That's unlikely. Anti-drone defences will only improve, yes, but autonomous drone swarms numbering in the thousands to tens of thousands are doable today, and few weapons systems can handle the rate of launch/fire required to combat that. Especially if there are follow-up waves mixing drones and heavy missiles against which your anti-drone defences wouldn't be enough.
> WW1 after all was based on exactly this thinking: surely the volume of an army would overcome the machine gun.
But building a cheap kamikaze drone costs much less than building a human.
They're now much closer to $3000 USD+ at the low end for an ISR vehicle. $8000+ for the more capable FPV kamikazes is the estimate for Russian models.
Which is comparable to a 155mm artillery shell. But with a lot less payload.
There's already literally millions of drones being produced and used per year in that conflict - and they've made a big impact, but the stability of the frontline also reveals that the impact of "swarms" is hardly overpowering (the obsession with them is also weird - if you had thousands of assets in the air, the last thing you'd do is put them all close together).
As Iran shows, you don't need overpowering. You need to hit the enemy where it hurts them, like strategic infrastructure.
> "swarms" ... (the obsession with them is also weird - if you had thousands of assets in the air, the last thing you'd do is put them all close together)
On the contrary, a swarm allows you to overwhelm the enemy air defences, which allows you to hit targets, including those same air defences, without having to disable them first. Cf. Iran destroying a THAAD radar.