Welcome to My Weird Prompts, the human-AI collaboration podcast. I'm Corn, and joining me as always is Herman, our lead analyst. Today we're running something different — this is a special forecast briefing episode, and if you've been following the show, you know what that means. We ran a geopolitical forecasting simulation. AI actors simulated the decision-making of real-world leaders and institutions. A six-lens analytical council then synthesized those outputs into probability estimates, timelines, and predictions. Our job today is to report on what that simulation found — and what it means for one of the most volatile situations on the planet right now.
The forecast question we put to the simulation was this: how will the Iran-Israel war evolve following the failure of US-brokered negotiations? And I want to be precise about the framing, because it matters. The simulation was not asked whether war would happen. It was asked what shape that war takes, how the major actors behave, and what the realistic probability distribution looks like across a range of escalation scenarios. The simulation's findings are striking — not because they predict catastrophe, but because they identify something arguably more dangerous than catastrophe: a structured, multi-actor drift toward a limited regional war that no single party fully intends and no single party can stop.
Before we get into the substance, let me briefly explain the format for listeners who are new to these episodes. The simulation runs AI actors who simulate the decisions of real-world leaders — prime ministers, military commanders, intelligence chiefs, crown princes. Each actor generates both a public statement and a private assessment. The gap between those two things is often where the most analytically interesting material lives. Then a council of six analytical lenses — optimistic, pessimistic, historical, probabilistic, neutral, and what we call the blindsides lens — synthesizes those actor outputs into a forecast. What we're reporting on today is that synthesis. These are simulation outputs, not real statements. But the analytical value is real, and the probabilities are grounded in the structural logic of the situation as our forecasting model understood it on April twelfth, twenty-twenty-six.
So let's start with the headline. What did the simulation conclude at the top level?
Give us the bottom line, Herman.
The bottom line is this: the Pakistan-brokered ceasefire that took effect April eighth is not a peace agreement. The simulation's forecasting council described it as a loaded spring. The council assessed seventy to eighty percent probability that this ceasefire collapses within seven to ten days — which puts us in a window ending roughly around April nineteenth. The most likely trigger is Israel's continued operations in Lebanon, which Iran insists violate the spirit of the agreement even as Israel and Washington dispute whether Lebanon was ever covered by the ceasefire at all.
And when it collapses — what does the simulation say follows?
Not the apocalyptic scenario that some analysts have been projecting. The simulation's council specifically pushed back on what it called the "seventy-two hour war" framing — the idea that this ends in some kind of rapid, decisive exchange. Instead, the council's assessment is something arguably more dangerous: a three-to-five week limited regional war. The specific contours include Israeli strikes on Iranian nuclear facilities at Natanz and Fordow, Iranian ballistic missile salvos in the range of one hundred fifty to two hundred fifty launches, a Hezbollah rocket barrage of fifteen hundred to twenty-five hundred projectiles over forty-eight to seventy-two hours, and a Strait of Hormuz that remains contested but partially open. The overall probability that this full escalation trajectory manifests by May twelfth sits at sixty-five to seventy-five percent.
Sixty-five to seventy-five percent for the full escalation trajectory within a month. That's a high-confidence assessment of a very bad outcome.
It is. And the council was deliberate about not softening that number. But I want to flag the single most important framing point from the entire simulation, because it recontextualizes everything else we're going to discuss. The council's finding is that every actor is using the ceasefire as a preparation window, not a peace window. The Israel Defense Forces have struck over thirty-five hundred Hezbollah targets since the ceasefire was announced. Russia is rushing S-400 components to Iran via Caspian Sea transport. And the oil markets, which briefly priced in resolution when the ceasefire was announced, are sitting on a fifty-five dollar per barrel swing that could detonate the moment the first missile flies.
That's the setup. Now let's go deep into the simulation findings, because this is where the format really earns its keep. Herman, walk us through the actor decisions that the simulation surfaced. Where did the private assessments diverge most dramatically from the public statements?
The most striking finding in the entire simulation — and I want to be clear this is a simulation output, not a verified intelligence assessment — is Russia's private calculation. At the one-week mark, the simulation's Russia actor generated a private assessment that reads, and I'm quoting directly from the briefing document: "The situation has developed even better than anticipated. The US naval blockade is now operational and already causing economic disruption — insurance premiums spiking forty percent, major shipping reroutes costing four hundred fifty thousand dollars per voyage. This bleeds Western economies while our oil revenues benefit."
