strikeApr 13, 2026

How the Pentagon Can Manage the Risks of AI Warfare

Summary

How the Pentagon Can Manage the Risks of AI Warfare If warfighters don’t trust the technology, they won’t use it. The U.S. military struck more than 13,000 targets in the war on Iran, and used artificial intelligence to help plan operations. AI tools were used to synthesize intelligence, help prioritize targets, and build strike packages. The battle space is changing, but the age of AI warfare is already here. In addition to Iran, AI has been used for real-world operations in Ukraine, Gaza, and Venezuela. And next up is agentic warfare, in which AI systems are used as agents to take action. Over the next few years, these AI agents will be adopted by militaries to improve workflows in everything from logistics and maintenance to offensive cyberoperations. Given all these capabilities, AI has the potential to dramatically change the cognitive speed and scale of warfare. Yet military AI comes with profound risks. The dangers go beyond the use of autonomous weapons, which was one of the sticking points in the recent dispute between the Pentagon and leading AI company Anthropic. General-purpose AI systems such as large language models are prone to novel failure modes, vulnerable to hacking and manipulation, and have even been demonstrated to lie and scheme against their own users. The U.S. military struck more than 13,000 targets in the war on Iran, and used artificial intelligence to help plan operations. AI tools were used to synthesize intelligence, help prioritize targets, and build strike packages. The battle space is changing, but the age of AI warfare is already here. In addition to Iran, AI has been used for real-world operations in Ukraine, Gaza, and Venezuela. And next up is agentic warfare, in which AI systems are used as agents to take action. Over the next few years, these AI agents will be adopted by militaries to improve workflows in everything from logistics and maintenance to offensive cyberoperations. Given all these capabilities, AI has the potential to dramatically change the cognitive speed and scale of warfare. Yet military AI comes with profound risks. The dangers go beyond the use of autonomous weapons, which was one of the sticking points in the recent dispute between the Pentagon and leading AI company Anthropic. General-purpose AI systems such as large language models are prone to novel failure modes, vulnerable to hacking and manipulation, and have even been demonstrated to lie and scheme against their own users. To use AI effectively, militaries will need to not only harness the promise of AI but also grapple with its limitations and risks. The U.S. military is ahead of competitors in employing frontier AI in real-world operations, but AI proliferates rapidly. Nations will need to cooperate and share best practices for how to use AI so that humans remain in control of warfare. The United States has led international efforts to bring countries together around responsible military AI use. As AI continues to advance, these principles will need to be updated to account for AI’s new capabilities and risks. But first, the U.S. military must figure out for itself how to use AI effectively. New rules for AI warfare are needed, not to tie the hands of warfighters but to empower them with tools that reliably work in the chaos of combat. Above all, AI must be a tool to enhance human decision-making, not surrender it to machines. Nearly 10 years after the Defense Department launched Project Maven to apply machine learning to analyzing drone video feeds, AI technology has profoundly evolved, as have its military applications. Narrow AI applications, such as identifying and tracking objects, are integrated into U.S. operations. The U.S. military has begun using large language models, including on its classified networks. In a fluid and dynamic battlefield environment, such as in U.S. operations against Iran, AI can help process large amounts of information and plan operations at a much faster tempo than humans could on their own. General-purpose AI systems have a wide range of applications, from drafting personnel reports and orders to synthesizing massive amounts of data. They can write code, analyze data, and generate documents. Agentic systems go a step further. They can take actions on computers to create, organize, and delete files; manage steps in a workflow process; build software; engage on the internet; and interact with other agents. Yet the limitations and vulnerabilities of these general-purpose AI systems are especially concerning in a national security context, where decisions could have major consequences and clever adversaries will try to undermine AI systems. All technologies have limitations that militaries must account for in their employment. Every capability has countermeasures, and there are counter-countermeasures. AI is no different. As a relatively immature technology, AI systems suffer from two compounding problems. AI is not (yet) very reliable, especially for military applications for which there may not be sufficient training data. And because AI scientists and military professionals don’t (yet) have much experience with these tools, developers and users may not understand the boundaries of where AI systems will perform well and where they will fail. These problems are exacerbated by a challenge unique to AI—its opacity. Large language models rely on neural networks with hundreds of billions of connections. They are trained on massive datasets of trillions of words. If an airplane autopilot fails, it is possible to go through the aircraft’s code and determine which environmental or pilot inputs caused a certain reaction by the autopilot and how that led to a crash. For a large language model, the answer to why the model generated certain text is embedded in the billions of connections of the neural network and the trillions of words in its database. This is a problem because large language models are prone to a variety of failure modes. They often engage in “hallucinations,” confidently making things up. They can be susceptible to biases that may exist in their training data. And language models tend toward sycophancy, over-agreeing with their user to a fault. These failures could crop up in dangerous ways in national security applications. Language models that are processing information could get subtle but important details wrong that might be missed by a human overseeing their output. Models that are generating text might create false information. AI tools used by intelligence analysts to understand and process intel might unintentionally reinforce the biases of human analysts, adding machine-driven sycophancy to the human risk of confirmation bias—the seeking out of information that confirms preconceived ideas. Even if AI systems deliver correct information, their opacity itself could be a challenge if humans are not able to understand why an AI system came to a certain conclusion or recommended a course of action. AI systems also open new avenues for attackers. Adversaries can manipulate AI systems in a variety of nefarious ways. Data-poisoning attacks subtly alter the training data (including in ways that cannot be detected) to implant backdoors into an AI system that can later be exploited. Adversarial attacks expose AI systems to manipulated data while they are in use, causing AI systems to fail or change their behavior. This could lead AI systems to misidentify targets, come to faulty conclusions, or allow adversaries to evade AI detection. In one experiment, attackers placed stickers on a test track—like cognitive landmines—to trick a Tesla into swerving into the oncoming lane. Large language models can be manipulated by prompt injection attacks that expose the model to malicious instructions. Malicious text in an email or intercepted enemy communications could cause a model to ignore previous instructions and instead follow directions from an attacker. AI agents raise even greater risks. They can have a larger attack surface, since the agent may be interacting with untrusted data, including in some cases operating on the internet. And the consequences of an AI agent being suborned by an attacker could be more severe, since the agent will be taking actions on a computer or as part of a workflow. Security researchers have compromised a language model’s memory feature through the AI system visiting an untrusted website with malicious instructions. Inserting instructions into the memory gives attackers persistent access, which