I was reading about that electrical fire at the Isfahan radar site back in January, and it struck me how we almost just accept those headlines at face value now. Another mysterious failure in a sensitive facility, another shrug from the international community. But when you look at the leaked reports suggesting it was actually a swarm of micro-drones launched from inside the city limits, the whole picture changes. Today's prompt from Daniel is about exactly this: the mechanics of long-term, deniable sabotage. He wants us to look at how these physical and cyber infiltrations of critical infrastructure serve as the real opening move of modern warfare, often years before a formal declaration of war.
It is a fascinating lens to look through because most people focus on the big, loud moments—the missile barrages, the troop movements, the declarations on state television. But as Herman Poppleberry, I have spent a lot of time looking at the gears turning in the background, and the reality is that the war is often decided during the years of silence that precede the first official launch. What happened in Isfahan two months ago was just the tail end of a campaign that has been running for over a decade. We are talking about a shift from overt military strikes to the invisible dismantling of a nation's backbone. If you can disable the radar before the jets even take off, you have already won the air war without firing a single traditional shot.
It is that doctrine of ambiguity that really gets me. Israel has mastered this art of neither confirming nor denying, which creates this bizarre geopolitical space where everyone knows what happened, but because no one officially claims it, the victim is stuck in what analysts call a retaliation vacuum. If you admit you were sabotaged, you admit your security is porous and your most sensitive sites are compromised. If you strike back, you are escalating against a ghost, which makes you look like the aggressor to the rest of the world.
That is the strategic feature of grey zone operations. You want to stay just below the threshold of a formal casus belli—a justification for war. If you blow up a laboratory using a remote-controlled machine gun, like what happened to Mohsen Fakhrizadeh back in November of twenty twenty, you are sending a message that is both incredibly specific and technically deniable. You are not just removing a person; you are demonstrating total atmospheric control over the enemy's home turf. You are saying, we can put a one-ton automated weapon on your highway, operate it via satellite from a thousand miles away, and you cannot stop us.
I remember we touched on this in episode seven hundred thirty-eight when we talked about the invisible front and the sabotage of Iranian air defenses during the Twelve Day War. But Daniel is pushing us to look deeper into the tradecraft itself. How do we get from a piece of intelligence to a physical explosion in a hardened, air-gapped facility? Because you cannot just hack a centrifuge that is not connected to the internet from a basement in Tel Aviv. There has to be a bridge.
This is where the intelligence-sabotage nexus becomes critical. You cannot destroy what you have not mapped with perfect precision. People think of the twenty eighteen nuclear archive theft as just a heist for information—a way to prove Iran was lying about its nuclear ambitions. But from a sabotage perspective, that was the ultimate reconnaissance mission. When the Mossad took those half-ton of documents and compact discs from a warehouse in Tehran, they were not just looking for proof of a weapons program. They were getting the blueprints. They were getting the names of the specific suppliers for every valve, the specific models of the Siemens industrial controllers, and the exact layout of the facilities at Natanz and Fordow.
So the archive theft was basically the user manual for every sabotage operation that followed. It gave them the "where" and the "how" before they even sent in the "who."
It provided the target list. When you know exactly which valve in a facility is the single point of failure, you do not need a thousand-pound bomb. You need a small explosive charge or a piece of malicious code tailored to that specific industrial controller. Look at the Natanz incident in April twenty twenty-one. Reports suggested that three hundred pounds of explosives were smuggled into the facility and detonated near the internal power system. This is a facility buried deep underground, protected by multiple layers of security and the Revolutionary Guard. You do not get three hundred pounds of high explosives into that room without months, if not years, of human intelligence work. You need someone on the inside, or someone who can bypass the physical security checks with a legitimate-looking crate of equipment.
That is the part that always blows my mind. We talk about cyber-attacks like Stuxnet as if they are magic, as if a line of code just travels through the air and breaks a machine. But even Stuxnet needed a physical bridge. Someone had to plug in a thumb drive. It is that transition from digital to physical that seems to be the hallmark of the elite agencies. They are bridging the air gap with human proxies or incredibly sophisticated delivery systems.
