It doesn't take an infinite amount of energy to generate gravity. Just the amount needed to create the mass that generates the gravity. That in itself is a very large amount by our standards (but still finite), so any method that can directly generate gravity by itself will probably be relatively cheap. In the end, unless we're tapping into an outside system (which conservation of energy takes into account - it just becomes a larger system), we're still not going to be getting extra energy out of it. At most we could do some weird stuff with time, which is where the real applications for something like this would be I think.
The problem is that we really don't know what gravity is, why it's so weak, etc. Until we do it's all speculation, unless you want to get together enough mass to make a black-hole, or find a way to make stable neutronium. Black-holes aren't exactly something you want to be walking around on, and even a smaller, condensed earth-sized mass is going to be majorly disruptive to the local solar system, nevermind natural and unnatural satellites ("Oops, we just made the moon spiral down to Earth, they're not going to be too happy.") and tides.
Basically, artificial gravity is one of those technologies that's probably not even as practical as, say, dyson spheres or the like. And if we really can generate large amounts of gravity using a relatively energy-efficient method, and we can create shielding to keep it in one spot, we're basically boned as soon as some homicidal idiot learns how to shut that shielding off. If this technology did come into being, we'd probably put any generators in deep space, where they couldn't do much to the inner system. Actually we'll probably end up changing ourselves and what we need (food, etc.) to not need gravity before we even know what it is.