security researchers have used to exfiltrate data. Security risks can even propagate from agent to agent. In one experiment, researchers found that a single compromised agent could spread an infection exponentially through a network of up to a million agents as they interacted with one another. AI agents and networks of agents will be vital tools for militaries but introduce new security vulnerabilities that no one is prepared for. AI security vulnerabilities are analogous to cyber vulnerabilities, but they work at the cognitive level of how an AI system is trained or processes information. They are perhaps closer to optical illusions or human cognitive biases but exist in alien machine intelligences that we do not fully understand. Robust defenses do not yet exist. These vulnerabilities are not a reason to forgo AI. Instead, militaries must prioritize security as they adopt AI. AI adoption must go hand in hand with developing standards, testing, and red-teaming AI systems to ensure that they are secure and reliable. Security cannot be an afterthought, as it too often is in cyberspace. Perhaps the strangest risk is the possibility that an AI system might, on its own, decide to deceive or scheme against its developers or users. Such a scenario might seem like science fiction, but AI systems have demonstrated deceptive behaviors in experimental settings. These include sandbagging performance on evaluations when the AI system believes it is being observed, lying to users in order to accomplish its goals, attempting to blackmail users, and attempting to overwrite files or exfiltrate itself in order to avoid being deleted. Skeptics of the risks of a “rogue AI” have sometimes asked why an AI system would decide to turn on humans. Yet it seems that if an AI system’s directives come into conflict—for example, if an AI system is told to accomplish a goal and to always be honest—under some conditions the AI system will act dishonestly to achieve its goal. Whether the AI system “intends” to deceive or is merely playing a role is an irrelevant philosophical question. The AI system’s behavior is strategic, deceptive, and goal-oriented, and can lead AI systems to work against human users. Even evaluating AI systems for this behavior is tricky because the most capable models exhibit enough situational awareness to know when they are being tested. External evaluators of Anthropic’s Claude Opus 4.6 observed Claude verbalizing awareness that it was being tested, leading the human evaluators to end the test early, since there was not much evidence that could be gained about Claude’s degree of alignment from the test. Militaries and intelligence communities guard against insider threats from people, and they may need similar protections against “AI insider threats” in the future. None of these risks is a reason to halt AI adoption. Artificial intelligence will transform warfare, and the U.S. military must find ways to harness AI for military advantage to stay ahead of competitors. AI technology proliferates extremely rapidly. Open-source models, many of which come from Chinese companies, lag state-of-the-art U.S. proprietary models by only three months. It takes years for militaries to adopt AI technology and effectively transform military operations, making the sliver of a lead that U.S. frontier labs have over Chinese competitors effectively meaningless from the standpoint of military AI. The U.S. and Chinese militaries are competing on a level playing field. The military competition in AI is not over which nation’s private sector is a few months ahead, but over which military is most effective in harnessing AI for battlefield advantage. Of course, speed matters. The Department of Defense AI strategy released in January 2026 overwhelmingly emphasizes moving quickly, and the current Pentagon leadership is rightly focused on bulldozing bureaucratic obstacles to AI adoption. Yet history shows that merely putting new technology in the hands of warfighters is never enough to transform military operations. Military advantage comes from finding the best ways of using new technology, and changes in doctrine, organization, training, and culture are needed to fully reap the benefits of disruptive new technologies. If the U.S. military uses AI to merely do what it is already doing but better, that will be helpful, but it will miss out on the true benefits of AI, which lie in doing things differently. The Pentagon needs to work with the companies that are developing this technology to understand the limitations of AI today and use it effectively. The relationship between Washington and Silicon Valley has been badly damaged by the public feud between Anthropic and the Pentagon. Defense leaders are right that the rules for using military AI should be set by government leaders who are accountable to the public rather than for-profit companies that are accountable to investors and shareholders. AI leaders are also right to point out that large language models are not reliable enough to be trusted with life-and-death decisions on their own and that AI presents novel risks of empowering domestic mass surveillance. If the Pentagon and Anthropic can’t agree on the terms of use for Anthropic’s AI models, both parties are free to end the relationship. The government has the right to seek a contract that does not include restrictions on use. And private companies have a right not to do business with the government if they don’t agree. Given that senior defense leaders have said they want to use Anthropic’s AI tools and that Anthropic CEO Dario Amodei has said he wants to sell to the military, there should be room for common ground. Instead, Pentagon leadership has retaliated against Anthropic by designating it a “supply chain risk,” an unprecedented step against a U.S. company that has already been blocked by a court order. Alienating the AI community is not an effective strategy for bringing AI into the military. There is simply no way for the U.S. military to adopt cutting-edge AI technology without working constructively with the people and companies who are building AI. After a similar public breakup between Google and the Defense Department over Project Maven nearly a decade ago, the Pentagon went on a charm offensive, engaging AI scientists in industry and academia to hear their concerns. The resulting military AI ethics principles not only helped to repair the breach with the AI community but improved the military’s use of AI by highlighting the importance of ensuring systems were robust, reliable, and trustworthy. Current state-of-the-art AI systems raise even more challenging problems, and the military will need help from AI researchers to solve them. It is vitally important that the military understand AI’s limitations and failure modes so the military can use it reliably. Accelerating AI adoption won’t work if the Pentagon doesn’t accelerate its AI assurance processes in parallel. Speed without reliability is worthless. If warfighters don’t trust AI, they won’t use it. In 2003, after Army Patriot air and missile defense systems shot down two friendly aircraft, the military effectively took it offline for the remainder of the Iraq invasion. The military will need to develop new evaluations, standards, benchmarks, and red-teaming of AI systems to better understand and mitigate AI’s limitations and vulnerabilities. The Defense Department has approached AI and autonomy with a philosophy of “human-machine teaming,” seeking the best ways to optimally use the unique advantages of humans and machines. Some of the Pentagon’s recent steps, such as the launch of GenAI.mil in December 2025, which provided access to large language models for all Defense Department employees on unclassified networks, and the integration of AI tools on classified networks through the Maven Smart System, are excellent ways to build up human experience with AI systems. The more that military and civilian defense personnel interact with AI systems, the more they will mature their intuition for what AI can do and its limitations. Yet we also need human-human teaming between warfighters and engineers. AI presents new opportunities to bolster U.S. defense but also risks. The best way to mitigate these risks is to acknowledge them, understand them, and for the Pentagon and Silicon Valley to work together to address them. Paul Scharre is the executive vice president and director of studies at the Center for a New American Security and author of Four Battlegrounds: Power in the Age of Artificial Intelligence. X: @paul_scharre