And the delivery systems are getting wilder as we move into twenty twenty-six. The assassination of Fakhrizadeh was a turning point because it used a robotics-sabotage hybrid. It was a one-ton automated machine gun equipped with artificial intelligence to account for the delay in satellite transmission and the recoil of the weapon. It had multiple camera angles and was assembled inside Iran piece by piece. After the hit, the entire mounting system self-destructed to leave as little evidence as possible. That removes the risk of an operative being captured on-site, which is the biggest nightmare for any intelligence agency.
It also removes the human element of hesitation. But let's talk about the infrastructure side of this, because that is where Daniel’s prompt really hits home for a lot of people in the tech world. He mentioned our recent discussion on microwave backhaul and how buried fiber is vulnerable. We saw this play out with the Baltic Sea cables in late twenty twenty-four. You had multiple data cables severed in Swedish waters, and the prime suspect was a Chinese freighter, the Yi Peng three, allegedly dragging its anchor.
That is a masterclass in low-tech, high-deniability sabotage. Over ninety-five percent of international data traffic moves through those undersea fiber cables. If you want to blind a region or disrupt a financial market, you do not need a sophisticated cyber-attack that leaves digital fingerprints. You just need a heavy piece of metal and a plausible excuse about a dragging anchor in a busy shipping lane. It is the ultimate physical bypass for any digital security measure. Even if you have the best encryption in the world, it does not matter if the physical glass carrying the signal is snapped in half.
It is almost cheeky in its simplicity. You spend billions on firewalls and encryption, and someone with a rusty anchor just resets the whole conversation. But what about the more permanent, dormant threats? We have been hearing about the Volt Typhoon group for a couple of years now, and the way they operate seems like the definition of pre-war sabotage. They aren't blowing things up yet, but they are "pre-positioning."
Volt Typhoon is the perfect example of the "sabotage in waiting" model. This is a Chinese state-sponsored actor that has been infiltrating United States critical infrastructure—utilities, transportation, water systems—not to steal data or intellectual property, but to sit there. They are looking for ways to maintain persistence in the network so that if a conflict breaks out, say over Taiwan, they can flip a switch and cause physical chaos in the American heartland. Imagine the pressure on a political leader if they know that the moment they sign a deployment order, the power grid in three major cities might go dark or the water treatment plants might stop functioning. It is a gun held to the head of the national grid.
It changes the definition of a peaceful period. If the infrastructure you rely on for water and power is already compromised and just waiting for a command, are we actually at peace? Or are we just in a very quiet phase of a continuous war?
We are in a state of permanent grey zone competition. The shift we have seen recently, especially with the escalation between Israel and Iran leading up to the current situation in March of twenty twenty-six, is that the sabotage is becoming more frequent and more kinetic. The July twenty twenty-five report about Mossad establishing drone bases inside Iranian territory was a massive leap. They were not just flying drones across the border from Azerbaijan or Iraq; they were staging them from within. Imagine the level of infiltration required to maintain a drone hangar, a maintenance crew, and a launch team inside a hostile, surveillance-heavy state like Iran.
It makes the Isfahan incident in January feel almost inevitable. If you have the staging bases, you can strike with total precision and leave no trail. But let's look at the other side of this, the information sabotage. The IRIB hijack in June twenty twenty-five, where the Iranian state broadcaster was taken over to show protest footage and slogans against the regime. That is sabotage of the national narrative.
It is coordinated with the physical side. When you combine a cyber-hijack of the state media with physical strikes on infrastructure, you are creating a sense of total vulnerability. It is what we call information attrition. We did a whole deep dive on this in episode eleven hundred ninety-three. The goal isn't always to destroy a building; sometimes the goal is to destroy the belief that the government can protect anything. If the state can't even keep its own television station on the air, how can it protect its nuclear facilities or its citizens?
I remember you saying that a failing missile can still win if it forces the enemy to reveal their radar positions or drains their interceptors. Sabotage does the same thing on a psychological level. If a gas pipeline blows up in early twenty twenty-four and the government cannot explain how it happened, every subsequent flickering light bulb becomes a potential Israeli agent in the mind of the public. It creates a "ghost in the machine" paranoia.
