Global military buildup

[Jacob]

I think it's clear that the "inevitable march towards the logic of liberal democracy" is if not an illusion, then at the very least a very long road ahead.

Whatever progress Russia may have had, it appears to have been mostly illusory. China is moving towards autocratic control and onto a path of conflict with the liberal world order. For much of the non-Western world the argument between liberalism and autocracy seems to be primarily about positional advantage for local elites, responding to local political imperatives. So yeah, I don't think there's anything inevitable about the the ascendancy of liberal logic.

I do agree that spending on the military and war engenders an opportunity cost for social and political development. Luckily, I think the advantages of open societies are enough that we can potentially find the right balance between robust defense and societal growth that still outpaces the autocrats.

[The Minsky Moment]

This probably merits a different topic, but the results of the Russian war to date seem to reinforce the Fukuyama thesis.  To the extent Putinism sought to pose as a viable alternative to the liberal democratic Western model, it has been badly discredited.

[Jacob]

This probably merits a different topic, but the results of the Russian war to date seem to reinforce the Fukuyama thesis.  To the extent Putinism sought to pose as a viable alternative to the liberal democratic Western model, it has been badly discredited.

As I see it, the liberal democratic Western model has proven the most efficient in producing a whole bunch of positive results - including economic growth, innovation, and general well being.

What is not clear to me is that that efficiency is means that the world will trend towards a liberal democratic social order, that the liberal democratic order will avoid collapse from internal stresses, or that we will necessarily prevail over challengers.

The West is better than China et. al., yes. That does not mean that China will not defeat the West, that the West won't kill itself somehow, or that unaligned countries will prefer the Western model over the Chinese one.

[The Minsky Moment]

On topic my thoughts are:
1) I would be careful about drawing very broad conclusion from 60 days of fighting by these particular forces in this particular war.
2) The costs of military procurement are driven to a great extent by figuring out the newest and best means to enclose human beings inside computerized hunks of metal of different sizes and shapes with different weapons attached - planes, ships, vehicles.  A lot of cost and vulnerability is driven by the need to enclose, protect and sustain the human beings inside.  Although there are difficult control and networking issues involved in having the human beings be outside rather inside the hunks of metal, there are very big potential returns.

[DGuller]

As far as drones go, I understand that they're far for invulnerable, but can they be countered in a cost-effective manner?  I can imagine a scenario where a strategy is to zerg rush with with cheap drones, both for military effect as well as for psychological effect (every drone the enemy sees can be a spotter for artillery aiming for you, but most aren't).  Can you take a cheap drone down with cheap means?

[The Minsky Moment]

What is not clear to me is that that efficiency is means that the world will trend towards a liberal democratic social order, that the liberal democratic order will avoid collapse from internal stresses, or that we will necessarily prevail over challengers.

It's a debatable point, but IMO the last 60 days have moved the needle towards the LD order. Military performance was supposed to be one of the defining positive attributes of Putinism and that has been badly exposed. On the flip side, the members of the LD order have displayed an ability to engage in effective collective action that some believed it no longer had. And while Ukraine may be fighting for nation and kin, the fact that they choose to define themselves as fighting for and as a part of the LD order is telling as to its continuing attractive pull.

[The Larch]

I see a lot of videos from drones filming or correcting artillery fire.

How far away from the target are they? Are they visible from the ground?

If they're only being used for filming they can be pretty high and far away, so they're hardly visible at all from the ground.

For instance, this is the Polish made FlyEye, which the Ukrainian army has been using since 2015 as a spotter drone for its artillery:



It has a wingspan of 3'6 m., and a length of 1'8 m. when fully set-up. It weighs 12 kg. and it can be launched by hand. It can be carried dismantled in a backpack and mounted in the field, and once launched it can operate for 2-3 hours, flying at 60 km/h at altitudes of up to 3 thousand m.

So, with that size and operating at that speed and altitude it can be assumed that nobody in the ground is even aware of it being there.

As far as drones go, I understand that they're far for invulnerable, but can they be countered in a cost-effective manner?  I can imagine a scenario where a strategy is to zerg rush with with cheap drones, both for military effect as well as for psychological effect (every drone the enemy sees can be a spotter for artillery aiming for you, but most aren't).  Can you take a cheap drone down with cheap means?

AFAIK the preferred counter-drone measure is jamming them, which can be done at considerable range, but can mess up your own communications, as well as any drones you might be flying yourself. A zerg rush style attack could, I think, be easily countered by mass jamming, if you're properly equipped for it.