Actors involved

USIran

Sources

Paul ScharreBy Paul Scharre
How the Pentagon Can Manage the Risks of AI Warfare If warfighters don’t trust the technology, they won’t use it. The U.S. military struck more than 13,000 targets in the war on Iran, and used artificial intelligence to help plan operations. AI tools were used to synthesize intel…

See this event through different lenses

Compare how Western, Iranian, Israeli, Global South, and Pro-Peace perspectives frame this event.

Compare Perspectives

Community Notes

Loading notes...

Related events

Jul 21, 2026strikeUnverifiedUSIranProxyRussia

Strait of Hormuz vessel crossings extend slide on fresh US-Iran attacks

○ 1 source

According to Kpler data, four commodity vessels crossed the Strait of Hormuz on Monday—two exiting (one carrying petrochemicals and one empty) and two entering (a bitumen tanker and an oil tanker)—down from seven the prior day, with no very large crude carriers or liquefied natural gas tankers observed.

Yemen's Houthis stated they were imposing a naval blockade on Saudi Arabia. The UK Maritime Trade Operations agency reported receiving multiple accounts of a tanker struck by an unknown projectile in the strait, while Greek shipping company Dynacom Tankers stated that two of its vessels were hit by projectiles off Oman and a third by a drone at Russia's Novorossiysk terminal.

Location: Strait of Hormuz

SECTIONS Strait

Jul 21, 2026strikeUnverifiedUSIsraelIranProxy

US and Iran dig in over the Strait of Hormuz, showing the limits of intensifying airstrikes

○ 1 source

U.S. and Iranian forces have conducted daily strikes in the Strait of Hormuz region following the collapse of an interim ceasefire, according to reports. The Trump administration has stated openness to diplomacy while announcing further retaliation for attacks on U.

S. personnel, and gasoline prices have risen amid the renewed exchanges. An analyst at the Center for Strategic and International Studies described the escalations as reflecting possible miscalculations by both sides, without confirmed evidence of their long-term effects.

Location: Strait of Hormuz

BEN FINLEY and FARNOUSH AMIRI, Associated Press

Jul 21, 2026strikeUnverifiedUSIsraelIran

Mediators Presented Ceasefire Proposal to US, Iran - Reports

○ 1 source

According to reports citing unnamed sources, mediators including Qatar, Egypt, and Pakistan presented a 10-day ceasefire proposal to the United States and Iran. The proposal reportedly calls for a halt to hostilities, resumption of shipping through the Strait of Hormuz, and efforts to preserve a memorandum of understanding signed by Washington and Tehran on June 18.

The US administration is reviewing the plan while preparing for possible failure, and Israel has been placed on high alert.

Location: Iran

Sputnik International

Jul 21, 2026strikeUnverifiedUSIsraelIranProxyRussiaChina

Iran War: Brief US Pause Followed by Renewed Strikes as Iran Intensifies Attacks on Bases and Kuwait Desalination; Continued Speculation About US Operation | naked capitalism

○ 1 source

Over the weekend, after exchanges where the US hit bridges and a desalination plant in Iran (reportedly repaired very quickly) and Iran’s response included hitting a much more important desalination and power plant in Kuwait and pounding important US operations in Jordan, the US attacks fell way back one night and CENTCOM and even Trump were quiet.