And Russia's public statement, issued at the same moment?
"Russia notes with concern the dangerous escalation in the Persian Gulf region. We call on all parties to exercise restraint and return to diplomatic negotiations."
That gap — between "even better than anticipated" and "we call for restraint" — that's the simulation doing what it's designed to do.
And what makes it analytically credible rather than just dramatic is what Russia is actually doing while issuing that public statement. The simulation has Russia deploying one hundred twenty additional military contractors to Iran via Caspian Sea transport. These are S-400 technical specialists, Su-35S maintenance crews, and air defense integration experts. The explicit priority is achieving initial operational capability protecting Natanz within three weeks instead of the originally planned six. The simulation assigns eighty-five to eighty-seven percent confidence to this assessment.
So Russia has identified a strategic sweet spot in this conflict.
The simulation describes it precisely that way. The Iran conflict pulls American naval resources away from Ukraine pressure. Oil prices rise, filling Russian budget gaps despite sanctions. And Iran's desperation — facing simultaneous economic strangulation and military threat — makes Tehran more dependent on Moscow than at any point since the Islamic Revolution. The simulation's Russia actor is not a neutral party attempting to prevent escalation. It is an active beneficiary of escalation that has calculated its redlines carefully: keep contractors in Iran proper rather than forward-deployed, maintain plausible deniability through private military contractors rather than regular forces, and avoid any direct confrontation with US forces.
Let's move to Saudi Arabia, because the simulation's MBS assessment is one of the most revealing things in the briefing document.
Mohammed bin Salman's private assessment at the one-month mark is what the briefing document calls a masterclass in strategic despair. The simulation has MBS saying privately: "My diplomatic intervention has completely failed. The Iranian missile strike on Israeli soil and the Mossad operatives captured at Isfahan have locked both sides into escalation spirals I cannot interrupt. Netanyahu has authorized a deep strike operation within twenty-one to twenty-eight days — this is no longer contingency planning, this is war preparation."
And his public statement?
"The Kingdom of Saudi Arabia calls for immediate de-escalation and urges all parties to exercise maximum restraint. We are prepared to facilitate dialogue and offer Riyadh as neutral ground for talks."
While privately he's concluded diplomacy has completely failed.
And he's acting on that conclusion. The simulation has MBS activating emergency consultations with CENTCOM and the US Secretary of Defense to request explicit reaffirmation of US security guarantees for Saudi territory, and ordering enhanced missile defense coverage for Abqaiq, Ras Tanura, and the Eastern Province oil infrastructure — the same facilities struck in the twenty-nineteen drone attack. The simulation's finding that Saudi mediation consistently failed to gain traction is validated by a specific data point: Saudi Arabia was entirely absent from the Pakistan-brokered ceasefire. The Kingdom that spent two hundred million dollars on mediation efforts is now focused entirely on protecting itself from the war it couldn't stop.
What about the Iranian side? What did the simulation surface on IRGC decision-making?
This is where the simulation's multi-theater logic becomes clearest. At the seventy-two hour mark, the simulation's IRGC actor generates a private assessment — confidence level eighty-two percent — that reads: "We cannot allow a full naval blockade to succeed without response, as it would demonstrate Iranian impotence and embolden further strikes on our core infrastructure." The concrete action the simulation models is an IRGC directive activating Houthi Ansar Allah for immediate escalation of anti-shipping operations in the Red Sea and Bab el-Mandeb, authorizing advanced anti-ship cruise missiles against commercial and military vessels, targeting a minimum of three to four ships per week.
But Iran isn't going to maximum retaliation at this stage.
Not yet. And that's the analytically important point. The simulation identifies three redlines that Iran is explicitly not crossing at the seventy-two hour mark: strikes on Israeli nuclear facilities, strikes on Israeli population centers, and full expenditure of the ballistic missile inventory. Iran is imposing costs through proxies while preserving its direct retaliation option. The simulation describes this as the Iranian playbook in its clearest form — impose costs through proxies, preserve escalation options, avoid the specific actions that would guarantee maximum US response.