[Malthus]

What is not clear to me is that that efficiency is means that the world will trend towards a liberal democratic social order, that the liberal democratic order will avoid collapse from internal stresses, or that we will necessarily prevail over challengers.

It's a debatable point, but IMO the last 60 days have moved the needle towards the LD order. Military performance was supposed to be one of the defining positive attributes of Putinism and that has been badly exposed. On the flip side, the members of the LD order have displayed an ability to engage in effective collective action that some believed it no longer had. And while Ukraine may be fighting for nation and kin, the fact that they choose to define themselves as fighting for and as a part of the LD order is telling as to its continuing attractive pull.

It seems to me that certain types of dictatorships have proved themselves bad at warfare in the modern world.

Perhaps the very things that allow the dictator to rise to power inhibit the ability to function well in war.

An addiction to lying, to the point that the truth becomes unimportant. Appointing cronies to important positions, and removing people of talent and ability who may become rivals. In military matters, preferring prestige projects designed to awe and intimidate, rather than concentrating on the boring and largely invisible-to-the-public fundamentals (like establishing a loyal an efficient NCO class, and an honest logistic corps).

Call it the Nasser effect. Or the Mussolini effect ... or the Putin effect.

[Zanza]

Is there any technical limitation to making drones fully autonomous? I mean there is obviously ethical considerations. But technology should allow auto-piloting the drone, acquisition of targets and launching of weapons. The hardest bit is probably sensors, but identifying military vehicles seems straightforward enough.

[Josquius]

Is there any technical limitation to making drones fully autonomous? I mean there is obviously ethical considerations. But technology should allow auto-piloting the drone, acquisition of targets and launching of weapons. The hardest bit is probably sensors, but identifying military vehicles seems straightforward enough.

I'd imagine all the same limitations behind self driving cars multiplied by 1000.
AI is not great at messy unpredictable situations.

[The Larch]

Is there any technical limitation to making drones fully autonomous? I mean there is obviously ethical considerations. But technology should allow auto-piloting the drone, acquisition of targets and launching of weapons. The hardest bit is probably sensors, but identifying military vehicles seems straightforward enough.

Some parts of drone operations can and already are automated. For instance you can set one of these spotter drones to do an area patrol mission, and I guess that automatically dispatching target info to a base could also be automated in some way. Then that base has to dispatch towards the selecter target or targets a separate drone to do the bombing itself (for instance a... Bayraktar!). Now, how much that decision-making of deciding which targets to dispatch attacks to can be automated I don't know. I guess it's possible, but don't know how desireable it is.

[celedhring]

Is there any technical limitation to making drones fully autonomous? I mean there is obviously ethical considerations. But technology should allow auto-piloting the drone, acquisition of targets and launching of weapons. The hardest bit is probably sensors, but identifying military vehicles seems straightforward enough.

I'd imagine all the same limitations behind self driving cars multiplied by 1000.
AI is not great at messy unpredictable situations.

Well, not quite. Aerospace is far easier to navigate than roads.

I think acquiring and deciding which targets to engage would be the most troublesome bit.

[The Larch]

Is there any technical limitation to making drones fully autonomous? I mean there is obviously ethical considerations. But technology should allow auto-piloting the drone, acquisition of targets and launching of weapons. The hardest bit is probably sensors, but identifying military vehicles seems straightforward enough.

I'd imagine all the same limitations behind self driving cars multiplied by 1000.
AI is not great at messy unpredictable situations.

Well, not quite. Aerospace is far easier to navigate than roads.

I think acquiring and deciding which targets to engage would be the most troublesome bit.

Target acquisition should be easy. It's deciding on engagements that is the tricky part. I doubt an AI can properly decide on which target's destruction brings the larger tactical advantage, you need a person to do that kind of decisions, as it's rather subjective.

[Syt]

As long as e.g. facial recognition is having issues with more heavily pigmented faces, I'm loathe to trust an automated recognition of enemy fighting vehicles and people.

[The Larch]

As long as e.g. facial recognition is having issues with more heavily pigmented faces, I'm loathe to trust an automated recognition of enemy fighting vehicles and people.

For vehicles I'd assume that thermal cameras are more practical, as a running engine is going to be much hotter than anything in the background (unless you're in an extremely hot environment). Now, I doubt that a camera can tell apart a, say, supply truck, from a tank.