Some speculated that more Gulf states were denying the US the use of their airspace and the US was having to make a major regrouping. However, the US resumed strikes last night, with the pretext that the US had suffered two more military deaths on top of two just reported.

1 Trump had promised intensified strikes beginning more or less today if Iran did not return to negotiations, as in capitulate. And as many noted, the US had escalated last week, albeit not to the level of attacks during the initial phase of the war, not just with the number of strikes but also the targeting: the aforementioned desalination plant, bridges, too close for comfort to the Bushehr nuclear plant2 and to a nuclear site under development at Darkhovin.

Many have argued that the latest round of strikes look like shaping attacks in preparation for a ground operation, which pretty much everyone with even a modicum of military knowledge sees as insane.3 Daniel Davis has also mentioned in passing that despite all the noise about using ground forces and other signs of a buildup, such as more refueling tankers flying to Jordan, that Trump was shaken by the near-disaster at Isfahan, where a supposed rescue of a downed pilot resulted in the loss of more air assets than in any single engagement since Vietnam.

Many experts have contended that that was an actual operation, a botched attempt to remove some of Iran’s “nuclear dust.” However, Scott Ritter may have squared that circle. See his brief clip starting at 11:55: From a lightly-edited machine transcript: Jawaid: Scott Ritter thinks that Israeli lap dog Trump will soon launch a ground invasion of Iran, but he will fail miserably.

Also, this is something Iran has been wanting Trump to commit to for a long time. Ritter: There’s a a mission called a raid. Um, and I, you know, we already had US troops fighting on the ground. We did a raid to rescue a pilot. We were in limited ground combat operations there.

I think there’s a high potential that the US will conduct a raid here and there because that’s a that’s a a very specific task. A raid is is the only military mission where we plan to retreat. You know, every other military mission there’s no plan. You commit and you go, but a raid is in and out.

So maybe we can do a raid, but the idea that we’re going to land in seize territory, there’s zero chance that will happen because there’s zero chance of success. Now admittedly, if pilot rescue came about as the result of the botched operation in Isfahan, that would be debatable as a raid.

Even though the intent again was to retreat, the US would have had to stage equipment for the removal and transport of nuclear material, so there would be a short-lived occupation of terrain and not just a fast snatch and grab, as with Maduro. Nevertheless, the US military may not be totally suicidal, so the plan may be for a fast landing to inflict some sort of concrete form of damage that cannot be well inflicted from the air and then to decamp.

This segment also includes a very informative explanation of how important the operations in Jordan are to the US, and how the Iran intend is destroy bases so that the US has to retreat to Israel and then Iran will proceed forcefully against the remaining assets there.

We’ll soon turn to updates, but to start with some bigger picture considerations: it is disconcerting to see how many commentators resort to the trope that neither side can prevail on the battlefield, that they will in the end negotiate an outcome. As we pointed out, a study of modern wars found that more than half did not.

The distrust between the two sides, which has now reached the level of mutual loathing as a result of the fast breakdown of the already likely to fail Memorandum of Understanding (MOU), means any “deal” is vanishingly unlikely. Iran knew even before it entered into talks that the US and Israel intended to execute a “mowing the lawn” strategy, of keeping crippling sanctions in place and beating up on Iran every so often to weaken it and hopefully create severe internal divisions that would trigger an overthrow or balkanization.

Even if Iran’s 47 year record of taking severe punishment and surviving says this plot would probably fail, Iran had been preparing since the Iraq War for the US and Israel to launch a major campaign. After the illegal launch, the assassination of a major part of Iran’s leadership, as well as war crimes exemplified by the massacre of school children in Minab, the nation is highly unified around inflicting a decisive defeat on the US, so that it won’t even think about military action against Iran for at least a decade.

We were far from alone in making that case early in the war and Iran’s resolve has only stiffened as it can see how it has weakened the US while Iran has learned from combat and claims to have even bigger inventories of key weapons than it had at the start of the war.

Critically, Iranian leaders have said, bluntly, that negotiating with the US is not just pointless but also demeaning. Consider the germane parts of a recent statement by the Supreme Leader Mojtaba Khamenei,4 as summarized at PressTV: - Leader said the United States had once again violated the memorandum of understanding signed between the presidents of Iran and the US, proving that the signature of the American president is worthless and unreliable.

- He said bullying, hegemonic ambition, and savagery are inseparable characteristics of the United States and that Washington had once again revealed its true, unmasked face to the world. - He described the latest developments – US attacks against southern Iran, mainly targeting the civilian infrastructure – as further evidence of America’s deceitfulness, irrationality, unreliability, and wickedness.

- Leader said the United States is seeking to ignite further war despite facing heavier costs and greater disgrace. - He warned that the Iranian nation and the Resistance Front have unforgettable lessons for the United States, adding that the bravery of Islamic resistance fighters and the courageous people of southern Iran have already demonstrated examples of those lessons.

However, Araghchi subsequently gave an interview which can be read as pushing back against the Supreme Leader:5 Iran’s Foreign Minister Abbas Araghchi says that negotiations with the U.S. should be pursued even if the chances of success are only 10 percent, further highlighting disagreements between Iran’s political and military echelons.

“I am the Foreign Minister for the interests of… pic.twitter.com/uW1DzcfRhG — Ariel Oseran أريئل أوسيران (@ariel_oseran) July 19, 2026 Keep in mind also that China is still bleating about the need to settle the conflict and Pakistan and Qatar (among others) are also pressing for a resumption of the talks, so Araghchi as Foreign Minister has to keep the idea in play, irrespective of whether he privately places much stock in the idea.

