If the incidents are not the result of "alien activity" perhaps the US Military does have supremely advanced technology under wraps and is demonstrating this technology to ambiguously create the idea that it could be "aliens".
The "laws of physics" are pretty wide. There are many kinetic phenomena, entirely consistent with them, whose dynamics can't be filmed even at thousands of frames per second. There are all kinds of potential optical and electromagnetic effects (whether natural or intentional) that could show essentially arbitrary scale and motion.
"Def[ies] the known laws of physics", is a good sign your imagination is too narrow, or your understanding of physics too small.
To the best of my knowledge, the only accepted laws of physics relate to the 3 laws of thermodynamics. Everything else is still considered a theory (to the best of my knowledge), even if the theory works really damn well (for instance relativity is still considered a theory).
I know there are accepted "laws" in other fields related to physics, such as Ohm's or Kerchov's laws in electrical engineering, but I don't think these are generally accepted scientific laws, more generally accepted as engineering laws where for nearly all cases for engineering, they apply to a sufficient accuracy to produce products that are sufficient.
How about conservation of momentum?
Thermodynamics on the other hand relies only on effects of large numbers (of hard balls) with no further assumptions. In so far I can understand the GP assertion although I am reluctant. You put in the hard ball assumption (particles can exchange energy) which may put it in theory territory. As well the laws hold only for equilibrium situations which can be reached in (sufficiently dense) baryonic matter but maybe not for non-baryonic matter.
And, it only takes a single counterpoint to disprove a theory. Hence theories are signifcantly harder to prove than disprove.
What you're referring to as laws are theories that work well enough in some circumstances that needing something better isn't needed. Such as relativity to implement GPS. Newton's theorems in Principia Mathematica are generally good enough for most day to day type stuff. E.g. we dont need to involve Relativity to understand the physics being acted upon a car (thats not a Tesla roadster launched into space).
Science isnt much about coining laws. It's more about obtaining an ever more correct model of reality, and as you say, make (reasonably accurate) predictions.
I don't know but I would assume they are usually referring to conservation of momentum being violated, rather than something like the 2nd law of thermodynamics or the Heisenberg uncertainty principle. Which is pretty easy to see unless you have reason to believe there is something rather crazy happening that you cannot see (like I imagine a neutron being emitted at >99% of the speed of light to cancel some change in momentum that you otherwise can't explain).
Hell, even if someone saw a B-2 today they'd probably think it's alien.
It refers to objects that start and stop moving instantaneously, or make turns at hard right angles, as if they have no inertia.
They'd probably think they're living in the matrix and several pixels of sky are glitching.
There was a great interview last night on Tucker Carlson where the expert said there are basically two possibilities that the military should explore: one of our terrestrial enemies has an aviation capability more advanced than ours, or an extraterrestrial civilization has capabilities more advanced than us. Of course, there could be some sort of natural phenomenon we don't know about. In any event, it is worth investigating in a rational way.
This doesn’t mean it’s aliens but this is pretty fucked up.