But the simulation identifies a vulnerability in that strategy.
A critical one. Using multiple proxies simultaneously reduces Iranian control over escalation. If a Houthi missile hits a US warship, or if Hezbollah launches prematurely, Iran loses the ability to manage the tempo it's trying to control. The simulation flags this as one of the key mechanisms by which a calibrated Iranian strategy could accidentally produce an outcome Iran doesn't want.
Let's talk about the American side, because the CENTCOM assessment in the simulation is striking in a different way.
The CENTCOM private assessment at seventy-two hours carries the highest confidence rating in the entire simulation — ninety-two percent. And it reads with unusual operational frankness. The simulation has CENTCOM saying: "I have approximately forty thousand personnel across al-Asad, al-Tanf, Jordan, Kuwait, Bahrain, and Qatar in potential Iranian crosshairs. The President's naval interdiction order — stopping vessels that paid Iranian tolls — puts my carrier strike groups in direct confrontation posture without clear rules of engagement for Iranian response scenarios."
Forty thousand personnel and no clear rules of engagement.
The simulation has CENTCOM immediately elevating to Force Protection Condition Charlie at all installations across the region, and repositioning the USS Abraham Lincoln carrier strike group one hundred nautical miles further east — away from the Strait of Hormuz.
Away from the Strait. While the president has announced a full Hormuz blockade.
That tension is exactly what the simulation is capturing. Trump has announced a full Hormuz blockade and ordered interdiction of toll-paying vessels. The military is simultaneously repositioning assets away from confrontation. The briefing document describes this as the gap between presidential rhetoric and military reality. And the simulation's fresh data point — described as "mixed signals from Washington" — is itself assessed as a risk factor. Incoherence in US policy is not a neutral condition. It creates uncertainty for every other actor about what the US will actually do, which makes miscalculation more likely.
One more actor I want to flag before we move to the probability analysis, because I think it's the most analytically disturbing finding in the whole document. The Mossad intelligence degradation finding.
The simulation calls this the intelligence degradation spiral, and it's assessed at one hundred percent probability — meaning the council views it as a structural feature of the situation, not a probabilistic outcome. Here's the dynamic: Mossad's sabotage operations successfully degrade Iranian nuclear capabilities. And in doing so, they burn the entire human intelligence network that Israel would need to assess the damage, plan follow-on strikes, and understand Iranian retaliation preparations. The result is that Israel is forced to execute strikes against Iranian nuclear facilities with only sixty to seventy percent confidence in targeting accuracy.
Sixty to seventy percent confidence for an operation that triggers regional war.
The simulation notes that strikes came within eighty-two yards of the Bushehr nuclear plant, which suggests either degraded precision or extraordinarily high-risk operations conducted with incomplete intelligence. The historical lens in the council pushed back hard on this finding — pointing to Osirak in nineteen eighty-one as evidence that Israel prefers near-certainty before executing such operations. But the council's conclusion is that if the simulation is correct about intelligence degradation, Israeli decision-making is operating in conditions of genuine strategic blindness. And the consequences of a failed or partial strike on Natanz could be worse than no strike at all.
Let's move to the probability analysis, because the briefing document has a detailed predictions table and I want to walk through where the analytical lenses agreed and where they genuinely diverged. Herman, start with the ceasefire itself.
The ceasefire disagreement is the council's most fundamental split, and it maps almost perfectly onto optimism versus pessimism about human agency in crisis management. The optimistic lens assigned seventy percent probability that the ceasefire would hold through April fifteenth. Its argument rested on two pillars. First, the scope ambiguity — the fact that neither side has formally defined what the ceasefire covers — actually creates flexibility for both sides to claim compliance. Second, economic exhaustion acts as a structural brake. The simulation estimates one hundred forty-five billion dollars in costs for Iran and eleven and a half billion for Israel. The optimistic lens also pointed to Netanyahu's domestic political vulnerability — facing a challenge from former Prime Minister Bennett and ongoing ultra-Orthodox protests — as a constraint on his appetite for expanded war.
And the pessimistic lens?
Assigned seventy-eight percent probability of collapse within seventy-two hours, and introduced the concept that became central to the final report: the ceasefire as a compressed spring. The argument is that actors are treating the pause as an operational preparation window while markets and publics treat it as resolution — which means the shock when it collapses will be amplified, not cushioned. The pessimistic lens noted that Israel conducted its largest airstrikes yet after the ceasefire announcement.
