Anyway, my country already has millions of young people who know how to use firearms.
Your rate of accidental firearms deaths suggest otherwise.
American civilians are really good at killing themselves. American soldiers are really good at killing other people.
But hey, power on, thanks to old Drumpf taking American back to isolationism, thank God, your daftness is no longer as much of a problem for the rest of us.
https://www.buzzfeednews.com/article/ryanhatesthis/its-not-just-food-and-hand-sanitizer-panicked-coronavirus
So your contention is that millions of laid-off, currently employed US soldiers will... what? Start killing civilians?
Ok, I'll bite. The link you offered showed civilians buying weapons, ammo, body armor, and other violence-related supplies. I would argue that the article supports a view that panicked civilians could start shooting people, not soldiers. But let's suppose your contention is right.
When are the laid-off soldiers going to start shooting people?
Are they going to attack the federal government, or local police, or civilians?
How many shootings does it take for us to say that your prediction is correct (other than the mathematical impossibility of laying off 2 million employees out of 1.3 million)? Are you going claim correct prediction if one laid-off soldier shoots her spouse / domestic partner, leaving me the avenue of saying that's only a drop in the bucket compared to the existing homicide rate? Or would you accept my criteria that the prediction isn't "right" unless it produces a substantial increase in homicides traceable to laid-off soldiers, such as 10% or more increase in the national homicide rate plus proof that the increase is due to the laid-off soldiers?
If your prediction doesn't happen, is there a date at which you would acknowledge that it didn't happen? Or is this one of these wild remarks that will be left open with no definable closure, no admission that the prediction was inaccurate?
I continue to be surprised this is a reaction in our presumably even-keeled forum. But if I'm wrong, I will acknowledge that. All I propose is that you specify a proof/disproof criterion in advance. What's your standard of proof, and your timeline for evaluating the prediction?