My biggest problem with the FSF's point of view is that they see these kind of eventualities as inevitable for
all non-free software.
I would claim that there is not only room for non-free open source software sans abuse, but even closed source software. When it comes to something you want to use as a core foundation on which to build (whether you are building software, general business, academic work etc.) I completely understand how this is very likely to lead to you being exploited. This applies for example to an operating system, or a programming platform (JVM, .NET, etc.). But there are plenty of places where you can use closed source software that don't put you in such a disadvantageous position.
One example is non-required tooling like an IDE. If Jetbrains goes Dr. Evil on everyone tomorrow (though I guess some would argue their recent pricing model changes have done that), there is still Eclipse and plenty of room for someone else to break into that space. Changing IDEs isn't nothing, but it's not like I'd have to rewrite my code.
Another is various general XaaS offerings. If Heroku (or more likely Salesforce) goes off the deep end tomorrow, worst case is that I have to administer some servers myself. Again, I don't lose anything that I would have had before anyway.
The current trend is already building everything on top of open source infrastructure. Even the JVM and .NET are open source now, as are most other pieces of the development puzzle (Databases, libraries of all shapes and sizes, etc.). In fact, I don't think even a big player like Microsoft would have much luck trying to get people to invest in another closed source platform. So if we walk this line between open and closed software, what do we lose by using some replaceable closed source tools or services? That is one question I think the FSF doesn't have a good answer for.