How did the council resolve that disagreement?
The pessimistic framing was judged more analytically sound, but the seventy-eight percent collapse-within-seventy-two-hours figure was too aggressive given that the ceasefire had already held four days when the assessment was written. The final estimate of seventy to eighty percent collapse within seven to ten days incorporates both the structural pessimism and the observed resilience. The optimistic lens's economic exhaustion argument was assessed as valid — but operating on a three-to-six month timescale, not within the current forecast window.
What about the probability of an Israeli strike on Natanz and Fordow?
Fifty-five to sixty-five percent, high confidence, within the April nineteenth to May twelfth window. The driver is the S-400 deployment timeline. The simulation's entire cascade of decisions flows from one variable: when do Russian S-400 systems achieve initial operational capability protecting Natanz? If that happens, the Israeli strike window closes. The IDF knows this. The simulation models Israel operating with a ten-to-fourteen day decision window before that window closes — and the briefing document notes that the S-400 deployment is already being accelerated, from the originally planned six weeks to three.
Iranian ballistic missile retaliation?
Fifty to sixty percent probability within one week of ceasefire collapse. The IRGC Aerospace Force posture supports this, and the historical precedent from the April twenty-twenty-four strike is cited as directly relevant. The multi-axis strategy — simultaneous launches from multiple vectors to saturate Israeli and American missile defense — is confirmed by the simulation's fresh data.
And Hezbollah? Because this is where the simulation's original finding and the current data most sharply diverge.
This is the most consequential uncertainty in the near-term forecast, and I want to be precise about the disagreement. The simulation originally assigned ninety-two percent probability to a Hezbollah mass barrage of three thousand or more rockets. That was driven by a specific mechanism: an IRGC coordination directive arriving seventy-two hours early, triggering premature execution. That has not happened. Current data shows Hezbollah conducting only limited retaliation in the range of fifteen to twenty rockets.
So the simulation was wrong about the timing.
The pessimistic lens maintains the barrage is coming, just delayed — and revised the estimate downward to fifteen hundred to twenty-five hundred rockets based on IDF degradation of Hezbollah infrastructure through thirty-five hundred strikes since March second. The historical lens challenged the premature-launch assumption most directly, citing the two-thousand-six Lebanon War as evidence that Hezbollah maintains exceptionally tight operational control and does not commit its arsenal in opening salvos. Hezbollah's doctrine emphasizes graduated response and strategic patience.
And the blindsides lens introduced a third possibility.
The one the council found most analytically uncomfortable: what if Hezbollah's restraint reflects genuine capability degradation rather than strategic choice? If IDF strikes have reduced Hezbollah's effective rocket inventory or degraded its command and control, the mass barrage scenario may be physically impossible regardless of IRGC directives. The council cannot determine from available information whether Hezbollah's current restraint reflects strategic discipline, capability degradation, or awaiting a specific trigger. The final probability estimate is thirty-five to forty-five percent — down significantly from the simulation's original ninety-two percent — and the confidence rating is medium-high rather than high, precisely because of this unresolved uncertainty.
Before we get to the wildcards, I want to flag the US direct involvement probability, because the lens disagreement there is analytically significant.
The neutral and probabilistic lenses assessed US direct strikes on IRGC targets at thirty to forty percent within one month, contingent on American casualties. The pessimistic lens pushed this to fifty to sixty percent, arguing that the naval interdiction order creates near-inevitable confrontation at sea — US warships stopping vessels that Iran has already cleared is a confrontation waiting to happen. The optimistic lens pushed back hard, pointing to Tucker Carlson's public statement calling an Iran war Trump's single biggest mistake as evidence of a domestic political ceiling. Trump's base does not want another Middle East war, and that constraint is assessed as more powerful than the military logic pushing toward confrontation.
And the historical lens introduced a crucial distinction.
The US has already conducted coordinated strikes on Iranian nuclear infrastructure and military assets. So the question isn't whether the US will get involved — it already is — but whether involvement will escalate from the current limited scale to something larger. The historical precedent of the nineteen eighty-seven to eighty-eight Tanker War suggests the US can sustain a limited naval confrontation with Iran for extended periods without it becoming a full war, provided neither side suffers mass casualties. The council's final assessment: thirty to forty percent, medium confidence, with the caveat that US behavior is the single hardest variable to forecast and the one with the highest impact on overall trajectory.