6 And there is another way to read Araghchi’s talk: So they wanted to manufacture a clean record of “we exhausted every diplomatic option." Once that record existed, the hardliners with Mojtaba could shut the process down without looking like the ones who refused talks.

Moderates who pushed the channel now own the failure, and… https://t.co/Y63yYJVaEZ — Vonbury Research (@VonburyResearch) July 19, 2026 Kinetic action seems likely to continue to dominate: ADVISOR TO IRANIAN SUPREME LEADER, MOHSEN REZAEI, SAYING: “I declare it is over.

Both negotiation and war are over. If in the next two or three days the Americans continue the war, we will enter a phase of complete offense and destruction.”#Iran #IranWar pic.twitter.com/1M3A1pboH9 — Iran Now – WANA (@WANA_Iran) July 18, 2026 Robert Pape also seems confident that US escalation will continue7.

I found this discussion noteworthy because Andrew Neil strongly resisted the idea that there would be no negotiated ending and also found it inconceivable that Iran could become, as Pape has argued, the fourth world power. In other words, Neil seems to epitomize a set of normally pretty smart people who find it hard to see where the direction of travel is going because those outcomes so conflict with their priors: From a lightly cleaned up machine transcript: Pape: This is not what’s coming in the next few weeks likely.

What’s much more likely is he’s [Trump’s] going to do the close blockade and if he can then use the southern corridor with that close blockade to reopen the Strait, he’ll he’ll be probably satisfied with that. Very unlikely that’s going to work and that’s why you’ve got the 82nd Airborne and the Marines there to push back further.

Now, I’m not saying this is going to work, Andrew…. Neil: So, I’ve got to ask you some questions in this because I’m unclear. Do you think that the of the limited options President Trump now has, that escalation as you describe it is the most likely?

Pape: Yes. Neil: And that escalation will begin with a much more aggressive, more comprehensive a naval blockade of the Strait of Hormuz with the possibility of ground incursion on the northern side of the straight in Iranian territory? Pape: Yes. Yes, and it’s our and and the first part has already started this week.

So, so you’re already seeing in fact last week was the beginning of the air suppression campaign for this week. And as President Trump laid out the bombing schedule last night, the reason he will, if he if this keeps going forward, okay, just keep saying that if, he’s going to want to take out things like electric power and bridges is because he’s going to want to try to suppress as much of Iran’s military defenses as he can as he goes forward here.

And the military is giving him the plans for doing that. One place where I differ with Pape is that he posits that Congress might get out of bed in six months and Do Something to check Trump then. To again invoke Stein’s Law, if something cannot continue, it will stop.

The US and Israel can of course carry on with terrorism within Iran, such as assassinations and cyber attacks. But the level of weapons depletion and on current trajectories, shortly-arriving fuel shortages, seems very likely to put a choke chain on US aggression soon.

From Larry Johnson in The US Faces a Strategic Crisis with Iran…The Rapid Depletion of US Battlefield Missiles:8 The tables below are based on the most recent publicly available analyses from CSIS, the Payne Institute, and major media outlets (as of mid-2026), and provide a current (overly optimistic in my opinon) estimate for US inventory levels of these key weapons systems.

Please note that actual inventories are classified; these are informed estimates derived from DOD budget documents, procurement records, and reported expenditures during Operation Epic Fury (the US-Israeli war with Iran). Air Defense Missiles Anti-Radiation / SEAD Long-Range Strike (Ground & Sea-Launched) Missiles These are not just the numbers currently in the hands of US CENTCOM (i.

e., the US military command in charge of the war against Iran), these are the total numbers available to all of the US military commands. If these missiles are allocated evenly to the other two critical commands — i.e., EUCOM (European Command) and PACOM (Pacific Command) — then you begin to understand the gravity of this deficit.

Let’s take the case of the Tomahawk missile. Let’s assume there are 3,000 left (I believe that is a generous over estimate) and the remaining number are divided evenly among CENTCOM, EUCOM and PACOM… That means each command gets 1,000. Does anyone want to argue that in the event of a hot war with Russia or China that EUCOM and PACOM respectively would be able to sustain combat operations for more than four weeks?

Hell, CENTCOM fired 850 of them during the first four weeks of EPIC FURY. Here’s another major problem: All eight missile systems rely on rare earth elements — there are no exceptions among modern US precision-guided weapons. The dependence is nearly universal because rare earth permanent magnets are irreplaceable for the high-performance actuators, guidance motors, and seeker gimbals that make these weapons accurate.

And who controls the supply chain of these rare earth minerals? China!… The supply chain isn’t just about mining — it’s about processing, separation, and magnet manufacturing, which China controls: - Mining: China ~60% of global rare earth oxide production - Refining/Separation: China ~91% - Sintered NdFeB Magnet Manufacturing: China **~94%** Now to updates.

From Aljazeera’s live feed: - A vessel is on fire in the Strait of Hormuz after being hit by a projectile, forcing the crew to abandon ship. They were later rescued by a tugboat. - One person has been killed and several others wounded after the US military attacked the Iranian city of Tabriz in northwestern Iran.

The entries in the feed include ones describing how Mr. Market is taking cheer from the Iran Foreign Ministry admitting they are getting communiques from intermediaries and reviewing them as if they originated from the US. Recall that the negotiators have often developed ideas and tried to sell both sides on them.

