The military sure do like to make themselves seem important. Super secret alien assault rifles being attacked by unspecified hackers from unknown locations. Yep, better keep funding new and exciting ways to kill people.
Maybe there's still some minuscule effect that can be detected, though. The sections on Magnus force in http://www.nennstiel-ruprecht.de/bullfly/index.htm seem to suggest that this is the case. But now I'm not so sure it's as simple as counting the number of periods of oscillation from the accelerometer.
IANAIBOEP (I am not an expert in ballistics or even physics)
When the round is fired, it will experience a huge acceleration from the propellant along its axis plus a very large rotational acceleration as the rifling spins the round from 0 revolutions per second to a large number of revolutions per second.
Once the round leaves the barrel, it is falling at a constant acceleration (9.8m/s, freefall, 0G, "weightlessness"), so it will not be able to measure the earth's gravitational force. It also will be spinning at a constant rate because there is no outside force increasing or decreasing the spin rate, thus no rotational acceleration. While there will be acceleration due to the centripetal effect of the rotation, that acceleration is constant and thus not usable with an accelerometer for counting revolutions.
That's different than measuring the change in acceleration.
Users usually calibrate the device at rest to get a good estimate of G so that it can be subtracted from accelerations measured later. Is that what you mean?
Perhaps measuring the negative acceleration caused by air friction could give you a velocity estimate (the velocity-to-friction function is probably non-linear). But this may vary a bit with air density, temperature, etc.
I'd be interested to see how it's done. My bet is that it's not directly measuring a near-impulse acceleration.
I'm more interested by the mechanism that programs each projectile's detonation distance. That seems like the tricky bit...
Since it's in the chamber protecting it from external sources of interference / attack (e.g. programmed to explode at the end of the muzzle) should be easy to protect against physically. I guess you could do it with some mechanically retractable pins or something, but electrical contacts and the inside of a bullet chamber don't seem like the sort of thing that should go together to me. You'd need to clean them constantly.
If not, I can easily imagine a "bullet jammer" gizmo that broadcasts a retargeting command. "Sorry, little grenade, you have another thousand meters to go...."