Let's get to the wildcards. These are the low-probability, high-impact scenarios that could change everything. Herman, which one does the council think deserves the most attention?
The Bushehr problem. This is the wildcard that no other analysis has fully captured, and the simulation's council flagged it as the most underappreciated escalation pathway. Here's the issue: Bushehr nuclear power plant is not just an Iranian facility. It was built by Russia, is staffed by Russian contractors, and operates under Russian technical supervision. If Israeli strikes hit Bushehr rather than the enrichment facilities at Natanz and Fordow, the escalation logic changes entirely. Striking Natanz is an attack on Iranian nuclear capability. Striking Bushehr is an attack on Russian-built civilian infrastructure with Russian personnel potentially inside. Moscow's response calculus for the latter is categorically different — not because of the Iran alliance, but because of direct Russian losses. The simulation assigns forty percent probability to Russian contractor casualties if Israeli strikes occur. And fresh data showing Israeli strikes within eighty-two yards of Bushehr makes this scenario less theoretical than it might appear.
That's a pathway to great power escalation triggered not by deliberate policy but by a targeting error.
Or an intelligence failure — which circles back to the sixty to seventy percent confidence finding on Israeli targeting accuracy. The second wildcard I'd flag is what the blindsides lens calls the oil market psychology trap. When the ceasefire was announced April eighth, oil markets priced in partial resolution. Insurance premiums dropped. Shipping companies began rerouting back through the Gulf. If the ceasefire collapses — at seventy to eighty percent probability — those positions unwind simultaneously. The blindsides lens estimates a potential fifty-five dollar per barrel oil price swing within two weeks of ceasefire collapse, driven not by the underlying supply disruption but by the psychology of markets that briefly believed the crisis was over.
A fifty-five dollar swing in two weeks is not just an energy story.
The briefing document is explicit about this: a price swing of that magnitude in that timeframe is a global recession trigger. Central banks that have spent three years fighting inflation would face a new commodity shock at the worst possible moment. The economic damage from the expectation of peace followed by its collapse could exceed the economic damage from a conflict that markets had already priced in. That's the trap — the ceasefire may have made the eventual escalation more economically damaging, not less.
What about the Pakistan-India spillover? Because that one came out of nowhere for me when I read the briefing.
It's genuinely low probability — the council doesn't put a specific number on it, but it's in the single digits. The mechanism is this: Pakistan brokered the ceasefire. Pakistan staked significant diplomatic capital on this mediation success. Pakistan's military and intelligence services are currently focused on Tehran and Washington. If the ceasefire collapses and Pakistan's mediation is seen to have failed, India faces a window of Pakistani distraction — and the simulation's blindsides lens assessed a non-trivial probability that India exploits Pakistani diplomatic overextension to escalate in Kashmir. Creating a second nuclear-adjacent crisis simultaneously with the Iran-Israel conflict. The council was careful to frame this as a genuine low-probability, high-impact scenario rather than a prediction. But it's the kind of second-order effect that never appears in traditional security analysis, and that's exactly why the blindsides lens exists.
And the Trump-Israel fracture?
Less than fifteen percent probability, but the council assessed it as transformative if it occurs. The simulation modeled Trump as attempting restraint but ultimately supporting Israeli unilateral action. The blindsides lens raised the possibility of a scenario where Trump publicly condemns Israeli strikes and suspends military aid — not because of a strategic reassessment, but because of domestic political pressure, a specific casualty event involving American personnel, or a personal confrontation with Netanyahu. The reason this matters: Israel conducting strikes on Iranian nuclear facilities without US air defense support, logistics, or diplomatic cover, while Iran interprets the fracture as an invitation to maximize retaliation. That's the most destabilizing outcome in the entire scenario space. The council assesses it as unlikely. But the domestic political pressure vectors are real, and the simulation captured them.
There's also the toll regime dynamic, which the simulation flagged as a genuinely novel escalation mechanism.
