context on where Square comes from since OP seemed confused about what they perceived as a ridiculous policy. The comparisons you picked (Comcast, Google) are a good example of why it's interesting to talk about fraud (which is a topic we rarely have in mind when building or using payment services): fraud is not a simple inconvenience for startups -- it's something that can potentially put the entire business at risk, so you have to have both the ability and the willingness to take on that risk. In the absence of controls I can think of many ways to defraud Square out of millions of dollars in the matter of hours. They're strong enough to take the hit it now, but one single such attack while they were younger could have put their success in considerable danger had they not taken some precautions.
Of course this is no excuse for not having a friendlier or more reactive support, but Square or not Square I just felt this is a topic worth bringing to the spotlight.