That psychological pressure is a force multiplier. But we should also talk about the risk of this strategy. Sabotage is not a cheap alternative to war. It requires a massive, multi-decade investment in human intelligence—HUMINT. You need people on the ground who are willing to take incredible risks. When we see the culmination of these efforts, like the assassination of Khamenei in February of twenty twenty-six, people talk about it as a single intelligence triumph. But that was likely the result of years of deep infiltration, mapping out the most private movements of the leadership, and finding the one physical vulnerability—perhaps a specific ventilation shaft or a delivery route—that could be exploited.
It brings us back to that idea of the intelligence-sabotage nexus. You can have all the satellites in the world, but if you want to know which room someone is sleeping in or which cable carries the most critical data, you need eyes on the ground. This is why the Mossad's investment in Iranian human intelligence over the last twenty years has been the deciding factor. They turned the country into a Swiss cheese of security holes.
And it is not just Israel. The Russian GRU has been running a massive sabotage campaign across Europe for the last three years. We have seen gas pipelines targeted, communication hubs, even arson attacks on shopping malls in Warsaw. They are using locally recruited proxies, often small-time criminals or people they find on Telegram, to carry out these acts. It provides the GRU with a layer of insulation. If the person is caught, they look like a common vandal or a disgruntled local rather than a state actor.
That proxy model seems like the most dangerous development because it lowers the barrier to entry. You do not need a highly trained operative if you can just pay a local guy a few thousand dollars to throw a Molotov cocktail into a transformer station. It is the democratization of sabotage. It makes the "attribution problem" almost unsolvable.
It is, and it makes response nearly impossible. If a shopping mall burns down, is it a tragic accident, a local crime, or a Russian intelligence operation designed to create social unrest? The ambiguity is the weapon. It creates a sense of unease without providing a clear target for retaliation. This is exactly what we saw with the Nord Stream explosions in September of twenty twenty-two. Years later, we are still seeing conflicting reports. Was it a pro-Ukrainian group? Was it a state actor? The fact that we are still debating it four years later proves how effective the operation was at creating a permanent geopolitical mystery.
It is the ultimate cold case. But the practical takeaway for people listening, especially those working in critical infrastructure or tech, has to be that the air gap is a myth. If your security model relies on being disconnected from the internet, you are only protected from the remote hackers. You are not protected from the guy with the bolt cutters, the smuggled explosive charge, or the drone that flies into your cooling tower.
Physical access is the ultimate bypass. If someone can touch your hardware, they own your hardware. We saw this with the gas pipeline sabotage in early twenty twenty-four. These were hardened systems, but they had physical points of vulnerability—valves and sensors located in remote areas—that were exploited through simple kinetic means. The lesson for the modern era is that security must be holistic. You cannot separate the cyber from the physical. You have to assume that the "pre-positioning" phase has already happened.
So, if you are a facility manager or a network architect in twenty twenty-six, you have to move to a zero-trust model for physical infrastructure. Just as we do not trust a device just because it is on our network, we should not trust a piece of equipment just because it is inside our building.
You need to audit your physical infrastructure with the same intensity you audit your code. Look for those single points of failure that can be reached with a pair of bolt cutters or a simple drone. Red-team your physical security. If you assume the attacker is already inside, how do you limit the damage they can do? This is the shift in threat modeling that Daniel’s prompt is really pointing toward. We are moving from a world of clear battle lines to a world where the front line is your local power substation or your undersea data cable.
It makes me wonder if this level of sabotage actually prevents larger wars or just makes them inevitable. On one hand, you are degrading the enemy's ability to fight, which might make them hesitate. On the other hand, you are creating so much pressure and humiliation that they might feel they have no choice but to go overt.
It is a delicate balance. In the case of Israel and Iran, the sabotage campaign arguably delayed a full-scale war for over fifteen years. It bought time. It forced the regime to spend billions on security and rebuilding rather than on expansion. But eventually, the cumulative effect of those operations—the "Operation Rising Lion" campaign—created a situation where the Iranian regime felt its very existence was at stake, which contributed to the escalation we are seeing now. Sabotage is a tool for managing time, but it does not resolve the underlying conflict.
