We have this already in Ukraine: <a href="https://mod.gov.ua/en/news/thirty-companies-have-already-joined-the-private-air-defense-initiative" rel="nofollow">https://mod.gov.ua/en/news/thirty-companies-have-already-joi...</a>
Unironically could be a legit business model.<p>Not just for datacenters either - warehouses, refineries, power stations, you could sell to anybody with a building full of expensive stuff.
In some countries maybe, but in west where all the valuable infrastructure is there is absolutely no way you could get a permit to shoot down anything.
Obviously your biggest customers would be in active warzones, or areas worried about soon becoming warzones.<p>Drone attacks on US soil are (hopefully!) not happening anytime soon.
> Drone attacks on US soil are (hopefully!) not happening anytime soon.<p>This would be incredibly stupid. An attack on American soil could be met with a literally nuclear response against Iran or whoever perpetrated it.
This is trivially falsifiable. We didn't nuke Afghanistan after 9/11.<p>(We did nominally succeed in regime change in the country that didn't attack us, though...)
Glad no one has taken the invitation from the President to “take out” San Diego/LA.<p>Prediction markets are betting on a false flag attack around midterms.
Ukraine showed you can deploy from shipping containers delivered inside hostile lands quite successfully. As much stuff gets smuggled into US ports on a daily basis, sneaking a few of these into the US isn't impossible. Honestly, I'm surprised it hasn't been done yet.
> areas worried about soon becoming warzones.<p>So...the west if we go by the recent statements of chancellor Merz
believe it or not but legislation can change pretty quickly when shit actually hits the fan
I don't think it issue is selling in the current climate, producing seems like the major roadblock. If Russia, the supposed number one in air defence can't even protect their military bases, ramping up production might be hard.
HN discover the defence industry…
Oil tankers..
MTCR, ITAR, EAR, NFA ...<p>The US State Department will throw everything at you if you dare tried to build or possess your own guided guns and rockets, even if you were not from or in the US - because there was a US ally that exported a near-orbital solid motor sounding rocket to a Balkan nation with full factory setup and manufacturing training services without asking anybody in particular, which resulted in the recipient nation achieving prototype production and successful live fire tests of an indigenous surface to air missile system comparable to Soviet S-75 strategic SAM that was eventually selected.<p>If it weren't for those pile of laws and leverages... building a SAM was doable since 1950s.
The genie is out of the bottle, over the next 100 years this is a real problem US, China, and Europe will face.<p>Though someone like Amazon will have to calculate, is the 50-year cost of protecting all facilities less than the 50-year cost of rebuilding the ones that are destroyed?
It's already happened to Amazon once, and they were not able to recover the data from the destroyed data centre, so they have already experienced this and have some data on which to make those decisions.
This is one of the factors driving the development of orbital satellite data centers. While satellites are also vulnerable, they are at least somewhat harder to hit than a stationary terrestrial data center. For now it's questionable whether such constellations are feasible due to concerns over launch costs and cooling but some people are taking it seriously.<p><a href="https://www.spacex.com/spacexai/starmind" rel="nofollow">https://www.spacex.com/spacexai/starmind</a>
In addition to all the engineering problems associated with cooling those data centers and the shields that you'd need against cosmic rays, those things would only accelerate Kessler syndrome.
Batch processing could be done in solar orbit.
The difference is that to hit stationary terrestrial data center in the USA you'd need to go through multiple layers of defense systems, not to mention it would probably be an act of war.<p>As for the satellites they can be jammed and there's not much you can do about it.
Just add standardized bremer walls and bulkheads to the interior?
There surely is a point where it's cheaper and safer to bury them underground.<p>I'd invest in digging equipment for the next war.
> it's cheaper and safer to bury them underground.<p>It did not help the Iranians and their nuclear facilities.
Cyberbunker was shut down too soon.
Look up the capabilities of GBU-28 and other Bunker Busters first.
These are a lot more expensive and difficult to deploy. They weigh several tons. You can't just strap one to a $200 quadcopter and fly it over.
How many $200 quadcopters delivering blow after blow to reach the equivalent damage to a single MOAB? Is it an impossible number to reach for those unable to deliver MOAB?
We're talking about bunker-busters - the MOAB is not a bunker buster. It's an airburst weapon designed to clear out large open spaces, and isn't effective against hardened bunkers.<p>No reasonable amount of quadcopters could match a bunker-buster. Surface explosions are not very effective against buried targets.<p>The weight and density of the bunker-buster is extremely important to its operation. When dropped from a high altitude, it breaks into the bunker with sheer kinetic force, and then explodes once inside.
Those are entirely different missions. A GBU-43 MOAB is used against soft targets on the ground. A GBU-28 is used against hardened buried targets. It's simply physically impossible to dig a big enough hole with a bunch of small drones or bombs: the only way to penetrate deeply enough is with a hard object moving really fast.
A MOAB apparently has the equivalent of 11tons of TNT. A drone carrying a grenade has up to 180 grams. So you need about 61 111 drones to deliver the equivalent explosive. At that point it might be cheaper to just bring the MOAB.
>How many $200 quadcopters delivering blow after blow to reach the equivalent damage to a single <orders of magnitude larger purpose designed munition>?<p>How long do I have to piss on a steel plate to get the same effect as a waterjet cutter?<p>About the same answer.
