Allow me to play the Devil's Advocate here.
No, no one can really explain it, which is why it's so mysterious. The comments I've read demand real-world testing (instead of lab testing) to see what actually happens and account for false-positives.
It sounds largely like this is the case, though how it got through NASA I've no clue. Their funding is pretty slashed up though...
Actually, explaining how it works is the easy part. The difficult part is figuring out whether or not it actually does work.
We KNOW that "virtual particles" exist and can produce measurable effects in experiments (such as the Casimir effect).
We KNOW we can ionize particles and create thrust using ionized particles as propellant, we've been using ion thrusters on spacecraft for more than a decade.
The question remains whether we can ionize these virtual particles and produce thrust with them, which depends on the exact nature of "virtual particles" (which is by nature, inexact). See, "virtual particles" aren't really particles at all, they're perturbations in various fields. BUT WAIT! "Real" particles aren't particles either, they're also just a ripple in a field, although they are regular disturbances rather than transient. At this point I start going crosseyed, but basically "virtual" and "real" particles don't seem to be so different that they couldn't be used in the same way as our regular propellants.
I guess the short answer is "Yes, if the prevailing theories are correct it could work, but we're on the bleeding edge of theoretical physics and we're not really sure what's going to happen."
And that is why we experiment. Evidently NASA saw enough merit to at least give it a chance, which they don't do for just any hairbrained scheme.
Even if it is *not* a false-positive or misinterpreted result and *is* a magic drive, it produces almost no thrust. So it's no really that great; you'd probably still need to use stages to go interstellar in any kind of reasonable (less than hundreds of years) timeframe.
Well, let's have a little context. From 2003 to 2010, the Hyabusa spacecraft went from earth orbit, intercepted an asteroid, returned to earth and de-orbited with ion drive. The maximum thrust of this drive was 32 millinewtons. The drive fired for almost 9000 hours. (almost 300,000 Newton-Seconds of impulse) The craft also carried 75 kg of propellant (15% of the entire mass of the craft), though it only used 50 kg.
Which is why a functional drive of only a couple of millinewtons thrust would still be a massive breakthrough. A drive that could fire even at very low thrust for years on end with no propellant storage would be the ultimate in specific impulse.
As anyone who's played Kerbal Space Program can tell you, the farther you're going, the less thrust matters and the more specific impulse matters.
Granted, even if NASA's 40 micronewton results are true (China's 750 millinewton claim seems patently ridiculous), that's still about 100x smaller than what one might consider useful. It is however, the very first test of a fledgling and ill-understood device, so if it does work in principle it is likely that future devices would perform much better.
(For that matter, I don't even know if the test device weighed 5 kg or 500 kg, that kind of makes a big difference in how significant those results are)