It is like putting a patch on a leaking pipe while the pressure keeps building. Eventually, the pipe is going to burst, no matter how many clever patches you apply. But I want to go back to the technical side for a second. You mentioned the remote-controlled machine gun. How much of this is moving toward fully autonomous sabotage? Where a system is programmed to identify a target and strike without a human in the loop at the final moment?
We are already there. The Isfahan radar strike in January was likely a pre-programmed swarm. Once they are launched from within the city, they do not need a continuous satellite link that can be jammed. They can use visual navigation—basically "digital eyes"—to identify the specific antenna array and fly right into it. This removes the risk of signal interception. We are entering the era of fire-and-forget sabotage.
Which makes the attribution problem even worse. If there is no signal to trace back to a pilot, you are left with a pile of generic drone parts and a lot of questions. It feels like we are watching the traditional rules of engagement just dissolve in real-time.
They are dissolving because the technology allows for a level of precision and deniability that was unthinkable twenty years ago. In the past, if you wanted to sabotage a nuclear program, you had to send a team of commandos on a suicide mission. Now, you send a piece of code or a five-hundred-dollar drone. The cost-to-effect ratio has shifted entirely in favor of the attacker.
It is a grim realization, but a necessary one. We have to look at these incidents not as isolated accidents, but as data points in a much larger, more coordinated campaign. Whether it is a cable in the Baltic or a radar in Isfahan, the mechanics are the same. It is about finding the physical lever that can move a geopolitical mountain.
And it requires us to be much more vigilant about our own vulnerabilities. We tend to think of sabotage as something that happens somewhere else, to someone else. But the global nature of our supply chains and our data infrastructure means that we are all connected to these potential points of failure. The pre-positioning we see from groups like Volt Typhoon is a warning that the next big sabotage event could be much closer to home.
I think that is a perfect place to wrap the core of this. We have looked at the tradecraft, the transition from intelligence to kinetic impact, and the global shift toward this grey zone warfare. It is a sobering look at how the world actually works behind the headlines. We promised to look at how this happens before the war starts, and I think we've mapped that transition from intelligence to physical impact pretty clearly.
It is the world we live in. The most effective weapons are the ones that never officially fired, but they have already changed the course of history.
Well, I am going to be looking at every flickering light and every news report of a minor fire a lot more suspiciously from now on. You have ruined my peace of mind, Herman, but I suppose that is the point of the show.
I prefer to think of it as being realistically informed. Peace of mind is often just a lack of information.
Spoken like a true nerd. We should probably get to the practical side of this before we leave everyone completely paranoid. If you are in the industry, what is the actual takeaway?
The takeaway is that you need to audit your physical infrastructure with the same intensity you audit your code. Look for those single points of failure that can be reached with a pair of bolt cutters or a simple drone. Red-team your physical security. If you assume the attacker is already inside, how do you limit the damage they can do? That is the mindset shift that is required in twenty twenty-six.
And for the rest of us, it is about being critical consumers of information. When you see a report of a mysterious explosion at a sensitive site, ask yourself what the strategic goal would be and who has the blueprints to make it happen. The truth is usually buried somewhere in that intelligence-sabotage nexus.
I think we covered a lot of ground today. From the streets of Tehran to the bottom of the Baltic Sea, the mechanics of sabotage are the hidden skeleton of modern conflict.
This has been a heavy one, but a really important look at the reality of modern warfare. Thanks for walking us through the technical weeds, Herman. I know you have been waiting all week to dive into those undersea cable specs.
It is a fascinating topic. I could talk about anchor-drag physics for another three hours, but I think our producer, Hilbert Flumingtop, would probably cut my mic.
He definitely would. Big thanks to Hilbert for keeping the show running smoothly behind the scenes. And a huge thank you to Modal for sponsoring the show and providing the GPU credits that power our research and production pipeline.
We really couldn't do this without them. This has been My Weird Prompts. If you enjoyed this deep dive into the world of deniable sabotage, please consider leaving us a review on your favorite podcast app. It really helps other curious minds find the show.
You can also find our full archive and all the ways to subscribe at myweirdprompts dot com. We will be back next time with another prompt from Daniel. Until then, keep an eye on your infrastructure.
And stay curious. Goodbye everyone.
See ya.