When will the payload be quadruped robots that can walk in through the door?
You don't dig a hole deep enough that Bunker Busters can't penetrate. You drive up to the root of a mountain where it meets flats and drill sideways. Getting your men out in case the entrance was bombed can be tricky though.
to use a GBU-28 or similar you need to establish air superiority first
The current crop of "cruise missiles" are cheap and low penetration. Burying things is not 100% safe, but it certainly increases the attack difficulty.
"Silo"
You could do a ghaza and park human shields over military relevant infrastructure? Introducing Host.Age housing the only housing with free heating, free warn water and free leftover tokens .
Sounds like Snow Crash becoming slowly a reality.
I saw a video the other day advertising AI-powered a anti-air/drone gun. Seems to be part of a defense trade show, but also seemed to indicate that it was available to certain "qualified" citizens, but I'm not sure what that means exactly
Wouldn't it be cheaper to just build a fence around the building? They are cheap and easily repaired.
Releasing drone swarms wrapping in chaff with IR lamps or laser diodes to account for different guidance methods. They cluster around predicted trajectory. Could be more cost effective than a Patriot battery
They don't all have to be drones in the conventional sense. A subset could be a set of balloons with different amounts of helium, an arduino/servos that control one or more fans for directional and limited altitude control, and a box/release with chaff inside.<p>Whatever isn't used/destroyed could be recovered wherever it lands after a day or two.<p>It should be less expensive a possibly more effective in certain circumstances because of increased carrying capacity.
I'm extremely surprised that isn't already a thing. I'm similarly surprised that all the datacenters which get press for the negative impact on the surrounding communities have remained untouched.
It is, or at least it's expected that Russian companies get their own air defense:<p><a href="https://www.themoscowtimes.com/2026/05/28/russia-lets-private-firms-buy-heavy-weapons-to-defend-sites-from-drone-attacks-rbc-a92876" rel="nofollow">https://www.themoscowtimes.com/2026/05/28/russia-lets-privat...</a>
I look forward to anthropic drone swarms defending against AWS mounted infantry and OpenAI cruise drones. SpaceX is at war with Pepsi. SpaceX had always been at war with Pepsi.
Russian companies have had mercenaries of different kinds for some time now. Im sure Redut has anti air batteries.
Russia recently ordered corporations to start funding their own air defense.
You can always exhaust anti air defenses faster than they can defend if sufficiently prepared. Decoys help, just as you would with an intercontinental ballistic missile. Defender has to succeed 100% of the time, attacker needs to succeed once.<p>If your strategy assumes you can succeed against Iron Dome (which, I argue, is the state of the art for available air defense system), you should always be able to win as an attacker. Just have to test defenses ahead of time to understand level of effort required (munition capabilities, cadence, and quantity required, broadly speaking).<p><a href="https://missilethreat.csis.org/countermeasures-penetration-aids-and-missile-defense/" rel="nofollow">https://missilethreat.csis.org/countermeasures-penetration-a...</a><p><a href="https://en.wikipedia.org/wiki/Iron_Dome" rel="nofollow">https://en.wikipedia.org/wiki/Iron_Dome</a>
Russian suicide drones are still like minimum $10k each. That's probably like ~1k rounds in 12.7mm/14.5mm or ~200 rounds in 20mm. Missiles are a lot more expensive, but the defense side cost can be substantially lower than the offense side if you could make it work with an auto turret rather than guided missiles.
Smart munitions for the early warning, signal intelligence, and radar, second wave is dumb, fast, cheap munitions for kinetic solving.
More DCs are than obviously required for the battery stores that these laser turrets will shoot.<p>Obelisks from Command and Conquer pinned to the roof of the DC would be amusing although personally I'm leaning towards prism towers.
Will we get degraded performance notifications when the DC's power suddenly prioritize the lasers over the GPUs? At 15:34UTC, we noticed degraded performance due to an incoming drone strike. Normal performance was restored at 15:42UTC once the strike was defeated.
The fundamental challenge is that it will never be economically feasible to provide the necessary defense hardware at scale. Iran beat back the US military (“most powerful military in the world” quote unquote) and Ukraine Russia’s military on a shoestring budget. You can protect some targets some or most of the time, you cannot protect all potential targets all of the time.<p>If someone wants to take out a data center, they will likely succeed assuming they have the resources (probably military drones, but perhaps more traditional ballistic artillery). The world has become a more dangerous, volatile, and unpredictable place, for <i>gestures broadly at everything</i> reasons.<p>(thoughts, opinions, and threat model always my own)
New Startup idea - A railgun that launches dictators and fascists into the sun.
Wouldn't work. They would just send enough drones to overwhelm it or run it out of ammos or Russia out of money. This is the lesson we are learning the hard way in the Middle East.
The last thing we need is more Silicon Valley funded military startups. This means the ecosystem is now profiting off of war, suffering, war crimes, and human death. What we need is startups focusing on peace and diplomacy. That solves way more problems in a net positive manner.
The irony of your username is not lost on me.<p>That said, I’m in agreement in principle. A lot of what you’re asking for is statecraft, and doing that with a profit model is brittle to corruption. But I’d greatly prefer to see folks trying to build things on hope then on bloodshed.
The industrial military complex is doing just fine thank you very much.
What do you suggest for Ukraine? What's their alternative if they don't want to be forced to become a Belarus?
[dead]