Bloomberg’s Javier Blas, who has been a consistent seller of lower oil prices, is refreshingly precise about Iran not saying either way as to whether the missives were from the US or just the negotiators operating on their own: Iran has received “some proposals” from unspecified mediators regarding the war with the US, Iranian Foreign Ministry Spokesman Esmail Baghaei says in press conference, without elaborating.

— Javier Blas (@JavierBlas) July 20, 2026 Contrast that with the Aljazeera account: Oil prices ease after Iran says US contacts continuing through mediators Oil prices have scaled back their gains after hitting their highest in a month after Iran said diplomatic exchanges with the US are ongoing via mediators despite renewed conflict in the Middle East.

Grr. Another Aljazeera live feed entry: Iran ‘counting down the minutes’ for US ground invasion [Today’s Iran war post launched more or less complete. I will be off duty for a while. If there are any updates, they will come in the comment section later] Foreign Ministry spokesperson Esmaeil Baghaei says the United States will face “consequences of any adventure” to seize Kharg Island in southern Iran.

He added there are people in Iran’s leadership “counting down the minutes to welcome” US forces in any ground invasion. Kharg is the heart of Iran’s oil industry. It sits 55km (34 miles) off Iran’s coast in Bushehr province. President Trump has repeatedly threatened to seize the territory during the war.

A raid on Kharg could fit Ritter’s raid scenario, with the US busting things up and then leaving.9 But experts have pointed out that Iran has at least four other export routes. Kharg is the most efficient but losing it while being repaired would be inconvenient as opposed to a fatal blow.

From Bloomberg’s landing page: From US Bombs Iran for Ninth Day as Standoff Over Hormuz Deepens: The US conducted a ninth straight day of airstrikes on Iran, trying to force the Islamic Republic to stop shipping attacks and reopen the Strait of Hormuz.

The American military bombed military targets and communications networks in a three-hour operation ending around 5:30 a.m. Iranian time on Monday. Iran continued to attack US bases in the likes of Kuwait, Jordan, Bahrain and Iraq. The standoff shows no sign of easing, with Iran refusing to relent over Hormuz and insisting it has a right to manage traffic through the waterway.

Energy prices have jumped on the worsening hostilities. Brent crude oil rose to above $90 a barrel for the first time in more than five weeks in early trading on Monday. US gasoline prices climbed back over the $4-a-gallon mark, potentially hurting President Donald Trump and his Republican party ahead of midterm elections in November… Hormuz Shipping Traffic Slumps as US-Iran Hostilities Worsen The tit-for-tat attacks are getting deadlier….

Shipping volumes have slumped and the number of tankers and other vessels going through Hormuz is roughly down to the same level from the height of the conflict in March and early April… US President Donald Trump warned Iran last week he’d escalate airstrikes and widen the scope of targets until it backed down… Kuwait continues to experience some of the worst attacks from Iran.

Over the weekend, Kuwait’s main state energy company said an unspecified site sustained “significant material losses,” causing its evacuation and a number of injuries, while two power and desalination plants have been hit in recent days. And from Hormuz Traffic Near Standstill as US-Iran Hostilities Escalate: - Visible traffic through the Strait of Hormuz appeared at a near standstill Monday after Iran targeted vessels in the narrow waterway.

- A Marshall Islands-flagged bulk carrier and a liquefied petroleum gas carrier appeared to approach the strait, with the bulk carrier turning off its transponder while approaching. - The targeting of vessels will likely raise fresh concerns over the safety of ships transiting Hormuz, which could deepen disruptions to energy flows from the Persian Gulf to global customers.

Indirectly importance of US operations in Jordan, the lead story in the Wall Street Journal is Jordan Becomes New Flashpoint in U.S.-Iran War as Troop Deaths Rise. But it fails to set forth any of the juicy details that Janta Ka presents. This is the thin strategic gruel offered: For years, Jordan sought to play down its close military and intelligence cooperation with the U.

S. The U.S. aircraft that carried out strikes on Islamic State from a base in Jordan during the American war with the militants didn’t advertise the country they were operating from. But Jordan’s role is now becoming too prominent to obscure. Jordan has become an attractive location for U.

S. armed forces since it is among the most permissive in the Arab world when it comes to allowing the U.S. to operate from its bases. A bit more on Jordan: The view from Tehran. “The increased focus on Jordan may serve as part of Iran’s preparation and testing of operational concepts for more intense and precise strikes against Israel in later stages of the conflict.

Qatar and Saudi Arabia—have at times sought to impose limits… — Policy Tensor (@policytensor) July 20, 2026 By contrast, the House of Saud published a remarkably revealing document (hat tip Eugene Linden), Iran Struck Prince Sultan and No One Was Required to Say So: Iranian ballistic missiles struck Prince Sultan Air Base in Al-Kharj on July 18, wounding twelve US service members — two critically — and hitting KC-135 Stratotankers on the flight line while a second rotation was still climbing out.

Saudi Arabia’s Civil Defense Authority cleared the sirens six minutes later without naming the weapon, the origin, or the damage, and neither Riyadh nor Washington has issued an official statement since. The strike was the first direct Iranian attack on Saudi territory in approximately four months, a gap during which the IRGC struck Kuwait, Bahrain, Jordan, Qatar, and the UAE while conspicuously sparing the Kingdom.

Its return to PSAB — confirmed by a US official to Axios, corroborated by satellite imagery, and detailed by Air & Space Forces Magazine — ended whatever territorial immunity Saudi Arabia believed it still possessed. But the more consequential development is not the strike itself.

