I don't have special knowledge of the particulars of the internship, but characterizing the effect as creating a ghetto makes sense to me, and I can try to explain why.
Have you ever heard the argument that donating free clothing to Africa is actually bad for Africans? The idea is that by providing free clothing, you destroy the local industries which allow Africans to create clothing for themselves, eliminating jobs, introducing dependency, and eliminating the possibility of a native African-style clothing industry flourishing and eventually exporting products and enriching the community. [1]
Charity comes with a counterintuitive hazard: any time you do something for someone that they could do for themselves, you make them weaker. You deprive them of the independence and control that doing it themselves would offer them, and you reduce the development of their own infrastructure for doing the task the next time.
Charity must attempt to make people stronger. Reduce the impact of a crisis while people recover. Give otherwise unattainable support while the person develops in ways they couldn't without it. Philosophically, funding startups, not donating products.
I'm not saying donating things is always bad . . . it depends on the effects. Charity is hard, I think more because it's hard to figure out how to help people, than because it's hard to be unselfish. I think there are a lot of cases where it's hard to anticipate what the effects will be, which is why it's so important to study what you're doing and see if you're actually helping. It's very easy to do harm while intending to do good. But the rough criterion I use comes from an old song about charity in Africa, and argues that the right attitude to have is, "I want to see you shine / See your light not mine." In a charitable effort -- is the focus on the generosity of the giver? Or is it about what the recipient can achieve and become, with the giver's help being a footnote?
A formalized mentorship seems like a marginal case, one you could argue worked either way. It's giving the new employee knowledge and extra help about the firm, a running start, if you will. Where's the harm in that?
The harm is that the new employee would normally acquire these things herself by asking coworkers, and in the process would learn who she could ask about what, and would begin establishing peer relationships that she will use throughout her career. The implication that she cannot do this for some reason isolates her, and giving her an artificial source of support weakens her ability to find her own support. This would be a bad thing if that artificial support ever went away, but in a way, it's almost a worse thing if it sticks around, because the result is the new employee relying on an inferior network to the one she would have developed herself.
The term ghetto to me implies that all the capable and competent people moved away, leaving the poorest ones stuck with just each other. A targeted system which isolates and coddles new employees will have just that effect -- they'll have the impression that it's the system and each other they can talk to, leaving them with a much, much poorer network than if they had just been introduced to the veteran coworkers as if that were perfectly normal which it is.
I'm right there with you in that this sort of thing shouldn't be taken personally. People intend it helpfully, and that charitable impulse should be honored (if perhaps gently educated, too). But I myself have always been very suspicious of programs targeting women, because there are a lot of people in the world who want to Help Women and I always suspect those programs are about their glory and pro-womanness or whatever, and more likely to harm me than do me any good, even if I can't see right away why that might be. I think my chances are better on my own, and anyway -- I'd prefer to have a similar experience to others in my field in the ways I can, to help me integrate into the new group.
Helping people develop and become stronger is difficult, even in the most concrete cases. It's not dissimilar to raising a child -- you argue one minute that relieving their frustration by making a task easier will give them the boost they need to do something even greater, and argue the next that letting them fail will make them stronger and they will eventually succeed on their own. It's an art more than it is a science, but I think as long as you are teaching something you understand and guiding someone toward mastery, you are pretty safe. But I definitely think it is much safer to devise solutions to help concrete cases than to help statistics. If I see you need help finding the break room, it is easy to conclude that giving you directions will be helpful. Providing you a buddy preemptively because certain classes of employee are less likely to ask for directions easily . . . seems more dubious in its effect. I understand the impulse, but I really think it is safest to stick to the concrete, and to help the people around you and solve the problems you can clearly identify directly.
[1] http://www.huffingtonpost.ca/mariah-griffinangus/africa-char...