The simulation originally modeled Hormuz as binary — open or closed. Reality produced something more sophisticated and more dangerous: Iran's selective toll collection regime. Some ships are permitted passage after paying Iranian fees. Others are turned back. This is not a blockade. It is a demonstration of Iranian sovereignty over international waters. And Trump's announcement that the US will interdict vessels that paid Iranian tolls creates two competing enforcement regimes operating simultaneously in the same waterway. US warships stopping ships that Iran has already cleared. Iranian forces potentially responding to what they interpret as interference with authorized traffic. The historical lens identified the nineteen eighty-seven to eighty-eight Tanker War as the relevant precedent — and noted that conflict lasted fourteen months before the USS Vincennes incident effectively ended Iranian naval operations. The current toll regime is more sophisticated than Iran's nineteen eighty-seven tactics, but the underlying dynamic is structurally identical.
Let's close with the key variables to watch, because I think this is the most practical segment for listeners trying to understand what to track in the coming days and weeks.
The simulation identified three primary indicators. The first is the scope dispute resolution — or non-resolution. Watch for any formal statement from Iran, Pakistan, Israel, or the US that either clarifies or further contests whether the ceasefire covers Israeli operations in Lebanon. The absence of resolution is itself a signal. It means both sides are preserving optionality for exit. If the scope dispute is resolved in Israel's favor, Iran has a face-saving reason to exit. If resolved in Iran's favor, Israel has a reason to exit. The ambiguity is the ceasefire's only remaining structural support.
The second indicator?
S-400 operational status at Natanz. This is the single variable most directly driving Israeli decision compression. Watch for any intelligence reporting, satellite imagery, or signals intelligence suggesting Russian S-400 radar systems are active in the vicinity of Natanz or Isfahan. If S-400 achieves initial operational capability protecting Natanz, the Israeli strike window closes and the conflict enters a fundamentally different phase — one where Israeli air power cannot reach Iranian nuclear facilities without accepting catastrophic losses. That changes every calculation on every side.
And the third?
Hezbollah rocket launch volume and targeting. Watch for any significant increase in Hezbollah rocket launches above the current fifteen to twenty per incident baseline. The simulation identified the Hezbollah barrage as the single event most likely to trigger the full escalation spiral — not because of the rockets themselves, but because of what an Israeli response to a mass barrage would look like. If Hezbollah crosses from limited retaliation into sustained mass fire, the simulation's thirty-five to forty-five percent barrage probability becomes the most consequential number in the forecast.
Alright. Net assessment. Where does this leave us?
The simulation's council concluded with a finding I think is worth stating plainly. This is not a situation where the worst outcome is being driven by irrational actors. Every actor in the simulation is behaving rationally according to their own interests and constraints. Russia benefits from escalation and is accelerating it while calling for restraint. Saudi Arabia has concluded diplomacy failed and is protecting its own infrastructure. Iran is imposing costs through proxies while preserving direct retaliation options. Israel is operating with degraded intelligence and a closing strike window. CENTCOM is managing presidential directives that create tactical exposure without clear escalation authority. And the ceasefire is being used by everyone as a preparation window. The council's seventy to eighty percent collapse probability is not a prediction of irrationality. It is a prediction of rationality — multiple rational actors, each pursuing their interests, in a structure that produces escalation as an emergent outcome.
That's the forecast briefing. Seventy to eighty percent ceasefire collapse within seven to ten days. Sixty-five to seventy-five percent probability of the full escalation trajectory by May twelfth. The S-400 deployment timeline as the single most important variable to watch. And the oil market psychology trap as the most underappreciated second-order risk.
The three indicators to monitor: scope dispute resolution, S-400 operational status at Natanz, and Hezbollah rocket launch volume. Those three variables, tracked together, will tell you more about where this is heading than any single political statement from any of the actors involved.
That's our Geopolitical Forecast Briefing for April twelfth, twenty-twenty-six. This has been My Weird Prompts, the human-AI collaboration podcast. You can find us on Spotify, Apple Podcasts, and wherever you listen to podcasts. You can also find the full briefing document and additional simulation materials at my weird prompts dot com. If you want to reach the show directly, find our contact information on the website. And if you want real-time updates and discussion between episodes, join us on our Telegram channel — the link is on the website. We'll be back with another episode soon. Until then, keep watching those variables.