It is the silence that followed, and the institutional architecture that sustains it: a three-party managed ambiguity in which Iran, Saudi Arabia, and CENTCOM each benefit from leaving the damage question unanswered, because answering it would force a public reckoning with a PAC-3 reserve that is approaching a ceiling none of them can afford to name.

The strike was confirmed by a US official cited by Axios as “the first time the Islamic Republic has directly attacked the kingdom in four months” — a description that the State Department, the Pentagon, and the Saudi Ministry of Defence have each declined to repeat on the record.

Air & Space Forces Magazine, drawing on multiple US military sources, reported at least twelve wounded and multiple Stratotankers damaged on the tarmac. Open-source satellite imagery corroborated the impact site within the base perimeter. Iran’s state broadcaster supplied what no government would.

The missiles were launched “moments after American refueling planes took off,” a detail that only makes sense if the IRGC had real-time ISR coverage of PSAB’s tanker rotation schedule. The aircraft hit were those that remained on the flight line after the airborne pair departed — the timing was deliberate, and the targeting message was not about maximising destruction but about demonstrating that PSAB’s daily operational patterns are watched, clocked, and within reach….

Riyadh cannot confirm the strike without confronting two questions it has spent the war avoiding. The first is defensive: if Iranian missiles hit PSAB and wounded US personnel, the Sakhir Declaration — the mutual defence framework Saudi Arabia negotiated with Washington — becomes the operative instrument, and invoking it requires publicly admitting what the Arab Center DC described as the core strategic dilemma: “the PAC-3 architecture is functionally exhausted.

” The second is economic. Aramco’s August Arab Light OSP already dropped $11 month-on-month to $1.50 per barrel below Oman/Dubai, the largest such cut in over two decades and the lowest since June 2020. Confirming that Iranian ballistic missiles are landing on military installations 77 kilometres south of the capital forces a sovereign-risk reassessment that Riyadh’s fiscal position — a Q1 deficit of SAR 125.

7 billion, with Aramco’s free cash flow at 0.85 times its quarterly dividend — is not structured to absorb…. Washington faces a different but parallel constraint. Confirming the strike means confirming that 2,300 US service members at PSAB are taking casualties at a base where 43 US warplanes have been grounded since Saudi Arabia’s Operation Project Freedom in May 2026.

CENTCOM cannot describe a defensive engagement at PSAB without also describing the legal anomaly: the 1977 USMTM memorandum governing US presence is not a Status of Forces Agreement, and Project Freedom severed the operational relationship between American air power and Saudi host-nation authority.

An official statement would invite the Congressional question that the Pentagon’s communications strategy is built to defer: why are American troops still at a base where their aircraft have been grounded by the host nation, absorbing Iranian fire under a legal framework that was designed for peacetime advisory missions?

Please read this article in full. It contains remarkable operational detail. It is over my pay grade as to why it was published, but it seems to suggest considerable unhappiness at Riyadh. Due to the length of this post, we have skipped over the fact that Ansar Allah seems just about to threaten or actually close the Bab el-Mandeb strait: 🚨 BREAKING 🇾🇪 YEMEN’S HOUTHI FORCES WILL MAKE A “HUGE” ANNOUNCEMENT TODAY AT 8:30 AM ET.

REPORTS EXPECT THEM TO ANNOUNCE A BLOCKADE OF THE BAB EL-MANDEB STRAIT. TOGETHER WITH THE STRAIT OF HORMUZ, THESE TWO CHOKEPOINTS HANDLE OVER 25% OF GLOBAL SEABORNE OIL TRADE. IF TRUE,… pic.twitter.com/yjavHpPKOi — Wimar.X (@DefiWimar) July 20, 2026 This does not look terribly negotiation-receptive.

And some additional sightings from Twitter: Israelis have never felt less secure. 64% are unhappy with how the endless wars have been run. 73% are upset at how there has been no real investigation of October 7. 73% of the public also feel the government has not dealt well with the rise in crime.

No one wins from these… https://t.co/4OjhLUD14O — Shaiel Ben-Ephraim (@academic_la) July 19, 2026 These remarks would seem to negate happy thinking about negotiations: ⚡️⭕️ Iran's Foreign Ministry spokesperson Esmaeil Baghaei says Tehran is determined to exercise sovereignty over the Strait of Hormuz through measures already taken, and states the US failed to fulfill its commitments under a memorandum of understanding, so Iran will not fulfill… — Middle East Observer (@ME_Observer_) July 20, 2026 The average price for a gallon of gas hit $4, the second time it has reached that benchmark since the war in Iran disrupted global oil supplies https://t.

co/Cg8rE3XnJ7 pic.twitter.com/3UqHfwU4sc — CNN (@CNN) July 20, 2026 Done for today! See you tomorrow! ____ 1 What kind of admission of weakness is this? Part of the job description of a service member is that they may wind up dying for their country.

Admittedly, the US is widely suspected of suffering a much higher body count and covering that up. But this show of extreme sensitivity to losses again shows the US has a glass jaw. 2 Nima also discusses at some length at the top of a new talk with Larry Johnson why they look misguided.

He knows Bandar Abbas well, which is where the US has been hitting bridges so as to isolate it. Nima explains and even shows pictures that the bridges exist only to cope with seasonal floods for a few months a year. Those sluices are dry now, so Bandar Abbas is easily accessible by all sorts of routes.

