As far as I know (and I haven't read too much about QC, so I might very well be totally off), QCs are only useful for a very specific set of problems. And none of those seem to map to FB problems right now.
Will that change? Sure. Will FB do internal research with QCs? Maybe, once they've progressed past the vaporware state. (Sidenote - if we observe QC less, will development work better? ;)
But a working QC will not put FB out of business any time soon. The hardware is only a tiny piece of the puzzle of quantum computing. Algorithms for QC are rare, hard to develop, and the field is in its absolute infancy.