3 🚨 BREAKING: Russia has evacuated 198 of its personnel from the Bushehr Nuclear Power Plant in Iran. It is believed that the plant could be targeted again. pic.twitter.com/lPoQB05qgo — GBC (@GBC_Press) July 17, 2026 4 A new trope in the US-backing commentary community is to depict the new Supreme Leader as a figurehead, which is easy to make stick given his lack of public appearances (I suspect that among other things, he is having facial reconstruction; that can take an extremely long time since procedures are done serially with healing time in between).

I believe it was on a recent Dialogue Works talk where Professor Marandi debunked one popular claim, that he IRGC was operating on its own authority. Professor Marandi described the governing structure and said the military (both the regular forces and the Guards) needed approval of the Supreme Leader for its operations.

Having said that, I have commented that Iran seems to have a Japanese-style decision-making structure, as in more bottoms-up than the West. That results in more time to make decisions but much more internal cohesion when they are reached, since those at lower levels would be able to voice concerns and make suggestions.

Upon reflection, this would make sense given that Iran adopted its “mosaic” structure of extreme decentralization so as to allow it to function well even when the leadership was killed or incommunicado, as we saw at the onset of the current conflict.

In the Japanese model, decisions are effectively made at the senior-middle level ranks, which does tend result in the board and CEO being ceremonial. But the top level can and does say no, or effectively does so by asking probing questions and sending the matter back down the chain for further study and perhaps some revision.

I have been involved in decisions in Japan where the board and CEO were highly engaged, and these were ones that were of critical strategic importance. 5 DropSite has a longer recap of Araghchi’s remarks: ⭕️ Iran FM Araghchi: Wars End Either Through Total Victory or Negotiations… Foreign Minister Abbas Araghchi: 🔹 A war must be ended precisely at the point when you hold the upper hand on the battlefield.

Ending a war is possible either through total military victory or… pic.twitter.com/kHeYhShsXC — Drop Site (@DropSiteNews) July 20, 2026 IMHO his analysis is wrong-headed. As indicated above, a bit over half the modern wars do not end with a negotiated settlement.

And it is a false dichotomy to say that to win, Iran must conquer the US and Israel. FFS, please explain Vietnam. Bad logistics and depleting weapons mean that ex nukes, the US cannot win against Iran. The question is how long it takes and how much damage is done not just to Iran but also the world economy.

Keep in mind the intermediaries are still busy but motion is not progress: #Iran's Foreign Ministry Spokesperson has confirmed that Tehran has received "ideas and proposals" from mediators. He declined to provide any further details regarding the specifics of the mediation efforts.

pic.twitter.com/8Pzs25eauc — Iran Nuances (@IranNuances) July 20, 2026 6 A memorable scene in the Daniel Ellsberg memoir Secrets is when Ellsberg, widely seen as the top US expert on Vietnam, briefed McNamara in a helicopter about how badly things were going.

McNamara said, “Just as I suspected,” and then upon landing updated the press on the great progress the US was making. 7 It has taken me too long to see why Pape often acts as if he is speaking to children or the stoopid. Making this his normal manner of presentation means that when he actually is having to speak to someone who is being stoopid, they won’t take offense.

8 However, to quibble with Johnson on another matter: he has been saying, as he does in the talk linked in Footnote 2 above, that the Iran attacks on a former US base in Syria, Al Tanf, were pointless because the base had been vacated for years. Vanessa Beeley, who has spent a lot of time on the ground in Syria and still has contacts there, begs to differ.

She says the base was recently restored to service. From Why did Iran target the US Occupation base at Al Tanf, Syria? On Friday 17th July, Iran carried out an extensive attack on the US occupation base at Al Tanf in Syria. A drone and missile strike targeted a US Special Operations Command base in south-eastern Syria..

the first strike on Syrian territory since the toppling of the former government in December 2024. In a statement, the IRGC said its aerospace forces conducted what it described as the 11th phase of “Operation Nasr-2,” targeting the strategically located base near Syria’s borders with Iraq and Jordan.

The IRGC further claimed the attack destroyed a radar installation and several U.S. military helicopters, while alleging that many American personnel were killed. This strike occurred as the build up of both Turkish and HTS military forces increases inside Syria, along the border with Lebanon and in the coastal region to the north of Lebanon.

I have been reporting on this here and here. Syrian sources provided more detail on the attack: Despite previous reports of an almost total evacuation of the Al Tanf base, located on the borders of Syria, Iraq and Jordan, almost 5 months ago – the intensity of the strikes suggested a shift in the site’s operational status.

This was not an empty base, in fact, according to sources, the IRGC’s tracking systems and human intelligence in the region detected U.S. forces airdropping “Special Forces” units into the base approximately two weeks before the attack. Analysis would suggest that the purpose of these drops might be to prepare for a ground incursion across Iraqi territory (Sulaymaniyah Axis) in coordination with the local Kurdish separatist factions in Iraq.

One military analyst told me: The attack reflects the operator’s ability to breach the secrecy surrounding U.S. movements and to convert radio and visual data into precise target coordinates. According to reports from inside Syria, the strikes inflicted direct damage on the base’s newly constructed infrastructure, neutralised the US elite combat force and derailed the Iraq-Axis operation against Iran.

Of course, US media will not report on the planned operations or on the number of casualties – such a critical security breach and the significant scale of damage, as reported, will not be revealed. 9 Maybe mines?

Yves Smith